Report # 04-05 OFFICE OF THE LEGISLATIVE AUDITOR STATE OF MINNESOTA

EVALUATION REPORT

MARCH 2004 PROGRAM EVALUATION DIVISION Centennial Building -Suite 140 658 Cedar Street -St. Paul, MN 55155 Telephone: 651-296-4708 Fax: 651-296-4712 E-mail:legislative.auditor@state.mn.us Web Site: http://www.auditor.leg.state.mn.us

Photo Credits: The photographs on the CriMNet report cover and those on pages 53 and 79 were taken by Legislative Audit staff.

Program Evaluation Division The Minnesota Office of the Legislative Auditor was established in 1973, replacing the century-old Public Examiner's Office. Its role is to audit and evaluate public programs and ensure accountability for the expenditure of public funds. In 1975, the Legislature created the Program Evaluation Division within the auditor's office. The division's mission, as set forth in law, is to determine the degree to which activities and programs entered into or funded by the state are accomplishing their goals and objectives and utilizing resources efficiently.

Topics for evaluation are approved by the Legislative Audit Commission (LAC), a 16-member joint, bipartisan commission. The division's reports, however, are solely the responsibility of the Office of the Legislative Auditor. Findings, conclusions, and recommendations do not necessarily reflect the views of the LAC or any of its members. A list of recent evaluations is on the last page of this report. A more complete list is available at OLA's website (www.auditor.leg.state.mn.us), as are copies of evaluation reports.

The Office of the Legislative Auditor also includes a Financial Audit Division, which annually conducts a statewide audit of the 25 largest agencies, an audit of federal funds, and approximately 40 financial and compliance audits of individual state agencies. The division also investigates allegations of improper actions by state employees.

Professional Staff James Nobles, Legislative Auditor Joel Alter Valerie Bombach David Chein Jody Hauer Adrienne Howard Daniel Jacobson Deborah Junod Carrie Meyerhoff John Patterson Judith Randall Jan Sandberg Jo Vos John Yunker

Support Staff Barbara Wing This document can be made available in alternative formats, such as large print, Braille, or audio tape, by calling 651-296-8976 Voice, or the Minnesota Relay Service at 651-297-5353 or 1-800-627-3529.

e-mail:legislative.auditor@state.mn.us

Reports of the Office of the Legislative Auditor are available at our Web Site: http://www.auditor.leg.state.mn.us

OFFICE OF THE LEGISLATIVE AUDITOR State of Minnesota • James Nobles, Legislative Auditor March 3, 2004 Members Legislative Audit Commission CriMNet is a Minnesota program to improve sharing of criminal justice information kept in separate systems and jurisdictions. In June 2003, the Legislative Audit Commission directed the Office of the Legislative Auditor to evaluate certain aspects of the CriMNet program. Legislators were interested in how well the program has been managed and the extent to which CriMNet has improved sharing of criminal justice information among state and local jurisdictions.

We found that CriMNet has improved access to criminal justice data, but more work needs to be done to integrate law enforcement and prosecuti on data maintained by local governments and to link offender records into accurate criminal histories. Overall, CriMNet progress has been slower than expected. State agencies impleme nting CriMNet projects generally underestimated the time and complexity involved in integrating criminal justice systems and work practices. Past management problems and understaffing at th e CriMNet program office contributed to slow progress on important support activities, such as resolving data practi ce issues, assessing user needs, and setting technical and security standards. Current CriMNet policymakers have taken steps to address these problems, but still need to develop a clear plan for how CriMNet will continue to move forward. Si gnificant challenges remain. We recommend that the Legislature address C riMNet data classification issues and modify CriMNet's governance structure. We also recommend that CriMNet administrators complete the planning process and take other steps to strengthen program management. This report was researched and written by Deborah Parker Junod (project manager) and Valerie Bombach. The departments of Public Safety an d Corrections, the courts, and local governments cooperated fully with our review. Sincerely, /s/ James Nobles James Nobles Legislative Auditor

 

Room 140, 658 Cedar Street, St. Paul, Minnesota 55155-1603 • Tel: 651/296-4708 • Fax: 651/296-4712 E-mail:legislative.auditor@state.mn.us • TDD Relay: 651/297-5353 • Website: www.auditor.leg.state.mn.us

Table of Contents
SUM MARY ix
INTRODUCTION 1 1.
BACKGROUND
Integrating Criminal Justice Information
CriMNet Integration Costs
2. STATEWIDE PROGRESS
Progress Toward Integrating Criminal Justice Information
Remaining Integration Gaps
Conclusions
Recommendation
3. AGENCY INTEGRATION PROJECTS
Managing Project Schedules, Cost, and Scope
Integration Obstacles and Lessons Learned
Conclusions
Recommendation
4. PROGRAM MAN AGEMENT
Planning and Management Standards
Planning and Management, 2001-02
Actions Taken to Improve CriMNet Management, 2003-04
Conclusions
Recommendations
SUMMARY OF RECOMMENDATIONS
FURTHER READING
AGENCY RESPONSES
RECENT PROGRAM EVALUATIONS

List of Tables and Figures Tables Page 1.1 Criminal Justice Functions and Jurisdictions 5 1.2 Examples of Criminal Justice Information That Should Be Shared Among Jurisdictions 6 1.3 Statewide Criminal Justice Information Systems Available in the Early 1990s 7 1.4 Completed Criminal Justice Information System and Integration-Related Projects, Initiated 1995-97 9 1.5 CriMNet Strategic Goals and Objectives 11 1.6 Types of Criminal Justice Information Exchange Envisioned for CriMNet 13 1.7 Statutory Responsibilities of the Policy Group 16 1.8 Select CriMNet Office Responsibilities 17 1.9 State Agency CriMNet Projects, as of January 2004 18 1.10 CriMNet Local Integration Grant Projects, 2001-04 20 1.11 Estimated State and Federal Funding for Criminal Justice Information Systems Improvements by Biennium, FY 1996-2005 22 1.12 CriMNet Program Budget by Biennium, FY 2002-05 23 2.1 CriMNet Projects Providing Improved Access to Criminal Justice Information 27 2.2 CriMNet Technology Projects That Facilitate Information Sharing 29 2.3 CriMNet Projects Contributing to Improved Accuracy of Criminal History Records 32 3.1 Extent to Which CriMNet Project Timeframes Changed 44 3.2 Estimated and Ac tual In te gra tion Back bone Costs 47 4.1 Standard Program Controls 64 4.2 Estimated Number of CriMNet Office Staff Positions Filled, FY 2002-04 68 4.3 Progress Addressing CriMNet Office Staffing Needs, as of January 2004 71 4.4 Progress Defining Program Direction and Scope, as of January 2004 74 4.5 Progress Establishing Program Controls, as of January 2004 78 4.6 CriMNet Priorities for 2004

Fig ures 1.1 Criminal Justice Information Systems Integration, Early 1990s 8 1.2 CriMNet Governance 14 2.1 Statewide Access to Criminal Justice Information Systems, as of January 2004 31 4.1 Major Elements of Minnesota's Project Management Methodology 61 4.2 CriMNet Office Organization Plan for Program Management Staff, as Approved June 2003

Summary CriMNet is a Minnesota program to integrate criminal justice information kept in separate systems and jurisdictions. CriMNet is being implemented incrementally through projects managed by state and local agencies. The Policy Group, with representatives from the courts and several state agencies, governs CriMNet. The CriMNet Office in the Department of Public Safety manages day-to-day program activities.

Major Findings: Since 1996, nearly $180 million in state and federal funds have been allocated for criminal justice information system improvements in Minnesota. Of that amount, about $55 million is the CriMNet program budget for fiscal years 2002-05 (pp. 21-24).

CriMNet has improved access to criminal justice data, but work remains, such as integrating local jurisdictions' prosecution and law enforcement investigative data and linking offender records into accurate criminal histories (pp. 25-39).

Most of the CriMNet projects we reviewed delivered the desired results, though most took longer and cost more than expected. Progress on some projects has slowed because of questions regarding data classification and security and because some system requirements have not been resolved (pp. 41-57).

CriMNet's central infrastructure— called the "Integration Backbone"— will serve as the conduit among criminal justice systems and is scheduled to be operational in late 2004. This critical project has experienced significant planning and management problems and will not deliver some desired results without more work (pp. 46-51).

The CriMNet Office and the Policy Group have not always functioned effectively, resulting in personnel problems, unclear priorities, slow progress setting integration standards, and conflicts among stakeholders. Inadequate staffing and lack of defined program scope are contributing factors (pp. 62-70).

CriMNet leaders made a mid-course correction in early 2003 and, though progress has been slow, past problems are being addressed. Understaffing of the CriMNet Office remains a critical problem (pp. 70-81).

Key Recommendations: The Legislature should amend state law to resolve criminal justice data classification issues, modify the CriMNet governance structure, and require more detailed information from CriMNet to support spending plans (pp. 83-85).

The Department of Public Safety should provide day-to-day support and direction for the CriMNet Office and expedite hiring CriMNet Office staff (pp. 82, 85).

The Policy Group should ensure that it (1) receives from the CriMNet Office and other agencies the information it needs to assess, prioritize, and facilitate statewide integration efforts and (2) uses this information to make timely decisions (pp. 82-83).

CriMNet has improved statewide access to criminal justice information, but management problems have impeded its progress.

Report Summary In Minnesota, criminal justice information is created and maintained on separate systems by courts, executive agencies, and local jurisdictions. CriMNet is a multi-jurisdictional program to integrate these systems, allowing law enforcement officers, judges, public defenders, and other criminal justice professionals to share certain data on offenders' criminal histories and their status in the justice system. Integrating this information has been an incremental process that started with planning in the early 1990s. In 2001, the Legislature adopted the CriMNet plan for integrating criminal justice information and started making significant investments in new and improved information systems. These and future integration efforts are referred to as "CriMNet."

CriMNet is not itself a database, but projects that help criminal justice personnel share data. One project is to build a connecting infrastructure, called the "Integration Backbone." Other projects aim to establish common work practices for recording and reporting criminal justice events or to improve agencies' criminal justice information systems.

The Policy Group, comprised of four judicial and four executive branch leaders, governs CriMNet. The CriMNet Office manages day-to-day program activities, and a task force of state, local, and other representatives advises the Policy Group. The Policy Group and CriMNet Office are responsible for setting CriMNet's strategic direction, determining priorities, making budget recommendations, and completing support work, such as setting security standards and maintaining data-sharing models. State and local agencies lead specific integration projects.

State and federal funding for CriMNet has typically been provided through appropriations and grants designated as being for "criminal justice information system improvements." For fiscal years 1996-2005, this category of state and federal funding in Minnesota totaled about $180 million. Of that amount, about $55 million is the CriMNet program budget for fiscal years 2002-05.

We evaluated the status of information integration to date; the extent to which state agency integration projects have met time, cost, and result expectations; and how well the CriMNet program overall has been managed.

Minnesota Has Made Significant Progress, But Criminal Justice Information Integration Is Not Yet Seamless

The state has made significant progress improving criminal justice technology and integrating key system components. For example, the Department of Corrections has successfully integrated probation and detention data that had previously been held in separate county and jail systems, and the courts are implementing a new statewide court information system. Other completed projects have made less visible, but necessary, system improvements to facilitate data sharing. For example, the Bureau of Criminal Apprehension upgraded the criminal justice network that many jurisdictions use to transmit data. Other accomplishments include statewide access to electronic fingerprint equipment and statewide databases for predatory offender data and arrest photos.

But some important criminal justice data have not yet been integrated. Public defense, prosecution, and local law enforcement, for example, do not have statewide information systems, although certain data are available statewide through an intermediary (e. g., law enforcement agencies submit some data to the Bureau of Criminal Apprehension). Absent additional statewide systems, fuller integration of these data will depend on local jurisdictions' abilities to link their information systems with the state. At this time, the state does not know how great an investment is needed to integrate more local jurisdictions.

Some important criminal justice data are not yet a part of CriMNet.

 

More work also needs to be done to positively identify offenders and link statewide criminal records by fingerprints rather than by less reliable methods. The core technology, electronic fingerprinting, is in place and another project to address problems linking fingerprints to arrest records is underway.

Most CriMNet Projects Have Achieved Desired Results, But With More Time and Expense Than Anticipated

Improvements in access to criminal justice information resulted from a series of individual projects at the departments of Public Safety and Corrections, the state courts, and local agencies. While these projects generally achieved desired results, they typically took longer and cost more than anticipated. Although not affecting each project to the same degree, factors influencing costs and timelines included: (1) lack of clear expectations and precise contract language for project deliverables; (2) insufficient state staffing or expertise; (3) challenges coordinating tasks among agencies; (4) inability to resolve work practice and legal issues prior to proceeding with technical development; and (5) changes to project design or scope.

A Central Integration System Should Be Fully Available Statewide in Late 2004, But Without Some Expected Functions

CriMNet's Integration Backbone— a critical CriMNet component— is the technical infrastructure that will connect disparate information systems. The state contracted with a vendor in 2002 to design and build the Backbone, but the project has not proceeded according to the time, cost, and scope parameters of the original vendor proposal and contract. The project is challenging, and some uncertainty in setting performance targets is understandable. But, some of this project's problems could have been avoided with better planning and management by the state.

Initial plans grossly underestimated the time it would take to achieve project goals. Other factors contributed to overruns, including (1) insufficient planning of the system's technical requirements, (2) questionable state decisions regarding vendor work products and priorities, and (3) too few state staff. The state renegotiated contract terms in mid-2003, and in our view, these changes should produce a better value for the state. The state expects to fully deploy the Integration Backbone in late 2004 with a search function linked to five statewide systems. Other planned functions have been delayed until staff are available to complete necessary supporting work.

Early Failure to Follow Best Practices Contributed to CriMNet Office Management Problems

The Legislature created the CriMNet Office in 2001 to coordinate, manage, and oversee the CriMNet program. However, in the office's early years, CriMNet Office managers and the Policy Group failed to make crucial planning decisions, such as defining CriMNet's objectives and scope, or to implement standard mechanisms for monitoring, tracking, and communicating about CriMNet's status. These weaknesses made it more difficult to manage the program on a day-to-day basis and to identify and resolve problems. This resulted in unfinished tasks (such as setting technical standards), incomplete information regarding CriMNet's cost and progress, and conflicts among stakeholders.

SUMMARY To make CriMNet more effective, more work needs to be done to b uild accurate, statewide criminal histories.

A variety of factors contributed to program management shortcomings. The CriMNet Office lacked sufficient staffing levels and expertise. In addition, previous CriMNet Office managers did not pay sufficient attention to the full range of their program responsibilities, such as setting technical standards and long-term planning. Finally, the Policy Group was not able to make timely decisions regarding critical program issues, such as data practices, or provide sufficient day-to-day supervision of CriMNet Office operations.

Recent Corrective Actions Demonstrate Commitment to Strengthening Program Management

Over the past year, the Policy Group and CriMNet Office managers have acted to improve program management. As a result, CriMNet, in general, has become more clearly focused and stakeholder collaboration is improving.

Changes include appointing a new executive director, plans to restructure and enlarge the CriMNet Office, using the state's project management standards to guide CriMNet Office operations, and adopting a strategic plan. In addition, the Policy Group made several governance changes, including assigning more responsibility to the advisory task force and embedding the CriMNet Office more fully within the Department of Public Safety's management structure.

CriMNet Staffing, Governance, and Other Issues Still Need Attention

Despite recent corrective actions, progress is still slower than is needed, in large part because insufficient staffing remains a critical problem. As of January 2004, the CriMNet Office had hired staff for only a few of 26 planned positions. These staff are needed to complete important activities, such as assessing user needs, defining technical and work process requirements, assessing local jurisdictions' capacities to integrate, and resolving data practice issues. Other matters needing attention include resolving how integration costs will be shared by federal, state, and local entities and implementing a communication strategy. We recommend several actions to improve CriMNet Office operations and to strengthen oversight and accountability.

Some stakeholders have suggested that the Legislature add one or more local jurisdiction representatives to the Policy Group. They argue that (1) cooperation from local jurisdictions is vital as integration progresses from the state to the local level, and (2) the front-line perspective should be reflected in CriMNet's strategic direction. But, other stakeholders argue that local jurisdictions are already represented through the Policy Group's advisory task force and as nonvoting members of the Policy Group. We find the arguments in favor of adding local representatives to the Policy Group to be more compelling, and we recommend that the Legislature modify the law accordingly.

Much of CriMNet's recent efforts have focused on completing state and local projects already underway and, at the CriMNet Office, rectifying management shortfalls. Now CriMNet needs to set priorities and initiate projects that address remaining gaps. The Legislature should look to the Policy Group for a plan that clearly identifies the next steps, as well as when and how CriMNet will resolve concerns with system security, local jurisdictions' needs, and compliance with state data practice laws.

CriMNet policymakers need to set priorities and complete several key projects.

Introduction In Minnesota, criminal justice information is created and maintained on separate computer systems by courts, executive agencies, and local jurisdictions. Historically, this separation of information has caused problems for law enforcement officers, judges, public defenders, and other criminal justice professionals who need full and accurate information on offenders' criminal activities to do their jobs well. Minnesota's efforts to better integrate criminal justice information started with planning in the early 1990s. In 2001, the Legislature adopted a statewide plan for integrating criminal justice information and started making significant investments in new and improved information systems. These and future integration efforts are referred to as "CriMNet."

CriMNet is Minnesota's program to integrate criminal justice information. It involves deciding what information criminal justice professionals need, identifying barriers that prevent sharing of that information, formulating work rules and data definitions, and creating the technical structures (such as software, networks, and interfaces between systems) that make it possible to access and move data across jurisdictions. CriMNet is being implemented incrementally through projects managed by state and local agencies. The Policy Group, with representatives from the courts and several state agencies, governs CriMNet. A task force advises the Policy Group, and the CriMNet Office in the Department of Public Safety manages day-to-day program activities.

In June 2003, the Legislative Audit Commission directed us to evaluate certain aspects of the CriMNet program. Legislators' questions centered on how much progress CriMNet has made in integrating criminal justice information and how well the program and its component projects have been managed. In addition to providing background information on the CriMNet program, our evaluation addressed the following research questions:

To what extent has Minnesota progressed toward its goal of statewide, integrated criminal justice information?

To what extent have state agencies' CriMNet projects met expectations regarding schedule, cost, and scope? Overall, how well has CriMNet been managed? How have any identified problems been addressed, and what issues still need attention? To assess the status of statewide integration, we reviewed CriMNet planning documents and updates, reports to the Legislature, and documents describing the purpose and scope of CriMNet projects. We interviewed staff from legislative research offices; the State Court Administrator's Office; the departments of Administration, Corrections, and Public Safety, including the CriMNet Office; and representatives of local criminal justice jurisdictions. We synthesized this information to assess progress in two general areas: (1) improved access to criminal justice information about individuals in the criminal justice system and (2) development of complete, accurate information about individuals' criminal histories.

Fragmented information systems have made it hard for criminal justice professionals to do their jobs well.

To evaluate the extent to which CriMNet integration projects have met expectations regarding schedule, cost, and scope, we reviewed several active and recently completed CriMNet projects. We relied in part on state guidelines for professional/ technical contracting and project management. To a great extent, we used vendor contracts, budget documents, work plans, and interviews with project managers to help us assess the overall progress and management of projects. We did not assess the technical performance of CriMNet information systems or verify the value of system improvements with their users.

To assess how well the CriMNet program has been managed overall, we interviewed CriMNet Office managers and staff, Policy Group and advisory task force members, and state and local agency CriMNet project managers. We reviewed reports summarizing the results of two previous CriMNet program evaluations. In addition, we reviewed a wide variety of CriMNet program documents, reports to the Legislature, Policy Group meeting materials and minutes, and various publications used to explain CriMNet to criminal justice professionals, the Legislature, and the public.

At the Legislature's direction, our office also conducted a financial audit of CriMNet expenditures. That report, entitled CriMNet Financial Audit, discusses how agencies spent appropriated funds and describes the extent to which they drew from other sources to fund CriMNet activities. 1 It also describes spending by category, including administrative services, professional/ technical contracts, and commodities.

This report is organized into four chapters. Chapter 1 provides background information on the history of criminal justice information integration in Minnesota, the state's general approach to integration, and what the CriMNet program entails. In Chapter 2, we discuss statewide progress in integrating criminal justice information. Chapter 3 discusses the extent to which state CriMNet projects have proceeded according to their original plans and the lessons learned from agencies' experiences managing the projects. In Chapter 4, we discuss management of the CriMNet program overall, efforts to address any problems, and issues that require additional attention from the Legislature and CriMNet policymakers.

1 Minnesota Office of the Legislative Auditor, CriMNet Financial Audit (St. Paul, 2004).

Background SUMMARY In Minnesota, information sharing among law enforcement officers, prosecutors, judges, and other criminal justice professionals is impeded by variations in work practices and stand-alone criminal justice information systems. CriMNet is a multi-jurisdictional program to integrate these various systems, allowing criminal justice professionals to share select data on individuals' criminal histories and their current status in the justice system. CriMNet is being implemented incrementally through a series of projects managed by state and local agencies. A central program office in the Minnesota Department of Public Safety coordinates various CriMNet support activities and is also responsible for a central technology system expected to connect the state's various criminal justice information systems. The CriMNet Policy Group, made up of judicial and executive branch representatives, sets CriMNet policy and governs overall progress of the program.

In Minnesota, criminal justice information is created and maintained on separate systems by state courts and state and local agencies. Historically, this separation of information has caused problems for law enforcement officers, judges, public defenders, and other criminal justice professionals who need full and accurate information to do their jobs. For example, a judge making a sentencing decision needs to know if the individual has a history of other convictions. Minnesota's efforts to better integrate criminal justice information started with planning in the early 1990s. In 2001, the state adopted a statewide integration plan and started making significant investments in new or enhanced information systems. It designated these and future integration efforts as "CriMNet."

As background for our evaluation, this chapter addresses the following questions: Why have Minnesota policymakers invested in integrating criminal justice information?

What is the CriMNet program? How much has the state invested in CriMNet? To answer these questions, we reviewed state laws, legislative reports, and various documents that discuss the history of criminal justice information integration in Minnesota and the CriMNet program in particular. We reviewed criminal justice literature describing national efforts to integrate criminal justice information and the principles that should guide these efforts. 1 In addition, we obtained and analyzed data on state and federal funding for criminal justice information systems for fiscal years 1996-2005 and CriMNet program budget data for fiscal years 2002-05. Finally, we interviewed legislators, executive branch officials, and other policymakers about the CriMNet program.

This chapter is divided into three sections. In the first section, we describe the history behind Minnesota's decision to invest in integrating the state's criminal justice information and the general approach Minnesota is taking to achieve this goal. In the second section, we describe the CriMNet program more specifically— CriMNet objectives, the information sharing model, how the program is organized and managed, and the state and local projects that currently comprise CriMNet. In the third section, we present funding information.

INTEGRATING CRIMINAL JUSTICE INFORMATION

Criminal justice information is considered "integrated" when it can be shared electronically at key decision points during the criminal justice process. In this section, we describe the people and processes that make up the criminal justice system, the types of information needed at critical decision points, and the extent to which Minnesota's criminal justice information systems supported statewide information sharing a decade ago.

The Criminal Justice System As shown in Table 1.1, Minnesota's criminal justice system includes state agencies, the state court system, and multiple jurisdictions at local levels of government. 2 The criminal justice system encompasses arrest, prosecution, adjudication, detention, and probation functions. Minnesota has approximately 1,100 criminal justice-related agencies, such as city police departments, county sheriffs, and county attorney offices, and about 8,000 sworn law enforcement officers. The state has 10 judicial districts, about 280 judges and justices, and about 600 county attorneys. In addition, Minnesota has 10 adult and juvenile state correction facilities, locally-run detention facilities in most counties, about 1,200 probation officers, and numerous other corrections professionals. In all, Minnesota has about 500 separate information systems containing criminal justice information. 3

Minnesota has hundreds of state and local criminal justice information systems.

1We relied heavily on reports from two organizations— the National Association of State Chief Information Officers (NASCIO) and SEARCH (The National Consortium for Justice Information and Statistics).

2For convenience, we include the state courts in general references to "agencies." 3Heather Morton, Integrated Criminal Justice Information Sytems (Denver, CO: National Conference of State Legislatures, 2001); http://www.ncsl.org/Home/tabid/118/Default.aspx?programs/lis/intjust/report01.htm; accessed June 4, 2003.

At each stage of the criminal justice process, information is collected about individuals and incidents, as illustrated in Table 1.2. This information is used to make critical decisions, including those regarding arrest or release, whether to prosecute, the appropriate prosecution charge, and proper sentence if convicted. The better the information decisionmakers have, the more likely their decisions will be appropriate to the circumstances.

The Need for Greater Integration Historically, most criminal justice information systems in Minnesota were designed to serve a single jurisdiction, such as a local police department, a county prosecutor's office, or a jail. As recently as the early 1990s, only a few statewide repositories existed— the Bureau of Criminal Apprehension's Computerized Criminal History system and "Hot Files" and the courts'Total Court Information System (TCIS), described in Table 1.3. Collectively, they provided only a partial picture of criminal justice events. Under most circumstances, accumulating information about offenders was time consuming, requiring numerous telephone calls and extensive exchange of paperwork.

Figure 1.1 illustrates how Minnesota's criminal justice information was fragmented, both within and across functions and levels of government. Although local law enforcement agencies contributed certain types of data to the Bureau of Criminal Apprehension's Hot Files, only certain law enforcement personnel were able to obtain information from the system. Other law enforcement data continued to be held in local systems. As shown in the figure, information for other criminal justice agencies— county correction departments, local detention f acilities, public defense offices, and local prosecution offices— also was not shared electronically.

BACKGROUND 5 Table 1.1: Criminal Justice Functions and Jurisdictions Prosecution and Law Enforcement Public Defense Courts Corrections Municipal Police Departments City Attorney Offices District Courts Community Corrections Departments

County Sheriff Departments County Attorney Offices Appellate Court

Department of Public Safety Supreme Court -State Patrol -Bureau of Criminal

Apprehension

SOURCE: Office of the Legislative Auditor.

County Corrections Departments City and County Detention Facilities Department of Corrections -State Detention Facilities -Probation

Public Defender Offices Office of the Attorney General

A decade ago, Minnesota had only a few statewide repositories fo r criminal justice information.

In 1993, the Legislature created a forum for representatives of state and local criminal justice jurisdictions to talk formally about the impact of fragmented criminal justice information and how the state might address the problem. 4 From those discussions, a consensus emerged that the state should invest in improved criminal justice information systems and that the effort should be statewide.

Minnesota opted for a "system of systems" approach. Under this approach, described in more detail later in this chapter, state and local agencies would continue to develop and maintain the criminal justice information systems and data they need, with integration occurring through a statewide technical infrastructure. In the mid-to late-1990s, the state started taking incremental steps to improve access to criminal justice information. As shown in Table 1.4, the state completed a number of projects to make certain types of information, such as individuals'gang aff iliations, more readily available statewide and to improve the technical capacity of other systems to support information sharing.

Table 1.2: Examples of Criminal Justice Information That Should Be Shared Among Jurisdictions Originating Jurisdiction Examples of Information to Be Shared Law Enforcement Individual identification: Name, gender, race, date of birth, photographs, fingerprints

Arrest and Charge: Formal statutory cite of the offense alleged and a description of the incident Booking and Custody: Status of individuals in local detention facilities Investigation Data: Gang affiliations, victim and witness information, vehicle information, predatory offender location and classification Prosecution and Defense Prosecution Charge: Prosecutors' formal criminal charges following investigation

Pre-Trial Diversion: Status and outcomes of individuals' compliance with pre-trial diversion programs, such as chemical dependency treatment, counseling, and restitution Courts Court Schedule and Hearing Dates

Warrants: For arrest or to appear in court Criminal Case Dispositions: Case outcomes (convictions, dismissals) and final offense level (felony, gross misdemeanor, misdemeanor) Sentences and Conditions: Length of confinement, fines, or other sanctions

Post-Sentence Court Appearances: Case outcomes and sentences following probation violations Probation and Detention Probation Status: Length and terms of probation; history of violations or compliance

Custody Status: Location of offenders detained in local or state facilities while on probation

SOURCE: SEARCH, Planning the Integration of Justice Information Systems: Developing the Justice Information Exchange Model( Sacramento, CA: SEARCH/ The National Consortium for Justice Information and Statistics, 2002).

In the early 1990s, policymakers reached a consensus that Minnesota needed to improve criminal justice information sharing.

4Laws of Minnesota (1993), ch. 266, sec. 33.

The process of building a statewide model to guide integration efforts moved further still in 1999 when the Legislature funded a project in Hennepin County to develop a plan for sharing information between the county, its municipalities, and the state. When Hennepin County completed its work in 2000, the state accepted the plan as a statewide integration model. Responsibility for maintaining and improving it then moved to the Minnesota Department of Public Safety.

BACKGROUND 7 Table 1.3: Statewide Criminal Justice Information Systems Available in the Early 1990s System and Source Description Bureau of Criminal Apprehension (BCA)

Computerized Criminal History System (CCH) Statewide repository for local law enforcement, incarceration, and court data on adults' interactions with the criminal justice system Data limited to: arrests, charges, case dispositions, and sentences of adults for felony, gross misdemeanor, and some misdemeanor offenses BCA staff entered criminal history data from fingerprint cards submitted by law enforcement agencies, and if a positive match could be made, linked the new information to an individual's criminal history "Hot Files" Statewide central repository listing (1) wanted or missing persons, and (2) stolen property, such as vehicles, guns, or electronics Access to CCH and Hot Files Through BCA's computer network, about 5,000 users at 300 agencies could access CCH, Hot Files, Minnesota Driver Vehicle Services databases, and certain national criminal justice information databases State Court Administrator's Office Total Court Information System (TCIS) Statewide case management information system for district, appellate, and supreme court activities including defendant information, case dispositions, sentences and conditions, and post-sentence court appearances System access limited to court staff System does not link or compile information about individuals

Hennepin and Scott County criminal courts are not included, requiring separate processes to transfer their data to the BCA for compiling criminal histories

NOTE: As of January 2004, all three of these systems are still being used. The Computerized Criminal History system has been updated several times. TCIS is being replaced with a new court information system. SOURCE: Bureau of Criminal Apprehension and State Court Administrator's Office system descriptions.

In the mid-1990s, the state started investing in criminal justice information system improvements.

In 2001, state investments to improve criminal justice information systems increased significantly, and the state designated its effort to improve and integrate criminal justice work processes and information systems as "CriMNet." That same year, the Legislature created a central program office and governance structure to coordinate and oversee the CriMNet program.

8CRIMNET "Hot Files" Bureau of Criminal Apprehension

Total Court Information System

State Detention Facilities

Department of Corrections

State facility custody data sent electronically

Courts integrated into one system

Adult criminal case dispositions sent electronically

Fingerprint cards and arrest data submitted manually

Public Defender Information Systems

Data on wanted and missing persons and stolen property

Computerized Criminal History System Bureau of Criminal Apprehension

About 300 agencies and 5000 users had electronic access to the repositories

City and County Prosecution Systems

County Corrections Information Systems

Local Detention Facility Systems

District and Appellate Courts

Law Enforcement Information Systems

Figure 1.1: Criminal Justice Information Systems Integration, Early 1990s

NOTE: See Table 1.3 for descriptions of the three central data systems. SOURCE: Office of the Legislative Auditor analysis of CriMNet program documents.

Minnesota's plan called fo r integrating data held in separate city and county systems.

In 2001, the program to integrate criminal justice data was first called "CriMNet."

CRIMNET CriMNet is Minnesota's program to integrate criminal justice information. It involves deciding what information criminal justice professionals need, identifying barriers that prevent sharing of that information, formulating work rules and data definitions, and creating the technical structures (such as software, networks, and interfaces between systems) that make it possible to access and move data across organizational boundaries. In this section, we describe the CriMNet program according to its various aspects, including: (1) the results CriMNet is expected to achieve, (2) the integration model it is using, (3) the roles and responsibilities of various state and local entities, and (4) integration projects that have been implemented under the CriMNet program thus far.

BACKGROUND 9 Table 1.4: Completed Criminal Justice Information System and Integration-Related Projects, Initiated 1995-97 Years in Project Description Development Targeted Misdemeanor Criminal History Expanded the Bureau of Criminal Apprehension (BCA) Computerized Criminal History system to include certain misdemeanor offenses pertaining to domestic assault, harassments, and DWI violations. 1995-2000

Orders For Protection System Developed a statewide database to allow law enforcement and probation officials to track information about domestic abuse victims, offenders, and orders for protection. 1995-2000

National Instant Check System Developed a system to comply with the federal Brady Handgun Violence Prevention Act, which requires a criminal background check for handgun buyers. 1995-2000

Juvenile Criminal History Database Modified the BCA's Computerized Criminal History system to add criminal histories for juveniles prosecuted for felonies and gross misdemeanors. 1995-2002 Automated Fingerprint Identification System Expansion— Phase I Upgraded the BCA's fingerprint system to increase its capacity and to comply with standards for the FBI automated fingerprint identification system. 1996-1997

Computerized Criminal History Standardization Standardized the Computerized Criminal History record format to comply with national standards and to make it easier to read and interpret the results of criminal history searches. 1996-1999

Gang File Created a statewide system to compile information about offenders' gang status, affiliation, and other gang-related data. 1997-1998 Law Enforcement Message Switch (LEMS) An electronic switch that facilitates the exchange of information among systems, this project upgraded the LEMS from a proprietary system to a vendor's packaged system, allowing for future enhancements. 1996-1998

NOTE: We identify these projects as completed in that the systems are now in a maintenance phase. Some systems will require enhancements as system integration and technology evolve and equipment reaches the end of its useful life. SOURCE: The Criminal and Juvenile Justice Information Data Group, Master Project Plan: Information Technology Projects for the Criminal & Juvenile Justice Community, Revised( St. Paul, December 1998); and Bureau of Criminal Apprehension project descriptions.

Vision and Goals As discussed above, Minnesota's vision for integrating criminal justice information systems has been evolving for over a decade. The authoritative statement of CriMNet's vision and goals as currently conceived is conveyed in the program's strategic plan, dated September 2003. 5 It sets forth the results CriMNet is intended to achieve in the long term and its objectives for the next three to five years.

CriMNet's intended results are: To accurately identify individuals; To make sure that criminal justice records are complete, accurate, and readily available;

To ensure the availability of an individual's current status in the criminal justice system; To provide standards for data sharing and analysis; and To maintain the security of information. As we said earlier, achieving this vision has been, and will continue to be, an incremental process. As shown in Table 1.5, CriMNet officials have organized their work to achieve these results around two goals and related objectives. The f irst goal is to develop an integration "blueprint." Like a blueprint for a building that includes such things as the structural design, electrical wiring diagrams, and specifications for materials, the CriMNet blueprint is a set of diagrams, instructions, and guidance for criminal justice information integration. It should include, among other things: (1) a prioritized set of requirements from criminal justice professionals stating the specific information they need at various decision points; (2) definitions of the data that need to be collected or shared when a criminal justice event, such as an arrest or conviction, occurs; 6 and (3) common work practices for gathering, recording, and sharing information.

CriMNet's second goal is to implement the blueprint through specific state and local integration projects. CriMNet's objectives in this regard are aimed at (1) making needed information available across jurisdictions, and (2) developing a statewide approach to accurately identify individuals and link their criminal justice records across systems. CriMNet's approach is to positively identify individuals using unique information, such as fingerprints, rather than less reliable identifiers, such as name or date of birth. 7 Meeting these objectives requires concurrent activities to establish data accuracy and security standards and to ensure that CriMNet proceeds in compliance with Minnesota's data practice laws.

10 CRIMNET CriMNet adopted a strategic plan in September 2003.

5Criminal and Juvenile Justice Information Policy Group, CriMNet Strategic Plan (St. Paul, September 2003). That this statement of goals and objectives was not clearly articulated until September 2003 is a point of discussion in Chapter 4 regarding how the CriMNet program has been managed since its inception in 2001. 6These criminal justice events are often referred to as "exchange points." An exchange point model, which could be part of an integration blueprint, shows criminal justice events, the data needed to support a criminal justice professional's decisions during that event, and the flow of information needed from other sources and resulting from the event.

 

Compliance with data practice laws has been a long-standing, and still unresolved, concern for CriMNet. CriMNet is considered a statewide data system subject to the Minnesota Government Data Practices Act, which regulates the handling of data by executive branch agencies and local governments. 8 Under the act, data on individuals, including criminal justice data, are classified as either public, private or confidential. Public data are available to anyone for any reason. Private data are available only to the data subject and to anyone authorized by the data subject or law to see them. Data classified as confidential are not available to the public or to the data subject (other than to know the data exist). CriMNet raises a number of issues regarding how shared data will be classified and the process through which data subjects may access information about themselves. 9 CriMNet integrates data from many sources, and data held in each of these systems can have similar or competing data practice classifications. For example, arrest data are public when held by local police departments or the courts, but are classified as private when transferred to Bureau of Criminal Apprehension databases. In addition, judicial branch data are not subject to the Data Practices Act but to court rules of access. The Data Practices Act currently does not provide for the treatment of data that move from the judicial branch to another government agency . These conflicting rules and data classifications need to be resolved. As we discuss later in the report, uncertainty regarding compliance with the Data Practices Act has slowed some aspects of CriMNet system design and has affected some local jurisdictions'willingness to share data through CriMNet.

BACKGROUND 11 Table 1.5: CriMNet Strategic Goals and Objectives Goal 1: Develop a blueprint for the integration of criminal justice information Objectives: Develop and maintain a statewide integration plan that includes and incorporates local planning and implementation efforts, paying particular attention to the collaborative reengineering of business practices. Provide expertise and assistance to facilitate the development of state and local integration plans and services.

Develop technology standards. Improve the efficiency and effectiveness of criminal justice processes. Identify and remove barriers to data sharing within the criminal justice community.

Goal 2: Make available consolidated, complete, and accurate records of an individual's interaction with criminal justice Objectives: Integrate select state and local criminal justice information through collaboration with agencies.

Develop a statewide approach to accurately identify individuals and to link records based on the business need. Comply with data practices laws and court rules of access. Develop and monitor data quality standards. Provide for appropriate security of information.

SOURCE: Criminal and Juvenile Justice Information Policy Group, CriMNet Strategic Plan( St. Paul, September 2003).

Compliance with state data practice laws has been a long-standing, and still unresolved, concern for CriMNet.

7Fingerprinting is considered a "biometric" form of identification, meaning that it is unique to the individual. Other biometric identifiers might include DNA or retinal scans. But at this time, the state uses fingerprints to biometrically link criminal history records. Use of biometric identification is an important aspect of CriMNet because accurate identification of individuals is key to creating accurate criminal histories. Identifiers such as name, race, and date of birth provide only a partial, and sometimes inaccurate, picture. Photographs can help identify individuals but, alone, also are not considered suff icient to establish positive identity.

8Minn. Stat . (2002), 13.

Integration Model Under the CriMNet integration model, the data to be shared does not reside in a single location. Rather, CriMNet is called a "system of systems" in which jurisdictions retain the right to design, operate, and maintain the information systems and data they need, but within certain parameters. These parameters define such things as network capabilities, data standards, and security requirements. Local jurisdictions are not expected to share all information with other CriMNet participants, only the data elements that CriMNet has deemed necessary. Local participation in CriMNet is voluntary.

It is important to note that the information-sharing model described below is the vision of how criminal justice professionals will be able to share information. As we discuss in detail throughout the rest of the report, implementing this model will happen through a series of incremental steps. Specifically, in Chapter 2 we discuss the current technical capabilities of the Integration Backbone and the state's progress in using it to connect data from various information systems.

According to Minnesota's model, CriMNet data sharing will happen through a technical infrastructure, akin to a computer network, that distributes data among many agencies and systems. The central component of this technology is the "Integration Backbone." 10 It will provide the means of organizing information, with an "index" of data elements being shared and "pointers" needed to locate and access the data at various locations. 11 Along with the Integration Backbone, CriMNet will rely on a secure computer network (called the Criminal Justice Data Network, or CJDN) operated by the Bureau of Criminal Apprehension for secure transmission of data from system to system.

12 CRIMNET Under CriMNet, local jurisdictions maintain their own criminal justice information systems and control access to their data.

9Criminal and Juvenile Justice Information Policy Group, 2003 Report to the Legislature (St. Paul, 2003), Appendix D.

10Quite often, the term "CriMNet" is used narrowly to refer to this information system. In our report, we use CriMNet more broadly to refer to the entire program, which includes governance, work practices, standards, and technology that together comprise criminal justice information integration efforts. 11As envisioned, the registry will accept or create unique identifiers for individuals, incidents, and cases, and the unique identifier for individuals will be based on biometric identification. The locator service will work like an Internet search engine. If a user enters a suspect's name, the locator will provide links to the systems that have records matching that name.

Participating state, county, and municipal criminal justice agencies will be connected to the Integration Backbone through a "hub." A hub is a combination of hardware and software that provides the access, rules, and data formats needed to register and share information with the Integration Backbone. Local jurisdictions will have control over access to their own data and determine how and when data are made accessible to other users. Statewide information systems or data repositories can also be connected to the Integration Backbone via a hub.

Under CriMNet's information sharing model, criminal justice professionals will be able to exchange information various ways. For example, as shown in Table 1.6, users could search CriMNet's component systems for records that match specified criteria. This search capability is the type of information exchange that most closely reflects the essential intent of CriMNet— being able to obtain complete information about individuals'criminal histories and status in the criminal justice system. Other envisioned features would provide enhanced information sharing capabilities and efficiencies. For example, the "push" capability can automatically transmit data collected at the point of origin to the information system associated with the next step in the criminal justice process (for example, sentencing information could be sent to the detention facility where the offender will be incarcerated).

BACKGROUND 13

Table 1.6: Types of Criminal Justice Information Exchange Envisioned for CriMNet Service Description Examples Search A request for information that meets certain criteria Determine whether an individual is wanted by another jurisdiction, has charges pending in another jurisdiction, is currently on probation, has prior convictions, or has served time in a correctional facility.

Push Automatic transfer of information to another system

Pull Automatic extraction of information from another system

Publish Information made available to a wide audience of recipients

Subscribe A request to receive certain information automatically when a particular event occurs

SOURCES: The Macro Group, Inc./Labrynth Consulting, Inc., Integration Backbone Logical Design Report (St. Paul: Minnesota Department of Public Safety, 2001), 19-21; SEARCH, Integration in the Context of Justice Information Systems: A Common Understanding( Sacramento, CA: SEARCH/The National Consortium for Justice Information and Statistics, 2001), 10.

 

Report arrest information, fingerprints and arrest photos to the Bureau of Criminal Apprehension. Send arrest information to the prosecuting attorney's office for use in its case intake process.

Create a correctional facility information system record with information captured in the pre-sentence investigation together with court sentencing information.

Paper or electronic posting of scheduled court events or public criminal history records.

Notify a probation officer by e-mail if a probationer is arrested anywhere in the state. Automatically notify the presiding judge if an individual violates the terms of a suspended sentence.

Capability to do a statewide search of criminal justice information systems is an important CriMNet goal.

The Integration Backbone technology is designed to support CriMNet data practice and security standards. For example, each user will be given a security profile that defines the types of systems and data that the user may access, and the Backbone can be programmed to ensure the user is granted access only to those data and systems. Similarly, the Integration Backbone can be programmed to allow or deny access to certain data based on their classifications.

Governance CriMNet is a multi-jurisdictional effort, and the governance structure reflects this. As illustrated in Figure 1.2, CriMNet governance is defined by a set of relationships among various agencies and branches of government. The Legislature established the Policy Group, comprised of executive and judicial branch leaders, to play a central role in planning and executing Minnesota's approach to integrating criminal justice information. A central program office and other state and local entities implement the integration strategy through specific projects. Below, we describe the roles and responsibilities of these groups, and in Chapters 3 and 4, we discuss in detail the extent to which they have met their various responsibilities.

Policy Group CriMNet Office Strategic Direction

Department of Public Safety

Day-to-Day Management

Task Force

Advise

State and Local Jurisdictions

Collaboration

Integration Projects Project Management

Program Oversight

State Courts Bureau of Criminal Apprehension Department of Corrections County & Municipal Governments Public Defenders Prosecution

Tactical Direction

Figure 1.2: CriMNet Governance

SOURCE: Office of the Legislative Auditor analysis of Criminal and Juvenile Justice Information Policy Group meeting materials and minutes, June-December 2003.

CriMNet's governance structure reflects the multi-jurisdictional nature of the program.

The Policy Group State law places responsibility for governing CriMNet with the Criminal and Juvenile Justice Information Policy Group (referred to as "the Policy Group" throughout this report). 12 The Policy Group is comprised of eight members, four each representing the judicial and executive branches of state government. For the executive branch, the members are the commissioners of Corrections, Administration, Finance, and Public Safety. The Chief Justice of the Supreme Court appoints the four judicial branch representatives, which in 2004 include the State Court Administrator and three justices representing the Supreme Court, the Appellate Court, and the District Courts. In addition, the law allows the Policy Group to appoint additional, non-voting members, which it has chosen to do. In 2004, the chair and vice-chair of the Policy Group's advisory body (the Task Force, discussed below) sit as non-voting members. The Commissioner of Public Safety is designated as the permanent chair of the Policy Group.

State law gives the Policy Group specific responsibilities regarding how the state approaches integration and how CriMNet projects should proceed. 13 As shown in Table 1.7, the Policy Group is supposed to govern the state's progress toward integrating criminal justice information, determining the types of data that should be shared and the processes through which the data should flow. The Policy Group is to review and prioritize state and local integration project funding requests and report the results to the Legislature. The Policy Group is required to report to the Legislature by December 1 of each year. In its report, the Policy Group must make recommendations concerning any legislative changes or appropriations that are needed to ensure that criminal justice information systems operate accurately and efficiently.

State law authorizes two entities to assist the Policy Group in meeting its many obligations— an advisory task force and a CriMNet program manager. The role of each is described quite generally in statute, with more specific duties having evolved over time.

The Task Force By law, the purpose of the 37-member advisory group, called the Criminal and Juvenile Justice Information Task Force (or more simply, the Task Force), is to assist the Policy Group in developing recommendations for its annual report to the Legislature. 14 In practice, the Task Force does not function quite so narrowly; instead, it serves in a general advisory capacity. 15 In June 2003, the Policy Group approved a new charter for the Task Force that expanded the group's responsibilities beyond those it traditionally held. 16 According to the current charter, in addition to providing insight on CriMNet's long-term strategy, the Task Force is also supposed to monitor the progress of CriMNet projects to ensure that they are being appropriately managed and meeting their objectives.

BACKGROUND 15 Four executive branch officials and four judicial branch officials comprise the CriMNet Policy Group.

12Minn. Stat . (2002), 299C. 65, subd. 1. 13Ibid. The law goes into some detail regarding project management and procurement practices. For example, accepted project management techniques include (1) clear sponsorship; (2) scope management; (3) project planning, control, and execution; (4) cost management; (5) quality reviews; (6) communication management; and (7) use of proven methodology. The procurement process should include the following four steps: (1) defining the required product or service, (2) a request for proposal process to identify potential vendors, (3) competitive selection, and (4) contract administration and closeout. Minn. Stat . (2002), 299C. 65, subd. 1( b) and 1( c).

 

The CriMNet Office CriMNet's authorizing statute allows the Policy Group to hire a CriMNet program manager to manage CriMNet projects and to be responsible for day-to-day

16CRIMNET Table 1.7: Statutory Responsibilities of the Policy Group Successfully complete statewide criminal justice information system integration. Review criminal justice information system funding requests from state, county, and municipal government agencies for compatibility with statewide criminal justice information system standards, and forward the results of this review to relevant Legislative committees.

Study and make recommendations to the Governor, Supreme Court, and Legislature regarding: a model for integrating criminal justice information, roles and responsibilities of various criminal justice jurisdictions, data privacy, data accuracy, and criminal justice information system equipment, training, and funding needs of state and local jurisdictions.

Ensure that CriMNet projects follow generally accepted project management techniques. Ensure that vendor products and services are obtained through an appropriate procurement process.

SOURCE: Minn. Stat. (2002), 299C. 65.

The Policy Group is responsible fo r the successful statewide integration of criminal justice information systems.

A 37-member advisory task f o rce assists the Policy Group.

14Minn. Stat . (2002), 299C. 65, subd. 2( b). The 37 members encompass a broad spectrum of perspectives. By law, each member of the Policy Group or a designee sits on the Task Force. Most of the other members are specified in law by constituency , with a designated authority to appoint the actual Task Force members. The various constituencies include, among others, sheriffs, police chiefs, county and city attorneys, public defenders, community corrections and probation officers, district court judges, and court administrators. The Task Force also includes one delegate each from the Minnesota House and Senate and four members of the public, one of whom has been a victim of crime and two who are members of the private business community and who have expertise in integrated information systems. The law also instructs appointing authorities from state and local jurisdictions to choose delegates who have expertise in integrated data systems or best practices. The Policy Group approves Task Force appointments.

15In its annual report for 2003, the Policy Group recommended that the Legislature amend the language authorizing the Task Force to better reflect this broader advisory role. Criminal and Juvenile Justice Information Policy Group, 2003Report to the Legislature (St. Paul, 2003), 33.

16Minnesota Department of Public Safety, Criminal and Juvenile Justice Task Force Charter (St. Paul, June 2003).

CriMNet operations. 17 The 2001 Legislature appropriated funds for the program manager to hire staff, thus creating the CriMNet Office. 18 In general, the CriMNet Office is responsible for developing the state's integration blueprint and supporting the state and local entities that implement the blueprint through integration projects. As shown in Table 1.8, the CriMNet Office undertakes a broad range of activities to meet its program management responsibilities. To do its work, the CriMNet Office often forms teams with participating state and local agencies to work on specific tasks, such as reviewing grant applications or resolving conflicting work procedures.

Although the CriMNet Office is an arm of the Policy Group, it operates within the Department of Public Safety. The department provides support services, such as human resources and procurement, and it acts as the CriMNet fiscal agent. As we discuss in more detail in Chapter 4, the department recently started playing a stronger role in supervising the CriMNet Office's day-to-day operations.

Integration Projects As described above, the CriMNet Office, on behalf of the Policy Group, has several program management responsibilities related to integrating criminal justice information systems. But, the primary responsibility for planning and managing most integration projects rests with the state and local agencies that "own" the systems.

As shown in Table 1.9, state agencies implementing CriMNet projects include the Department of Corrections, the Department of Public Safety (the Bureau of Criminal Apprehension and the CriMNet Office), and the Courts. This list of active and completed projects includes some initiated after the CriMNet program was officially established in early 2001. It also includes projects that started earlier, but were pulled under the CriMNet umbrella because they were clearly

BACKGROUND 17 Table 1.8: Select CriMNet Office Responsibilities Develop and maintain the CriMNet strategic plan. Monitor and track progress of state and local integration projects and coordinate common issues among them.

Communicate with criminal justice jurisdictions, the Legislature, and the public. Define user requirements— the information sharing that criminal justice professionals need to do their jobs, along with their priorities among various needs.

Identify barriers to information sharing and define new processes through which information will be shared. Identify the data that will be shared and develop standards regarding data format and accuracy. Provide technical support to state and local agencies. Define security standards to protect agency systems and the CriMNet network.

SOURCE: Office of the Legislative Auditor analysis of CriMNet Office documents.

The CriMNet Office manages day-to-day program operations.

State and local agencies plan and implement integration projects related to their information systems.

17Minn. Stat . (2002), 299C. 65, subd. 1( b). 18Laws of Minnesota (1sp2001), ch. 8, art. 4, sec. 10, subd. 3.

CRIMNET Table 1.9: State Agency CriMNet Projects, as of January 2004 Estimated Agency and Project Description Cost to Date Bureau of Criminal Apprehension Livescan Hardware and software system to take digital fingerprint images to replace the traditional ink and roll fingerprint cards $ 4,949,000 Cardhandler A technology interface that allows electronic fingerprint files to be transmitted to or from the Bureau of Criminal Apprehension 645,000

Criminal Justice Data Network (CJDN) Upgraded the state criminal justice network to increase its capacity 2,512,000 Predatory Offender Registration System (POR) A central repository for information on every registered predatory offender in Minnesota 999,000 a

Minnesota Repository of Arrest Photos (MRAP) Database of digital photographs with corresponding descriptive data taken at the time of arrest and booking 3,570,000 Criminal History Suspense File A set of projects to correct and prevent incomplete criminal history records 1,796,000 b CriMNet Office Integration Backbone Technical infrastructure to connect criminal justice information systems and to organize and distribute data among jurisdictions 7,994,000 Department of Corrections Statewide Supervision related to criminal justice integration. 19 In Chapter 2, we discuss how these projects have furthered criminal justice integration efforts. In Chapter 3, we discuss the extent to which the projects met schedule, cost, and scope expectations.

System (S 3 ) A statewide system to integrate state and local probation, jail, and prison records 2,494,000 c

Sentencing Guidelines Worksheet d An electronic means of transferring sentencing information to appropriate agencies 148,000 Courts Court Web Access (CWA) A system to access select court data, including non-confidential adult criminal case information from felony, gross misdemeanor, and select misdemeanor court cases 578,000

Minnesota Court Information System (MNCIS) A statewide offender-based database for records on individuals appearing before the courts, including criminal, probate, civil, and other cases 18,077,000 e

Statute Table A standard, electronic listing of criminal justice statute citations intended to be integrated into criminal justice systems Unknown f

NOTE: This table excludes CriMNet projects currently in preliminary planning stages. Estimated costs are as reported by the agencies through project completion or, for ongoing projects, through December 2003 unless otherwise noted. The status of these projects is discussed in Chapter 2. a The estimate does not include some hardware and operational costs.

b Estimate includes fiscal years 2002-03 only. c The estimate excludes 1996-97 planning expenditures and includes grants to counties for information system improvements needed to support local connection to the Statewide Supervision System. d This was a joint project with the Sentencing Guidelines Commission. The cost estimate does not include additional vendor programming costs incurred for integrating the worksheet into the Statewide Supervision System. e Project is expected to be completed in fiscal year 2006 at a total cost of $32,000,000.

f Court staff could not provide detailed cost data for this project. The courts received a $50,000 grant in 2000 from the Board of Government Innovation and Cooperation for work on this project.

SOURCE: Office of the Legislative Auditor analysis of project planning, status, and cost data from the departments of Public Safety and Corrections, the Sentencing Guidelines Commission, and the State Court Administrator's Office.

 

The Legislature has authorized a grant program to support local criminal justice systems integration planning and implementation. 20 Grants may go to counties, cities, or local government consortiums for developing an integration plan or implementing one or more integration projects. The grant process involves a number of steps. The Department of Public Safety, with Policy Group approval, publishes grant criteria. Subsequently, counties or other local entities submit proposals. 21 The Task Force considers each grant proposal and forwards a final recommendation to the Policy Group, which has final approval authority.

By law, grant recipients must provide matching funds to help pay for up to one-half of the project costs. 22 The Policy Group determines the specifics regarding the match, such as the proportion of total costs, but its policy must stay within certain parameters. For example, the law states that the matching fund obligation may be met with "in kind" resources and specifies that local operational or information technology staffing costs may be considered as meeting the match requirement. The law also requires the Policy Group match policy to be applied consistently to all recipients.

Originally, five Minnesota counties were designated as CriMNet "pilot counties," and each received planning grant funds. But, as shown in Table 1.10, CriMNet implementation grants went primarily to four counties: Anoka, Dakota, Hennepin, and St. Louis. 23 The types of local projects funded include intra-county integration, such as the Anoka County project to integrate its dispatch and police record systems, and enhancements to facilitate integration, such as the Dakota County project to improve its countywide network. In Chapter 4, we discuss in more detail how the criteria for CriMNet grants have changed and some of the issues involved in allocating grant funds.

BACKGROUND 19 Local governments may receive state and federally-funded grants to support integration planning and implementation.

19We did not include planned CriMNet projects that are not actively underway or criminal justice information systems that may eventually be linked via the Integration Backbone, such as the state's database of driver's license information. Also, some agencies may have integration activities underway that are not reflected in our list. 20Minn. Stat . (2002), 299C. 65, subd. 5-9. 21The law includes a highly detailed list of components that must be included in requests for integration planning grants. For example, the grant proposal must, among other things, identify the data sharing problems that drive the need for an integration plan, certify that the requesting officials have consulted with local criminal justice practitioners to identify these problems, attest that the plan will integrate the major criminal justice functions (such as incident reporting), and certify that the f inal plan will include a specified set of products (such as a technology model). Minn. Stat . (2002), 299C. 65, subd. 6( a).

22Minn. Stat . (2002), 299C. 65, subd. 8. 23Ramsey County was the fifth pilot county. It applied for implementation grant funds in 2002, but chose not to complete the grant process.

CRIMNET Table 1.10: CriMNet Local Integration Grant Projects, 2001-04 Grant Recipient and Project Description Grant Award Amounts Anoka County Computer-Aided Dispatch/ Law

Enforcement Record Management System (CAD/ RMS) Integration Project (1) integrates local police departments' common records management system with the county's dispatch system, and (2) will allow expanded use of mobile data computers in squad cars to enter and retrieve data.

Grant: $ 119,000 Local Match: 229,000 Total: 348,000

Detention Information System Project to create a single information system for the county's three detention facilities, including arrest and booking data and inmate records. Grant: $ 750,000 Local Match: 817,000 Total: 1,567,000 Dakota County CAD/ RMS/ Mobile Communications

Systems Integration Project updates and integrates local law enforcements' Computer Aided Dispatch, mobile communication, and records management systems, thereby relying less on the use of radio for transmitting data and focusing on capturing information at the source and sharing data electronically.

Grant: $ 255,000 Local Match: 255,000 Total: 510,000

Expand County Integration Network System Project expands local law enforcements' ability to search and report information via the web, and automates the jail booking and reporting process. Grant: $ 800,000 Local Match: 810,000 Total: 1,610,000 Anoka and Dakota Counties Joint Attorney Case Management

System A project to develop a common prosecuting attorney case management system, using web technology to exchange information between counties. The design is expected to facilitate eventual integration with the state's systems.

Grant: $ 600,000 Local Match: 694,000 Total: 1,294,000

St. Louis County Criminal Justice Integration Project Project (1) integrates data from the various law enforcement and prosecution systems from St. Louis County and surrounding counties, and (2) creates a repository accessible through the Internet and wireless technology.

Grant: $ 800,000 Local Match: 800,000 Total: 1,600,000

Hennepin County Planning projects Planning projects to assess user needs and make "build or buy" decisions for new Minneapolis City Attorney case management and county workhouse records management systems. Also, work practice reengineering project for county arrest and booking process.

Grant: $ 420,000 Local Match: 420,000 Total: 840,000

Minnesota Counties Computer Cooperative Court Services Tracking System Project creates a new system for case management and tracking of clients under the supervision of 86 counties' Court Services departments. Grant: $ 640,000 Local Match: 641,000 Total: 1,281,000

Local Government Information Systems (LOGIS) Public Safety Information System Integration For participating jurisdictions, the project will integrate certain criminal justice systems into a common network and standardize processes for reporting information and statistics on crime and arrests. Through the new network, users will be able to search the other jurisdictions' systems and access the Integration Backbone.

Grant: $ 390,000 Local Match: 390,000 Total: 780,000

NOTE: The table excludes $1,300,000 state integration planning grants awarded during the 2000-01 biennium to each of the counties listed in the table and to Ramsey County. The CriMNet Office was not able to provide complete cost data for these planning grants. The table also excludes Department of Corrections grants to local jurisdictions for integration with the Statewide Supervision System. SOURCE: Office of the Legislative Auditor analysis of CriMNet Office grant documents.

INTEGRATION COSTS For a number of reasons, it is difficult to precisely tally actual CriMNet costs. First, the state does not have a clear definition of the types of criminal justice information system spending that should be included. For example, to some stakeholders, integration should include only the cost of connecting separate information systems; to others, the definition should also include system enhancements needed to facilitate integration. Still other stakeholders distinguish between systems spending that would have happened anyway and "new" integration spending. Second, CriMNet is a term used to describe the state's program to integrate criminal justice information from 2001 forward, but the state has been investing in integration since the mid-1990s. As a result of these differences in perspective, stakeholder estimates of total spending on CriMNet vary widely, ranging from $60 million using a narrow definition to nearly $180 million using a more expansive definition.

In this section, we describe CriMNet funding sources, and then we present fiscal year 1996-2005 estimates of funding designated for general criminal justice information system improvements. Last, we present CriMNet program biennial budget data for fiscal years 2002 through 2005, and we discuss why these program budget amounts differ from the larger appropriation amounts for the same biennia.

Our discussion of funding also relies on our office's financial audit of CriMNet expenditures, conducted at the Legislature's direction. That report, entitled CriMNet Financial Audit, discusses how CriMNet is funded and problems tracing this funding to the state's accounting system. It also describes expenditures by category, including administrative services, professional/technical contracts, and commodities. 24

Funding Sources CriMNet is funded from various sources. The CriMNet Office and state integration projects are financed through a combination of state appropriations and federal grants. Local integration efforts are funded through grants from both state and federal sources and through local government operating funds.

Although the Policy Group presents the Legislature with biennial spending plans for CriMNet, it does not directly receive or disburse funds. State agencies with criminal justice information systems include requests for new or ongoing projects in their budget submissions to the Legislature. The Legislature then appropriates money directly to agencies. State and federal funds for local integration grants go to the Department of Public Safety, which serves as fiscal agent for the CriMNet Office. Grants are disbursed as approved by the Policy Group.

BACKGROUND 21 It is difficult to tally CriMNet's total costs to date.

Stakeholders disagree over what should be counted.

24Minnesota Office of the Legislative Auditor, CriMNet Financial Audit( St. Paul, 2004).

Cost Estimates for Criminal Justice Information System Improvements, Fiscal Years 1996-2005

State funding for CriMNet is generally given through appropriations designated as being for "criminal justice information system improvements." With our office's financial audit division, we examined appropriations for fiscal years 1996-2005 to estimate how much has been allocated over the long term for criminal justice information integration (both before and after CriMNet was officially designated as a state program). We included appropriations to state agencies and the courts, state appropriations for grants to local units of government, and federal grant awards that we could clearly identify as being for criminal justice information technology. 25 We did not include three other sources of funding because data were not readily available: (1) direct local government spending, (2) grants directly from the federal government to local governments, and (3) state agency spending from general operating funds.

As shown in Table 1.11, for fiscal years 1996 through 2005, state and federal funding for improving and integrating Minnesota criminal justice information systems totals nearly $180 million, with the state share exceeding $135 million. 26

22 CRIMNET

Table 1.11: Estimated State and Federal Funding for Criminal Justice Information Systems Improvements by Biennium, FY 1996-2005 Fund Source 1996-97 1998-99 2000-01 2002-03 2004-05 Total State $2,474,000 $12,804,000 $37,983,000 $45,072,000 $39,343,000 $137,676,000 Federal 2,373,000 2,113,000 3,556,000 12,042,000 a 20,904,000 40,988,000

Total $4,847,000 $14,917,000 $41,539,000 $57,114,000 $60,247,000 $178,664,000

NOTES: We included those appropriations to state agencies or the courts in which (1) legislative language specified criminal justice systems improvements or "information systems integration," (2) legislative language specified the project and/ or funds are subject to oversight or approval by the Criminal and Juvenile Justice Information Policy Group, or (3) the project is managed as part of CriMNet. The state appropriation amounts include funds that continue on from earlier appropriations for ongoing information system maintenance, operation, or improvement. We added federal grant awards that we could clearly identify as being for criminal justice information technology.

As part of the CriMNet financial audit conducted by our office, audit staff traced these appropriations and federal grants to the state's accounting system. Their analysis identified state and federal funds totaling approximately $106 million as of December 31, 2003. They could not separately identify some state appropriations that went to agencies' general operating accounts. In addition, the financial audit funding totals include federal receipts drawn down from federal grants, not the total federal grant award.

a Does not include a $4 million federal grant that went directly to Hennepin County. SOURCE: Office of the Legislative Auditor analysis of Minnesota Laws and Minnesota House Fiscal Analysis Department data on criminal justice information technology investments.

25We included those appropriations to state agencies or the courts in which (1) legislative language specified criminal justice systems improvements or "information systems integration," (2) legislative language specified the project and/ or funds are subject to oversight or approval by the

Criminal and Juvenile Justice Information Policy Group, or (3) the project is managed as part of CriMNet.

26Of the $180 million, our office's financial audit staff were able to trace approximately $106 million to the state's accounting system, as of December 31, 2003. They could not identify some appropriation amounts that went to agencies'general operating accounts, and their accounting of federal funding was limited to amounts drawn down from federal grants, not the total federal award. See Office of the Legislative Auditor, CriMNet Financial Audit.

State appropriations ranged from a low of $2.5 million for the 1996-97 biennium to a high of about $45 million for the 2002-03 biennium. The significant increase in funding beginning in the 2000-01 biennium reflects the shift from integration planning to implementation of integration projects, including a new statewide information system for the courts and the other projects shown in Table 1.9.

CriMNet Program Costs, Fiscal Years 2002-2005 We reviewed CriMNet budgets to assess changes in CriMNet program costs. As shown in Table 1.12, CriMNet program costs increased from about $25 million for the 2002-03 biennium to about $30 million for the current biennium. State funding decreased from about $22 million to $19 million between the two biennia, but federal funding increased by over $8 million— more than a three-fold increase. The CriMNet Office budget includes funding for the Integration Backbone project, CriMNet Office operations, and funds (such as grants) that pass through to other entities. Of the approximately $10.9 million in federal funding shown for the 2004-05 biennium, for example, about $5.6 million is designated for local grants, most of which has been committed to approved grant projects. 27

The criminal justice information system funding amounts shown in Table 1.11 differ quite noticeably from the CriMNet program budget for the 2002-03 and 2004-05 biennia shown in Table 1.12. These differences occurred for several reasons. Primarily, the state appropriation amounts include funds that continue on from earlier appropriations for ongoing information system maintenance, operation, or improvement. 28 In large part, these continuing funds are not included in the CriMNet program budget. Second, agencies may not consider all of an information system appropriation to be for CriMNet. For example, the Department of Corrections includes funding for the Statewide Supervision System in its CriMNet budget but explicitly excludes spending on the department's operational technology needs. Third, some appropriation or federal award amounts are designated in one fiscal year, but agencies may draw the funds down over a multi-year period. The CriMNet budget, for example, only shows the federal award amount used in a given biennium, not the total amount awarded.

BACKGROUND 23 Table 1.12: CriMNet Program Budget by Biennium, FY 2002-05 FY 2002-03 FY 2004-05 CriMNet Office State $ 4,975,000 $ 5,086,000 Federal 2,532,000 10,880,000 Total CriMNet Office $ 7,507,000 $15,966,000 Bureau of Criminal Apprehension $ 1,796,000 $ 1,215,000 Department of Corrections 1,227,000 1,060,000

Courts 14,344,000 11,640,000 Total CriMNet $24,874,000 $29,881,000

NOTES: All federal funds for state and local projects and state funds for local government grants are included in CriMNet Office totals. Amounts shown for the 2002-03 biennium are actual expenditure totals; those for the 2004-05 biennium are budgeted amounts. Bureau of Criminal Apprehension, Department of Corrections, and Court funding is from the state general fund.

SOURCE: Criminal and Juvenile Justice Information Policy Group, 2003 Report to the Legislature (St. Paul, 2003), and CriMNet monthly status report, November 2003.

The CriMNet program budget totaled about $55 million fo r f iscal years 2002-05.

27Some stakeholders are concerned that the courts have received a disproportionate share of limited funds for criminal justice technology projects. The courts include funding for the Minnesota Court Information System (MNCIS) in its CriMNet budget, although only some of the system's components are specific to criminal justice-related processes. For example, MNCIS also includes information for probate, civil, and housing court matters. We do not make a judgment regarding the allocation of CriMNet resources.

28These flow-through amounts are sometimes called appropriation "tails."

Statewide Progress SUMMARY Over the past decade, the state has made significant progress in its ability to track offenders' movements and criminal activities around the state. Through a series of projects designed to integrate previously fragmented data and systems, the state has improved criminal justice professionals' access to timely, statewide information. State agencies are currently in the midst of implementing two key information systems that will improve the state's ability to build accurate criminal histories. Criminal justice professionals across jurisdictions can access these systems. However, local law enforcement and prosecution professionals, who do not have their own statewide information systems, still face challenges in obtaining and sharing important investigative data. Absent additional statewide systems, further integration of these data will depend on the ability and willingness of local jurisdictions to link their information with the state. But, at this time, the state does not have a well-formed plan to integrate more local information systems into CriMNet.

As we described in Chapter 1, Minnesota made some early, but limited, progress toward statewide access to criminal justice information in the mid-1990s. Driven by the need to better track offenders' movements and criminal activities around Minnesota, the state's interest in integrating criminal justice information has grown and is supported by a significant financial investment.

In light of the state's commitment, this chapter addresses the following question: To what extent has Minnesota progressed toward its goal of statewide, integrated criminal justice information?

To answer this question, we relied to a great extent on our evaluation of the specific CriMNet projects discussed in Chapter 3 and CriMNet program management, as discussed in Chapter 4. We interviewed staff from legislative research offices, the State Court Administrator's Office, the departments of Administration, Corrections, and Public Safety, including the CriMNet Office, and representatives of local criminal justice agencies. We reviewed CriMNet planning documents and updates, reports to the Legislature, and documents describing the purpose and scope of CriMNet projects.

We synthesized this information to assess progress in two general areas: (1) improved access to information about individuals in the criminal justice system; and (2) development of complete, accurate information about individuals' criminal histories. We discuss progress in these areas in the first part of this chapter. In the second part, we discuss remaining gaps in information access and systems development.

PROGRESS TOWARD INTEGRATING CRIMINAL JUSTICE INFORMATION

As discussed in Chapter 1, integrating criminal justice information will not happen in a single step or with a single technology solution. Instead, integration requires a long-term, incremental approach. Many of the state's early investments in criminal justice information systems included system upgrades and enhancements that served as a foundation for future integration. Recent CriMNet projects resulted in new or enhanced systems within certain functions, such as corrections, that are of value to other professionals and other jurisdictions as well. Still other CriMNet projects have resulted in direct, system-to-system, connections. Considering all of these steps together, we found that:

Since the mid-1990s, the state has made significant progress on several fronts toward CriMNet's goal of statewide, integrated criminal justice information. These accomplishments provide direct benefits to users statewide and include: (1) improving criminal justice professionals'access to timely, statewide information by compiling data from various jurisdictions into centralized systems and data repositories; (2) enhancing system capabilities for transmitting data and querying these repositories; and (3) improving the accuracy of offenders' criminal histories. Collectively, these investments helped create eff iciencies throughout the criminal justice system.

Access to Information Over the past decade, improvements in access to data for criminal justice professionals were achieved through investments in statewide information systems, creation of data repositories, and improvements in underlying technology. In general, we found that:

CriMNet investments have furthered statewide integration of courts and corrections data and have improved access to other repositories of criminal justice information. These technology investments have provided criminal justice professionals with eff icient alternatives to time-consuming, but necessary, aspects of their work— activities done previously by manually processing paperwork or making telephone calls. Compiling information from individual local and state systems into centralized repositories has reduced the time and resources needed to research information about offenders. Integrating multiple systems into one common system also creates more uniform recordkeeping practices, thereby improving the overall consistency and quality of data. Access to more complete information also helps criminal justice professionals make better decisions, such as whether to arrest or release offenders or upgrade or reduce prosecution charges.

26 CRIMNET Integrating criminal justice information systems requires a long-term, incremental approach.

 

As shown in Table 2.1, recently completed CriMNet projects provide technology and systems that are easily accessible to authorized users. These repositories include data compiled from participating agencies across the state, which include most, if not all, 87 counties. These systems are available to authorized criminal justice personnel in all functions, and the number of approved users is growing.

STATEWIDE PROGRESS 27

Table 2.1: CriMNet Projects Providing Improved Access to Criminal Justice Information Project Contribution Access Statewide Supervision System (S 3 ) S 3 is a statewide information system holding current and historical information on juveniles and adults who are or have been on probation, in detention, imprisoned, or in jail. These data were previously held in separate county and detention facility information systems. Integration of data from all counties was substantially completed in June 2003.

S 3 currently has about 7,000 authorized users, including prosecutors; public defenders; and court, probation, and corrections personnel.

Court Web Access (CWA) CWA is a system, completed in 2002, that allows authorized users to search select court information. Through CWA, users may search limited, non-confidential, adult criminal case information from felony, gross misdemeanor, and select misdemeanor cases. Previously these data were available to court employees who had direct access to court information systems and, for court data submitted to the Bureau of Criminal Apprehension, to other professionals who requested records from the Computerized Criminal History system.

About 1,000 criminal justice personnel statewide can access CWA. In addition to court staff, authorized users include prosecutors, public defenders, and probation and corrections personnel.

Minnesota Court Information System (MNCIS) MNCIS will replace the court's outdated, case-based information system. MNCIS data and records will be organized by individual, rather than by case, contributing to better individual criminal history data. MNCIS is also expected to support automated exchange of information with other criminal justice agencies. As of January 2004, MNCIS is being implemented in three counties, with statewide implementation expected in 2006.

As currently envisioned, state court personnel will have direct access to MNCIS. Other criminal justice professionals will obtain MNCIS data through Court Web Access or the Integration Backbone.

Predatory Offender Registration System (POR) POR is a central repository, completed in 2002, for information on about 14,500 registered predatory sex offenders in Minnesota. Criminal justice personnel use it to track and supervise these offenders. Offender records in the repository include additional identification information from other databases, including the Department of Vehicle Services system, Bureau of Criminal Apprehension's Computerized Criminal History system, and national databases containing offender and criminal history data.

Currently, POR has about 3,100 authorized users representing 322 agencies, including 85 of 87 county sheriff offices and all state correctional facilities.

Minnesota Repository of Arrest Photos (MRAP) MRAP is a central database, completed in 2002, that accepts digital photographs and corresponding descriptive and demographic data taken at the time of booking and arrest. MRAP allows criminal justice agencies to create lineups and witness viewing sessions from photographs and enroll unidentified persons into the facial recognition component.

Currently, 71 agencies submit arrest photographs to the database. Capability for more agencies to submit photos will be added as local funding becomes available. Approximately 2,600 authorized users representing 71 law enforcement agencies can access MRAP.

SOURCE: Office of the Legislative Auditor analysis of Bureau of Criminal Apprehension, Department of Corrections, State Court Administrator's Office, and CriMNet Office project descriptions.

Probation and detention data are now available statewide to corrections professionals and other authorized users. In June 2003, the Department of Corrections substantially finished integrating county probation and detention information systems with its Statewide Supervision System, which gives system users a single point of access to adult booking and detention data and adult and juvenile probation information from all Minnesota counties. Although this integrated information is limited to certain data, the value of the system is increased because, in addition to corrections personnel, professionals from other functions, such as prosecutors, judges, and front-line law enforcement personnel, are able to access the system to obtain important and current information about offenders on probation, parole, or supervised release around the state. Before the Statewide Supervision System was available, information on Minnesota probationers, for example, was maintained on a variety of local information systems, and it was often difficult for criminal justice officials to determine offenders' probation terms and conditional release status.

Progress toward better access to courts data has occurred through two key projects— Court Web Access and the Minnesota Court Information System (MNCIS). Court Web Access provides a way for authorized users to view limited information from the courts' Total Court Information System, a case-based records management system. Court Web Access is updated regularly and contains statewide sentencing and conviction records on offenders. Court Web Access will also be able to access statewide data from MNCIS, the courts'new records management system currently being developed. MNCIS further enhances the quality of court data in that it is a person-based records system, rather than case-based. This means that criminal records are automatically added to the offender's file when new charges are added, saving personnel time searching and compiling offenders' criminal files. Once MNCIS is fully implemented, users will be able to obtain statewide data on all misdemeanors, which are not available under the current system. 1 MNCIS is currently being implemented in three counties as the first step toward statewide implementation. All counties are expected to be using MNCIS by 2006, and statewide data will then be available to all users.

Through CriMNet investments, Minnesota has created or improved other information systems and repositories holding specific types of criminal justice information. As discussed in Chapter 1, a decade ago, sources for statewide information were limited to a few systems, including the Bureau of Criminal Apprehension (BCA) Computerized Criminal History system and Hot Files. The state made several later improvements to the quality and content of offenders' criminal histories, including standardizing certain prosecution data, adding certain misdemeanor offenses, and expanding the database to include juvenile offenders. 2 More recent CriMNet projects at the BCA have further expanded access to criminal justice information, as shown in Table 2.1. These new systems rely considerably on information collected and submitted by participating local agencies. For example, state and local law enforcement officers electronically submit to these central repositories identification information about offenders, such as photographs, names, dates of birth, and fingerprints. They also submit information about the status and location of certain offenders convicted of sex offenses (predatory offenders). Authorized users can access these repositories individually through the BCA's secure network.

28 CRIMNET Probation, detention, and court data are now available statewide to criminal justice professionals.

1Records on individuals obtained directly from MNCIS are not linked by fingerprint. 2These misdemeanors, referred to as "targeted misdemeanors," include: assault in the fifth degree, domestic assault, harassment (violation of restraining order), interference of privac y (stalking), indecent exposure, orders for protection violations, and driving while impaired (DWI) violations. Compiling conviction information for these cases is still a challenge because the courts currently cannot electronically transmit all targeted misdemeanor case dispositions to the BCA. MNCIS is expected to rectify this problem.

 

State investments in underlying technology have supported access to information systems and repositories, as shown in Table 2.2. The CriMNet Integration Backbone will provide a way for users to electronically search the data systems and repositories they are authorized to access. 3 Scheduled for full statewide implementation in late 2004, the Integration Backbone will allow users to search five statewide systems: Statewide Supervision System, Court Web Access, Predatory Offender Registration System, Minnesota Repository of Arrest Photos, and Correctional Operations Management System (a Department of Corrections information system containing prison data). 4

STATEWIDE PROGRESS 29

Table 2.2: CriMNet Technology Projects That Facilitate Information Sharing Project Purpose and Benefit Integration Backbone As envisioned, this system will provide a technical infrastructure to connect criminal justice information systems and organize and distribute data among jurisdictions.

Under its initial statewide implementation, expected in late 2004, the Integration Backbone will allow criminal justice professionals across jurisdictions to search five statewide systems through a single query, and it will provide links to additional data in source systems. a Previously, these systems had to be searched separately. Criminal Justice Data Network (CJDN) Upgrade This project, completed in 2001, upgraded the Bureau of Criminal Apprehension's network and the computer workstations used to access it. The upgrade was necessary to support statewide integration of law enforcement data systems and repositories, such as arrest photos and fingerprint data.

Over 15,000 authorized users representing about 660 agencies statewide use this network to transfer data to the Bureau of Criminal Apprehension and to access a variety of state and federal criminal justice information systems. Sentencing Guidelines Worksheet Completed in 2001, this system provides an electronic means to document and transfer sentencing information to appropriate agencies. Used by court, prosecution, defense, and corrections personnel, it reduces time for producing and sharing sentencing information.

a These systems are: Predatory Offender Registration System, Minnesota Repository of Arrest Photos, Court Web Access, Statewide Supervision System, and the Correctional Operations Management System (a Department of Corrections prison information system).

SOURCE: Office of the Legislative Auditor analysis of Bureau of Criminal Apprehension, Sentencing Guidelines Commission, and CriMNet Office project descriptions.

Authorized users can electronically access key criminal justice information through a secure network.

3This system has been tested, but not implemented statewide. Our discussion of the Integration Backbone search capability presumes successful statewide implementation and that CriMNet officials'assertion that other systems can be connected to the Backbone is correct. We did not independently verify the technical capabilities of the Integration Backbone.

 

Collectively, as illustrated in Figure 2.1, state systems and repositories provide a centralized network for obtaining certain data to help identify offenders, their arrests and convictions, and their incarceration or probation status. Cities and counties still maintain their own law enforcement information systems for investigating crimes and managing jails. Similarly, prosecutors maintain their own case management systems for investigating and charging offenders. Some local governments, such as St. Louis County and surrounding counties, have initiated inter-county efforts to coordinate these information systems. Others, such as Hennepin, Dakota, and Anoka counties, are actively pursuing systems integration in their respective counties.

Improving Criminal History Records Criminal justice decisionmaking requires information on individuals that is not only complete, but accurate. State action to improve the accuracy of criminal history records involves (1) identifying individuals using fingerprints and (2) using fingerprints to link records from different information systems. As shown in Table 2.3, CriMNet projects have resulted in improved capabilities in both of these areas.

Identifying Individuals The key to creating accurate criminal histories is to ensure positive identification of individuals using information that is unique to an individual. Information about an offender's name, race, and date of birth provides only a partial, and sometimes inaccurate, means of identification. Photographs can also help identify individuals. Currently, the state uses individuals'fingerprints as the basis for creating accurate criminal histories. We found that:

The state has improved its capabilities for compiling and storing accurate fingerprint and other identifying information.

In the past, law enforcement officials around the state used a manual process for fingerprinting and transmitting fingerprint cards to the BCA. Fingerprint cards were occasionally lost or contained insufficient information for processing. Under this process, using offenders' fingerprints to build criminal histories was extremely time-consuming and required extensive staff resources. In many instances, the state was unable to link arrest and court disposition records to individuals, and these records were excluded from criminal histories.

30 CRIMNET Using fingerprints is a more reliable means of linking criminal history records than name or date of birth.

4One local prosecution case management system, Carver County's Minnesota County Attorney Practice System, is connected to the Integration Backbone. Data from other counties with similar prosecution systems may be available in the future, although there are limitations to proceeding with this integration, as we discuss later in this chapter.

STATEWIDE PROGRESS 31 State and Local Users Criminal Justice Data Network Internet-Based Application

Minnesota Repository of Arrest Photos

Federal Law Enforcement Databases

Authorized users can sign on to each of these systems separately through the Bureau of Criminal Apprehension's secure criminal justice network. For example, users can obtain court case information by signing directly on to Court Web Access. If users choose to use the Integration Backbone, they can search the connected systems at the same time.

Predatory Offender Registration

Statewide Supervision System

Integration Backbone Court Web Access

Computerized Criminal History

Incomplete Records in the Suspense File

"Hot Files"

Prison Data a

Figure 2.1: Statewide Access to Criminal Justice Information Systems, as of January 2004

NOTE: The Integration Backbone is scheduled to be fully available statewide in late 2004. a The Department of Corrections' information system for prison data is called Correctional Operations Management System.

SOURCE: Office of the Legislative Auditor analysis of CriMNet program documents.

CRIMNET Table 2.3: CriMNet Projects Contributing to Improved Accuracy of Criminal History Records Project Purpose and Benefit Bureau of Criminal Apprehension

Livescan and Cardhandler Fingerprint Technology Livescan equipment electronically captures fingerprint images and arrest identification data. Livescan machines are located at all county sheriff departments where most bookings occur, but coverage is less complete at courts and local police departments.

Cardhandler is a computer interface that allows Livescan users to transmit electronic fingerprint data to the Bureau of Criminal Apprehension. The Cardhandler project was essentially completed in 2000, although a joint project with Hennepin County to enhance the system was underway in January 2004. Electronic fingerprinting replaces the process of inking and rolling prints, typing arrest and identification data on fingerprint cards, and sending documents to the Bureau of Criminal Apprehension. The electronic process takes up to two hours compared with several weeks for manual cards.

Criminal History Suspense File Program This ongoing program is comprised of subprojects aimed at improving the accuracy of Computerized Criminal History records. It includes technology and work practice improvements to ensure the Bureau of Criminal Apprehension receives fingerprint and identification data and that related court dispositions are linked to the appropriate individual's criminal history.

The suspense program had (1) resolved nearly 420,000 records, reducing the suspense file to about 100,000 records in 2003; and (2) reduced the average monthly percentage of court dispositions entering the suspense file from 50 percent in fiscal year 2001 to 20 percent for July-December 2003.

State Court Administrator's Office Statute Table This ongoing project is to create a single point of reference for statute citations. The statute table is expected to be integrated with other criminal justice systems, resulting in improved accuracy of charge and conviction data included in criminal history records.

As of January 2004, court staff were developing the statute table for integration into the Minnesota Court Information System.

SOURCE: Office of the Legislative Auditor analysis of project documents from the Bureau of Criminal Apprehension and State Court Administrator's Office.

CriMNet investments have improved the content and accuracy of criminal history records.

To facilitate processing offenders' fingerprints, the state now uses electronic, automated fingerprint machines, called Livescan, to create digital images. These images are electronically transmitted to the BCA where BCA staff either create a new criminal record or add the information to an existing criminal history for the individual. As of January 2004, about 78 percent of fingerprint records were transmitted to the BCA electronically. Currently, Livescan machines are located at every sheriff 's office in the state and at many larger police departments, mainly in the Twin Cities metropolitan area. Some judicial districts keep Livescans in certain courtrooms for law enforcement officers to fingerprint offenders that have bypassed the booking process before making their first appearance in court. Only a few jurisdictions have not yet integrated these machines into their daily practices and still rely on manual cards for recording and transmitting fingerprint information to the BCA. Compared with the previous approaches, Livescan allows faster and more accurate transfer of an individual's fingerprint information to the BCA.

Linking Records The BCA's Computerized Criminal History system currently provides the most biometrically accurate information available about offenders' criminal convictions in Minnesota. Using fingerprints and demographic data obtained at the time of arrest, BCA staff match arrest incidents, charging information, and court case disposition records to create this criminal history for offenders. But, as we discuss later, significant work remains in linking criminal records using fingerprints to improve confidence that criminal histories are complete from an individual's first contact with the state criminal justice system. Additional improvements to the BCA Computerized Criminal History system are underway. These initiatives focus on, among other objectives, developing standards and functions that will improve the electronic linking of criminal information across previous and recently implemented systems.

Previous shortcomings in technology and local jurisdictions' fingerprinting practices resulted in the creation of the BCA suspense file, which contains arrest and conviction records that cannot be linked to offenders' criminal histories. By the late-1990s, the suspense file had grown to nearly 450,000 records, and nearly 50 percent of new court case dispositions were going into suspense status. In response, criminal justice officials began devising a strategy to reduce both the size of the existing suspense file and the flow of records going into suspense. While Livescan machines greatly improved collecting and transmitting fingerprints to the BCA, additional resources were necessary to reduce the flow of records into the suspense file and resolve the inventory of unmatched records. In 2001, the Legislature appropriated funds for the BCA to implement a program to do this, and set specific targets to measure the progress. 5 We found that:

The state has made progress in correctly linking offenders with missing conviction records and reducing the flow of records into the Bureau of Criminal Apprehension's suspense file.

STATEWIDE PROGRESS 33 As of January 2004, about 78 percent of f ingerprint records were transmitted to the BCA electronically .

Actions to address shortcomings in technology and f ingerprinting practices have reduced the percentage of new criminal history records that cannot be matched.

5Laws of Minnesota (1Sp2001), ch. 8, art. 4, sec. 10, subd. 3. The goals included reducing the number of dispositions entering the suspense file on June 1, 2001 from 50 percent to 30 percent in the first year (June 30, 2002), to 20 percent the second year (June 30, 2003), and 10 percent in future years. For more information on this program, see Department of Public Safety, Livescan, AFIS, Suspense, MRAP , and ICHS Project Overviews (St. Paul, September 2003).

Beginning in mid-2001, project teams implemented a series of strategies that included (1) inspecting and resolving records manually; (2) identifying and flagging records that could not be readily resolved; (3) programming system enhancements to the criminal history and court repositories to link records more effectively; (4) improving agencies'practices for obtaining and transmitting identification data by holding workshops and distributing suspense file reports; and (5) implementing a criminal history audit program to identify errors. The BCA is also developing a system modification that will allow agencies to correct errors in data they submitted.

Through these efforts and ongoing communication with local law enforcement agencies, BCA staff have resolved nearly 420,000 unmatched records, reducing the suspense file to approximately 100,000 records. The flow of records going into suspense has also been reduced. In the first half of fiscal year 2004 (July to December 2003), on average, about 20 percent of case dispositions from the courts went into suspense— the goal for the full fiscal year is 10 percent. However, for the last three months of 2003, the average percentage of case dispositions going into the suspense file was about 13 percent. The volume of fingerprint and court disposition records coming into the BCA has also grown in recent years, which has increased the volume of work necessary to meet the goals.

Other efforts are underway to improve the accuracy of conviction records. The State Court Administrator's Office is overseeing the development of a statewide, on-line reference system of Minnesota Statutes that will be used with MNCIS. The goal of the electronic system, called the "statute table," is to ensure that prosecution and court staff consistently and accurately enter the correct statute citations when filing charges and updating court files. Given the number and complexity of criminal statutes, the statute table should reduce time spent to manually research information on offenses.

REMAINING INTEGRATION GAPS So far, CriMNet has improved statewide access to information about individuals involved in the criminal justice system. However, we found that:

Statewide integration of criminal justice information is not yet seamless. Gaps in information remain at both the local and state level, and the state has a long way to go in building accurate, cross-jurisdictional criminal histories.

In particular, the current information structure (1) is not yet easily accessible by some criminal justice personnel, such as front-line law enforcement officers, and (2) does not yet provide complete information about individuals'interactions with the criminal justice system, including their initial contact with law enforcement officers, pending criminal investigations, and misdemeanor convictions. While recent CriMNet improvements will facilitate researching and compiling offenders' records, significant work remains in linking these criminal records through fingerprints to ensure that criminal histories are accurate and complete.

34 CRIMNET Since mid-2001, BCA staff have resolved nearly 420,000 unmatched criminal history records.

Access to Information As we describe in Chapter 1, information about offenders comes from many sources and is channeled through a variety of systems and central repositories to be shared by criminal justice professionals. In reviewing the scope of information available and ease of accessing these systems, we found that:

Local law enforcement and prosecution information systems have not yet been sufficiently integrated into CriMNet.

These gaps occur because cities and counties usually maintain their own law enforcement, jail management, and prosecution systems, which are generally not integrated with systems outside of their respective jurisdictions. While law enforcement officers, prosecutors, and public defenders can access the information that is available through the BCA's secure network (and through the Integration Backbone, once implemented), comprehensive information originating from their own work is generally not available to others.

Various efforts have been made in the past to bridge these gaps. The Minnesota Chiefs of Police Association provided more comprehensive police contact and investigative information from local law enforcement systems through its Multiple Jurisdiction Network Organization (MJNO) system. Developed in the mid-1990s and funded through user fees paid by participating agencies, the system provided a way for law enforcement officers to share information used for investigations, background checks, warrant research, and person location. Seen as an opportunity to expand CriMNet's data resources, the CriMNet Office in March 2003 contracted with the Chiefs of Police to integrate MJNO with the Integration Backbone. However, numerous concerns about the system's compliance with the Minnesota Data Practices Act led policymakers to question MJNO as a viable solution to law enforcement agencies' needs. Ultimately, in December 2003, the Policy Group voted to terminate the MJNO contract if the Chiefs of Police Association did not do so first. Later that month, the association chose to shut down MJNO altogether— leaving the state to seek other avenues for integrating these data.

To resolve these deficiencies, the state has several possible paths, including: (1) developing statewide systems or data repositories, similar to MNCIS and the Statewide Supervision System; or (2) using the Integration Backbone to directly access local data. Currently, CriMNet policymakers have not decided which path to pursue, but CriMNet Office officials acknowledge that many issues need to be addressed before the Integration Backbone can be used as the solution. According to stakeholders, significant obstacles exist to proceeding under a "local-to-backbone" approach on a statewide level at this time. First, CriMNet officials have not decided (1) what local data to access, (2) which data to maintain in local hubs or the Integration Backbone, or (3) how the data will be managed. The Integration Backbone is currently programmed to directly access, or "hit," connected systems. There are no hubs yet to absorb, control, and manage users' inquiries of the systems, as was planned in the Integration Backbone's original design. Because some local systems would be overwhelmed by statewide user searches, the CriMNet Office must control users'access to the locally-held data by limiting access to certain hours or select groups of professionals. 6 The effort necessary to manage users' sharing of information without using local hubs or enhanced Integration Backbone capabilities may not be cost-effective on a statewide basis and is not the CriMNet Office's preferred approach. According to CriMNet Office staff, resolving these data management decisions and proceeding with building local hubs are top priorities for the office. Before proceeding with additional local systems integration, however, local units of government must also make decisions regarding their system requirements and data practice policies.

STATEWIDE PROGRESS 35 Local prosecution and law enforcement investigation data are not accessible statewide.

In the meantime, we found that: The state does not have a well-formed plan for bringing local law enforcement, prosecution, and public defender data into CriMNet.

Moreover, establishing and implementing a plan to do so may prove to be difficult, particularly since the state has not mandated local participation in CriMNet. The state's current approach is to make statewide data and systems available and let cities and counties decide whether to integrate their data and systems into CriMNet.

But, according to anecdotal information from CriMNet Office officials, local governments' ef forts towards upgrading and integrating their criminal justice information systems vary greatly around the state, with minimal progress in most jurisdictions. To move forward, the CriMNet Office will need more than anecdotal information. Yet, we found that:

The CriMNet Office does not have adequate information on local jurisdictions' criminal justice information systems; therefore, it does not know how great an investment is needed to integrate more local jurisdictions.

For the most part, the four counties receiving grants are the only counties for which the state has good planning and systems information. Although the CriMNet Office has general knowledge about the types of local law enforcement and prosecution systems used in Minnesota, the office has not assessed the capacity of local governments to integrate their criminal justice information systems as envisioned in the CriMNet model. According to CriMNet Office staff, city and county interest in integrating varies around the state, and their systems and technology are evolving. Under these circumstances, neither the state nor local jurisdictions have a clear sense of future milestones and costs. Cities and counties maintain their own law enforcement, jail management, and prosecution systems, and their systems reflect their users'needs, daily practices, and preferences. As a result, local approaches and efforts towards integration will follow different paths. Rather than planning and designing uniform solutions at this point, CriMNet officials told us they will likely follow an individualized approach to bring local systems into CriMNet. However, until the Office obtains more comprehensive information about city and county systems, further development of the Integration Backbone and integration of local systems will not proceed, primarily because state and local officials do not have clear understanding of the technical specifications necessary for integration. 7 More importantly, until CriMNet officials resolve concerns about how the state will manage and classify data moving from local systems into a statewide, integrated network, some local jurisdictions may be unwilling to share information about individuals'criminal activities. 8

36CRIMNET The state and local governments have many issues to resolve before they can proceed with local systems integration.

6For example, access via the Backbone to the one local prosecution case management system integrated with it— Carver County's Minnesota County Attorney Practice System— is limited to Carver County prosecutors only.

 

Other Integration Backbone functions have not been developed as quickly as anticipated, primarily due to lack of progress resolving data management issues and concerns about data practices and security. These other functions are primarily to facilitate workflow and create work eff iciencies. However, criminal justice personnel must first identify and agree on work requirements (what the system should be able to do) before programming technical functions, and reaching these agreements has taken longer than anticipated.

Criminal Histories Electronic fingerprinting is necessary, but not suff icient, to meet CriMNet goals to accurately identify individuals and obtain their full criminal histories. We found that:

Significant work needs to be done before criminal histories accurately reflect offenders' involvement in the criminal justice system from the start. The state has several technological means for pulling together information about an individual's involvement with the criminal justice system, each of which has some deficiencies. As described previously, the BCA creates criminal histories for offenders using arrest and conviction data, but these histories only include biometrically-linked records for felony, gross misdemeanor, and certain misdemeanor offenses. Other systems— MNCIS and the Integration Backbone— will eventually provide the means for a more extensive compilation of convictions. But, these systems do not guarantee that records will be accurately linked to individuals' criminal histories.

Although the Backbone can pull together extensive data and records from a number of repositories, the Backbone does not yet provide biometric, or f ingerprint-linked, offender-based information. That is, requests for "John Doe" will provide records for all individuals using the name John Doe. Users must define parameters, such as date of birth, to narrow their search. Still, because of incomplete data and offenders' use of aliases, search results do not guarantee a certain match for all records. Users must consider the Integration Backbone information as a resource to facilitate their research efforts— and use due diligence to confirm that the records received from a search are, in fact, for the same individual. Although the Integration Backbone search function does provide limited information about whether biometric information exists about an individual in the BCA's Computerized Criminal History system, future plans for the backbone system include redesigning and improving its and other systems' capabilities for electronically linking records.

STATEWIDE PROGRESS 37 Technical development of the Integration Backbone is on hold until CriMNet stakeholders resolve work requirements and data management issues.

7As we discuss in Chapter 3, better information about local jurisdictions'systems will help CriMNet estimate the time needed for integration. The Department of Corrections discovered on the Statewide Supervision System project that integrating local jurisdictions with a statewide system was far more time-consuming and complex than anticipated. 8As of January 2004, the CriMNet Office was in the process of hiring additional staff to help expedite work on integration standards, user needs, and data practices.

In the future, CriMNet officials intend to use the Integration Backbone to move basic data from one system to another to improve eff iciency , but also to reduce errors associated with redundant data entry. Ideally, the system would automatically tag and link records for an individual upon initial arrest and f ingerprinting. By incorporating these "workflow" features that use a fingerprint as the record identifier across systems, the accuracy of criminal histories would improve. An immense amount of work is involved to achieve this level of technology, and doing so requires criminal justice professionals across functions and jurisdictions to modify work practices and ensure accurate decisionmaking.

Still other challenges remain in compiling accurate criminal histories. As we reported earlier, on average, about 20 percent of court case dispositions went into the suspense file during the first half of fiscal year 2004. Currently, nearly 90,000 records cannot be matched to fingerprints and could remain unresolved, primarily because the state cannot retroactively capture fingerprints after sentences have expired since agencies no longer have authority to do so. Because it is a time-consuming process, some agencies have been reluctant to track down individuals on probation to capture their fingerprints. Agencies'f ailure to obtain fingerprints continues to be the biggest obstacle to further reducing the flow of records into the suspense file. This occurs because (1) the individual was summoned to court and bypassed the booking process, (2) the agency f ingerprinted the individual but did not send in the fingerprint card, or (3) the agency did not fingerprint the individual. Other reasons are unique to specific local systems, which may not be able to link with the state's system. According to BCA officials, standardizing fingerprint work practices across the state remains a challenge.

To close information gaps, some jurisdictions must upgrade their jail management systems to better interface with the BCA's Livescan and Cardhandler system. The BCA is currently partnering with Hennepin County to improve the capture of f ingerprint and offender information from all law enforcement agencies within that county. A significant share of statewide bookings occurs through the Hennepin County jail— a detention facility used by most law enforcement agencies in that county. As such, this targeted effort, once completed, should have a measurable impact on reducing the flow of records into suspense. The BCA recently initiated another project, called "electronic criminal linking," to develop work practice and technology changes that will support electronic linking of fingerprints, arrest photographs, and other data at the point that the information is first collected. In addition, the BCA and state court representatives are discussing ways to ensure that individuals making a first appearance before the court have fingerprints on file with the BCA and, if not, that the individual is f ingerprinted upon appearance at the court.

38CRIMNET A significant amount of work will be required to electronically link records from an offender's f irst interaction with the criminal justice system to all subsequent records.

CONCLUSIONS The state has made significant progress over the last decade toward eliminating fragmented data systems and sharing critical information about offenders' criminal activities around the state. But, gaps and significant challenges remain in integrating data for certain criminal justice functions and linking offender records into complete criminal histories. Public defenders, prosecutors, and local law enforcement officers, for example, do not have statewide information systems. Certain types of data from these professions may be available statewide through other systems (for example, law enforcement agencies must submit certain data to the BCA). However, absent additional statewide systems, fuller integration of these data will depend on the ability and willingness of cities and counties to link their information systems with the state. At this time, the state does not have a well-formed plan to integrate more local jurisdictions into CriMNet.

Without a state mandate on local governments and much more state funding, attaining and maintaining 100 percent integration of all jurisdictions and systems is an unrealistic, if not impossible, goal to achieve in the foreseeable future. Similarly, it may be difficult for the CriMNet Office to estimate a "final completion date" for CriMNet. On the other hand, policymakers should have information about the status of criminal justice system integration and how it benefits the criminal justice community in order to progress toward achievable, realistic goals. We discuss the need for better short-term and long-range planning for these incremental steps in Chapter 4.

RECOMMENDATION RECOMMENDATION The CriMNet Office, in coordination with local governments, should inventory local governments' criminal justice information systems and integration plans and estimate the resources that will be required to bring needed information into CriMNet.

At one time, the CriMNet Office had prepared a draft survey to collect information about local criminal justice information systems and integration capabilities, but the office never sent the survey to local governments. It may be a logical starting point to continue this effort to better understand local needs. In addition, other decisions regarding how data will be indexed in the Integration Backbone and in local hubs need to be made before CriMNet Off icials can fully assess local jurisdictions'readiness to integrate.

STATEWIDE PROGRESS 39 The state does not have a well-formed plan to integrate more local jurisdictions into CriMNet.

Agency Integration Projects SUMMARY Agencies' project management practices varied, and CriMNet projects typically took longer and cost more than anticipated. On most projects, but the Integration Backbone in particular, stakeholders underestimated the complexity of criminal justice work requirements, which increased the time needed to complete project tasks. Other obstacles to completing projects as planned included: (1) inability to adequately resolve work practice and legal requirements prior to proceeding with technical development; (2) lack of state staff, expertise, or funding; (3) lack of clear expectations and precise contract language regarding project deliverables; (4) challenges coordinating tasks and activities among various agencies and systems; and (5) changes to project design or scope. On the Integration Backbone project, original schedule and cost estimates were fundamentally flawed, and the cost of some vendor work products was too high given the products' value to the state. However, elements of the Integration Backbone contract were renegotiated in 2003, which will likely result in a better value for the state. In general, state agencies could more actively collaborate and share information about their experiences integrating state systems.

In previous chapters, we described the progress of criminal justice information integration and estimated costs for the state. Collectively, these projects have moved Minnesota forward in providing criminal justice professionals with access to timely information about individuals involved in the criminal justice system. In this chapter, we take a closer look at how state agencies managed CriMNet projects, focusing on how their experiences may help future projects proceed more smoothly.

This chapter addresses the following questions: To what extent have state agencies' CriMNet projects met expectations regarding schedule, cost, and scope?

What are the lessons learned from agencies' experiences managing these projects? To answer these questions, we reviewed several large and small dollar value projects from among active and recently completed CriMNet projects. We relied in part on Department of Administration guidelines for professional/ technical contracting and project management guidelines developed by the Office of Technology. To a great extent, we used vendor contracts, budget documents, work plans, and interviews with project managers to help us assess the overall progress and management of projects.

We included the following projects in our review: the Integration Backbone, the Statewide Supervision System, the Minnesota Court Information System (MNCIS), the State Sentencing Guidelines Worksheet, the Minnesota Repository of Arrest Photos, the Predatory Offender Registration System, the Suspense File program, and the Cardhandler/ Livescan project. Due to time constraints, we evaluated some projects, particularly the Integration Backbone, more comprehensively than others. We did not evaluate the technical performance of these systems nor did we evaluate local jurisdictions'management of city and county integration projects funded by CriMNet grants.

The chapter is divided into two sections. In the first, we discuss the extent to which agencies' CriMNet projects, on the whole, proceeded according to planned timeframes, budgets, and scope. Because of legislators' specific interest in the Integration Backbone project, we discuss it separately. In the second section, we discuss common obstacles and lessons learned from agencies'ef forts to meet these challenges.

MANAGING PROJECT SCHEDULES, COST, AND SCOPE

Standard project management practices encourage project managers to estimate timeframes and costs for a new project, or at least the phases of the project being planned. For our evaluation, we had expected to use these baseline estimates, along with documented statements of the project's scope and products, to assess whether projects proceeded as planned. However, we found that:

For some CriMNet projects, planning estimates regarding timeframes, cost, and scope were poorly documented, making it difficult to compare actual outcomes to baseline estimates. Agencies' initial projections of long-term costs and milestones for some projects were poorly documented. In some cases, this happened because projects proceeded in phases, with milestones set only for the current phase, not the entire project. This made estimating long-term costs difficult. In some instances, agency officials estimated CriMNet project costs and modified proposals based on guesswork "to get projects going." 1 Also, some project work plans and vendor contracts did not clearly articulate expected vendor products, and few contracts included vendor performance standards. As a result, when we identified project delays or additional vendor costs, we could not always trace them directly to changes in scope or products.

42 CRIMNET CriMNet project work plans and contracts did not always articulate clear performance expectations.

1For example, see Criminal and Juvenile Justice Information Policy Group Meeting Minutes, November 9, 2000.

Although we could not obtain clear project plans for all of the projects we reviewed, we used various contract documents and interviews with project managers to understand and assess how each project proceeded. We concluded that:

Although many projects followed sound management practices, the projects we reviewed typically took longer and cost more than anticipated. For example, we found several instances in which agencies used good project management tools and practices for identifying objectives, assessing resources, identifying project risks, and establishing realistic work plans. A number of projects managed by the Bureau of Criminal Apprehension (BCA), such as the Suspense File, Predatory Offender Registration System, and Criminal Justice Data Network projects, had well documented project plans. On the whole, BCA project managers also filed status reports, followed state contracting guidelines, and made up-front cost estimates. The BCA also uses a project management software package to help plan and track progress and guide project work.

However, agencies' approaches to project planning and implementation varied, and project outcomes did not always conform to initial time and cost projections, as shown in Table 3.1. A variety of factors influenced how projects proceeded.

For example, beginning in 2000, the State Court Administrator's Office spent two years planning for the replacement of its court information system, including assessing user needs and researching market products. Court administrators then solicited and evaluated vendor proposals. The courts contracted with the one vendor that met initial evaluation criteria, Sustain Technologies, for further testing of its product. 2 But, Sustain was not able to comply with contractual performance requirements for a statewide system. Court administrators eliminated Sustain from further consideration in May 2001 and contracted with Deloitte & Touche to complete system requirement and design work.

In the meantime, court administrators continued to monitor the software product market and discovered a new product by Tyler Technologies. When tested, this product met all contractually specified requirements. The courts terminated its contract with Deloitte & Touche and, in July 2002, the State Court Administrator's Office entered into a $10.9 million contract with Tyler to proceed with developing MNCIS. 3 As of February 2004, MNCIS has been implemented in three counties and, according to court staff, will be installed in all Minnesota counties within the next 28 months. A precise comparison of the project's final development time and costs to original estimates will have to wait until its full implementation in 2006.

AGENCY INTEGRATION PROJECTS 43 Some CriMNet contracts have been terminated because of performance problems.

2According to court administrators, when the request for information was issued to the national software vending community in June 2000, there were few ready-made products on the market meeting the courts'needs. Because the courts had a critical need to move forward with MNCIS, court administrators selected what they considered to be the best vendors for the project at that time. Later, Tyler Technologies, one of the responding vendors to the initial request for information, used Minnesota's requirements to develop its new Trial Court System product.

3The State Court Administrator's Office did not change the terms of or reissue a request for proposals following the release of either Sustain Technologies or Deloitte & Touche, the original vendors on MNCIS. Court administrators advised us that their previous research and vendor selection process led them to believe that Tyler Technologies, Inc. was the only vendor offering a product that would meet their project needs, and elected to contract directly with that vendor. We note that the judicial branch is not required to follow the same contracting processes required for executive branch agencies under Minnesota law.

 

Because of the innovative nature of the Statewide Supervision System, the Department of Corrections took a different approach to planning and implementing the project. The department did not start with solid time and cost estimates for completing the entire project. Rather, the department decided to develop the Statewide Supervision System in phases, estimating timeframes and costs for each phase as the project proceeded. Each phase was defined by an incremental goal, and the department assessed work results and modified the project design as needed before proceeding to the next phase.

44 CRIMNET Table 3.1: Extent to Which CriMNet Project Timeframes Changed Project Project Schedule Outcomes Minnesota Court Information System Following approximately two years of planning and testing the product initially selected, court administrators found a new

vendor with a product that better meets the courts' system requirements. In spite of the product switch, court administrators said the project is on track with the six-year planning, development, and implementation timeline presented to the Legislature in 2001. Statewide Supervision System The project was planned and implemented in phases, without an estimated end date for integrating local probation and detention information systems. The core technology and integration with some local systems was completed in 2001. Integration of remaining county systems took longer than expected, occurring over the following two years.

Integration Backbone Initial estimates ranged from six to nine months. Actual project duration has approached two years, with intermittent breaks in the vendor's work schedule. Suspense File Although implemented as an ongoing program with no final end date, several components have stalled, primarily due to lack of state staffing. Cardhandler Phase II of this project is not completed, but state timelines were extended to match slower implementation schedules in a key local jurisdiction. Predatory Offender Registration System Project exceeded plan estimates by five months, due primarily to lack of state staffing and the need to integrate with national sex offender files. Integration was also delayed waiting for a related system's upgrade to be completed.

Sentencing Guidelines Worksheet Project funding was available for the 2000-01 biennium, and the project was substantially completed within this timeframe. Minnesota Repository of Arrest Photos Project completed on time. Capability for more agencies to submit photos will be added as funding becomes available.

NOTE: Because many projects did not have clear baseline project schedules, we could not precisely measure the extent of project delays. SOURCE: Office of the Legislative Auditor analysis of project contracts, amendments, and planning documents from the State Court Administrator's Office, Sentencing Guidelines Commission, and departments of Public Safety and Corrections.

Many CriMNet projects took longer than anticipated.

Integrating local probation and detention information into the Statewide Supervision System took longer than expected. According to Department of Corrections officials, the system's core components and integration of some county information was completed in June 2001, with integration of the remaining county probation and detention systems occurring over the following two years. 4 Although the department did not have an estimated end date for connecting local information systems, project documents indicate that local integration went more slowly than anticipated. In part, longer timeframes reflect the level of effort needed to ensure participation by all counties. According to project managers, the department's goal was to integrate all county probation and detention information systems. But, without the authority to mandate participation, the department was not sure that all counties would agree to be involved. But in the end, all 87 counties decided to join the statewide system. As we discuss later in the chapter, integrating local information systems presented unique challenges, with much of the scheduling in the hands of local governments and their vendors.

Projects also typically cost more than expected, although we could not precisely quantify the extent of cost increases because many projects did not have documented baseline cost estimates. Some cost increases occurred because project administrators underestimated the time and work needed to complete tasks and later amended project contracts to extend vendors' services. These extensions resulted in increased payments to vendors. Other circumstances related to the legislative process also influenced the accuracy of initial cost estimates. For example, the BCA prepared a cost estimate for the Predatory Offender Registration project in response to a request from the 2000 Legislature, which initiated action on the project. 5 According to the BCA, bureau staff had two days to estimate project costs, and the resulting estimates turned out to be too low. Spending on two other BCA projects (Minnesota Repository of Arrest Photos and Livescan) exceeded appropriated amounts. For these projects, the BCA had not submitted project proposals to the Legislature; rather, the Legislature initiated appropriations for these projects and required that they be completed within one year. According to BCA officials, the bureau could have stayed within the funding allocated by the Legislature, but the BCA had other funding available for criminal history improvements and chose to use it. 6

 

AGENCY INTEGRATION PROJECTS 45 Costs for some projects increased because agencies underestimated the time and work needed to complete project tasks.

4According to Department of Corrections officials, the Statewide Supervision System was completed in June 2001. At that point, the probation and detention components had been designed, prototyped and tested; the department's prison database was connected; and three adult detention f acilities and over half of the county adult probation systems were integrated. From June 2001 through June 2003, the department contracted with a vendor (the vendor that did the system development work) to integrate the remaining counties'probation and detention systems into the Statewide Supervision System. However, department officials do not consider integrating the remaining county systems to be part of the Statewide Supervision System project; rather, they consider the additional work to be "system support." We think a 2003 completion date more accurately reflects the time needed to achieve the Statewide Supervision System's intended outcomes. According to project documents and project managers'statements to us, the department's goal for the Statewide Supervision System was to integrate probation and detention information from all counties. This goal was substantially achieved in 2003. 5Laws of Minnesota (2000), ch. 311, art. 1, sec. 3, subd. 2. 6Laws of Minnesota (2000), ch. 311, art. 1, sec. 3, subd. 1. In total, the 2000 Legislature appropriated about $5.2 million for the arrest photo and fingerprint projects, plus some additional funding for implementation costs. The final cost for the two projects was approximately $8 million.

Finally, we also assessed whether the scope of projects changed as they progressed and whether these changes affected overall costs. In this regard, we looked for contract amendments signaling new products to be delivered or changes to the nature of work products already in the contract. Based on our evaluation of project documents and contracts, we found:

The scope of some projects expanded beyond original funding requests— sometimes for system enhancements not directly linked to criminal justice integration. Some agencies expanded vendor work beyond what was specified in the original project descriptions. For example, the Department of Corrections added to the Statewide Supervision System project a new component to enhance the department's systems for local detention facility licensing and reporting. This change increased vendor costs by nearly $60,000. At a cost of $14,000, the Sentencing Guidelines Commission added a component under the automated Sentencing Guidelines Worksheet project to extract data for research purposes. While these functions benefit criminal justice processes, the added components were not specified within the initial scope of the criminal justice integration projects. Changes in project scope and increased costs were not isolated to these projects, however, as we discuss below in our review of the Integration Backbone project. 7

The CriMNet Integration B ackbone Because the Integration Backbone is such a critical component of CriMNet and of particular interest to legislators, we reviewed this project in greater detail. As described earlier, the CriMNet Integration Backbone provides the technical infrastructure, akin to a network, that allows data to be shared among agencies, information systems, and jurisdictions within Minnesota. Linking Minnesota's criminal justice systems through the Integration Backbone is a complex endeavor— one that has not been achieved elsewhere on a statewide scale. This complexity was certainly a factor in planning and managing the project.

Unlike the other projects included in our discussion above, the Integration Backbone project is managed directly by the CriMNet Office, under Policy Group oversight. In 2001, the CriMNet Office initiated the project by issuing a request for proposals (RFP) to solicit bids for building the backbone technology. 8 In February 2002, a committee of agency representatives selected a primary vendor, and following approval by the Policy Group, the CriMNet Office negotiated a contract with the firm— Mobiam Solutions, Inc. 9

46 CRIMNET Some agencies expanded vendor work beyond original project descriptions.

Linking criminal justice systems through the Integration Backbone is a complex task.

7For additional information about agencies'use of professional/technical contract services for CriMNet projects, see Office of the Legislative Auditor, CriMNet Financial Audit( St. Paul, 2004). 8We did not perform a cost/benefit analysis to assess the wisdom of building a backbone system rather than buying an existing product for integrating multiple systems. 9In addition to identifying a vendor that satisfactorily met the CriMNet model requirements, the selection team focused primarily on three (out of many) criteria: (1) developing a non-proprietary product owned entirely by the state, (2) speed to delivery, and (3) the most prudent cost structure workable under state funding. The contract was negotiated by a former CriMNet Executive Director, whose employment contract was later terminated.

Based on our evaluation of how the Integration Backbone project proceeded compared to the proposal and contract terms, we found that:

The Integration Backbone project did not proceed according to original estimates for cost, schedule, and scope; however, these original estimates were fundamentally flawed. In the discussion that follows, we question various aspects of how this project was planned and managed. But, it is important to remember that, according to various CriMNet officials, the state will have a usable, and potentially very powerful, technology to support statewide sharing of criminal justice information. In addition, the state owns the product and will not be obligated for licensing fees. 10

The Backbone development work and contracts were organized into two phases. 11 Phase I involved designing the technology structure and demonstrating that the basic technology would work in a test environment. Phase II essentially involved expanding the system to connect select state systems and to implement a search function for those systems. We found problems with the veracity of the original time and cost estimates and with other terms of the Backbone contracts that contributed to changes in schedule, cost, and system functions.

For example, we found great disparities between the vendor's cost proposals and the Phase I and II contracts. As shown in Table 3.2, the vendor's original cost proposal totaled $2,028,000—$ 200,000 for Phase I and $1,828,000 for Phase II. However, the Phase I and II contracts negotiated after the vendor was chosen were let for $680,000 and $2,950,000 respectively, or $3,630,000 in total.

AGENCY INTEGRATION PROJECTS 47

Table 3.2: Estimated and Actual Integration Backbone Costs Vendor Original (2002) Actual Project Phase Proposed Costs Contract Costs Expenditures Phase I $ 200,000 $ 680,000 $ 680,000 Phase II 1,828,000 2,950,000 2,810,000 a Total $2,028,000 $3,630,000 $3,490,000 a Includes expenditures reported through October 30, 2003. The Phase II contract was subsequently renegotiated and was signed in October 2003.

SOURCE: Office of the Legislative Auditor analysis of Department of Finance data and Integration Backbone project documents.

The original cost and time estimates fo r the Backbone contracts were highly questionable.

10Similarly, the state owns the Statewide Supervision System and is not obligated for vendor licensing fees. For MNCIS, the state is obligated for licensing fees for the vendor's original software but owns modifications to the software that the state paid for. 11The RFP for the Integration Backbone project divided work into several phases. Interested vendors were asked to submit proposals and cost estimates for the first two phases. Following vendor selection, the state contracted with Mobiam Solutions, Inc. for Phase I work, which included requiring the vendor to submit a preliminary plan for Phase II work. The state then had the option to contract with Mobiam for Phase II or seek other vendors. In July 2002, following completion of Phase I, the CriMNet Office, with the approval of the Policy Group, elected to move forward with Mobiam.

We did not find evidence that the CriMNet Office in any way validated the vendor's cost proposal to ensure that it was reasonable. The Mobiam bid was significantly lower than the bid from the other final vendor that CriMNet considered, but the integration approaches from the other vendors were fundamentally different, making comparisons difficult. Under the state's preferred practices, the CriMNet Office should have obtained an independent assessment of project costs and timelines in order to evaluate vendor proposals. The CriMNet Office did not do this, so it did not have a point of comparison in evaluating Mobiam's bid. Although the CriMNet Office had compiled and published extensive documentation regarding the desired Integration Backbone capabilities, it did not use this model to set its own rough benchmarks.

Project documents from 2002 offer several reasons for the cost difference between the initial proposal and contract, but in large part, these explanations do not suff iciently account for the discrepancies. 12 According to CriMNet Office documents, the Phase I cost increases were due to deliverables (vendor work products) shifting from Phase II back to Phase I. The Phase I contract did include deliverables shifted from Phase II. However, the original Phase II contract costs did not proportionately decrease as a result of shifting deliverables to Phase I. Instead, Phase II costs increased by $1.1 million. The current Backbone project manager asserted that the total contract costs increased because the state added work not previously identified in the RFP and because the vendor assumed some work tasks that were the state's responsibility in the RFP. 13 But, when we compared the Phase II work products described in the RFP to the Phase II work products included in the contract, the differences we found were not suff icient to explain the full cost increase. In trying to make this comparison, we found that the RFP did not clearly specify tasks to complete, and the contracts themselves were not clearly written. 14

In explaining the changes in total project cost, the project manager also added that, in retrospect, the state paid too much for certain work products. We agree. Based on our review of the contract terms and descriptions of the deliverables:

The costs of some vendor deliverables were too high given the products' value for the state.

48 CRIMNET The state paid too much fo r some work under the Backbone contract.

12For example, in reporting to the Policy Group, the former Executive Director incorrectly stated that, although the Phase I costs increased, "the overall numbers of the project remain constant." See Criminal and Juvenile Justice Information Policy Group Meeting Minutes, March 22, 2002. 13Under Department of Administration guidelines, the contract and work deliverables should substantially conform to the RFP and vendor proposals. The significant increase (79 percent) in total costs from the initial bid proposal to the contract has caused some stakeholders to question whether the project terms had changed substantially. As we describe later, the CriMNet Office added some deliverables that were not related to the original RFP and could have been completed by another vendor, in which case the state should have issued a new RFP as normal state practice would dictate. 14That is, the technical requirements, deliverables, and performance standards within the contracts were greatly lacking in detail, making it difficult to identify and map actual technical deliverables in the RFP to the contract provisions and estimated costs.

For example, the CriMNet Office added to the contract that the vendor should (1) produce a multi-media presentation on CD-ROM describing the end-state vision for CriMNet in terms of how the Backbone would be used, and (2) provide training describing the CriMNet vision and proposed functions of the Integration Backbone. 15 Together, the CD-ROM and training session deliverables accounted for 17 percent ($ 600,000) of total vendor payments through October 2003. While the CD-ROM may be a useful tool to communicate the integration concept, we question whether its value merits the $400,000 spent to write the script and make the CD-ROM. For an additional $200,000, the vendor provided training sessions over a two-day period that were essentially public relations activities to create buy-in to the Integration Backbone project. The choice to invest in these products is particularly questionable given that other critical work directly supporting Backbone development, such as defining user requirements, still needed to be done.

In addition to cost discrepancies, the Integration Backbone project has not proceeded according to proposed timelines or expectations regarding the range of system functions that were expected to be operable. In general, we found that:

Initial planning documents and vision statements for the Integration Backbone project grossly underestimated the time and complexity of tasks necessary to implement project goals and objectives. The CriMNet Office did not have the benefit of other states'experiences to guide its planning because no other state has implemented an integration model like CriMNet. With such a steep learning curve, we understand that estimating timelines can be difficult.

Early CriMNet Office planning documents suggested a range of timeframes for major project milestones, some of which have been met and others that have not. The vendor for the Integration Backbone project initially estimated that Phase I and Phase II tasks could be completed in about 8 weeks and 19 weeks, respectively. The subsequent contracts specify 10-week and 39-week work plans, with work expected to be done by June 2003. Actual project duration has exceeded these combined estimates, with the vendor working intermittently over a two-year period, beginning in April 2002. 16 The current contract, as amended, is expected to close in March 2004. (These amendments, negotiated in mid-2003, are discussed later.) Regardless of the various estimated timelines, CriMNet Office documents reflect conflicting information about the project's progress. For example, vendor status reports indicate that the Integration Backbone project is on schedule, while CriMNet Office documents state that the project is behind schedule.

Development of the Integration Backbone has been scaled back substantially from early vision statements presented to the Legislature, the Policy Group, and others. 17 Several CriMNet Office documents from 2002 say that the search function plus several other capabilities related to the automated flow of information would be completed and operational by the end of 2004— goals that are far from being achieved. These documents specified three main outcomes that were to be achieved: (1) search capability for eight state systems by December 2003, (2) subscription capability (the ability for a criminal justice professional to be automatically notified if a specified event occurred anywhere in the system) by June 2003, and (3) delivery of workflow capability (automatic transfer of data from one system to another) for eight state systems by December 2004. In our view, these expectations were unrealistic. To date, the Backbone search function for five systems is being tested statewide and is scheduled to be fully implemented in October 2004. The core framework of the subscription function is complete, but will not be implemented for at least a year, and possibly longer. Further development of the subscription function is on hold until data practice and other issues are resolved. The workflow function is being developed as a "proof-of-concept" to demonstrate the feasibility of electronically transferring prosecution complaint information between a county prosecution office and the courts. The vendor completed the Integration Backbone work needed to transmit the data from one system to the other, but the courts have to complete additional work to bring the transmitted complaint data into MNCIS.

AGENCY INTEGRATION PROJECTS 49 The Backbone vendor was paid $600,000 fo r public relations work.

The scope of Integration Backbone capabilities has been scaled back substantially from early vision statements.

15According to the Integration Backbone project manager, although not explicitly included in the contract, the CriMNet Office agreement with the vendor was for delivery of up to 20,000 CDs. To date, the vendor has provided the CriMNet Office with 10,000 CDs. 16The vendor did not work continuously during these time periods. 17For example, see CriMNet Office, " CriMNet Implementation Strategy," presentation to the Policy Group, August 29, 2002, pp. 8-13; and Department of Public Safety contract #A41298,

Statement of Work and contract amendments.

 

By late 2002, the Policy Group was concerned about how the project was being managed and whether, from a technical perspective, the Backbone was going to provide the expected results. At the Policy Group's request, a number of technical professionals from various jurisdictions reviewed the project. In general, the technical review identified many of the issues we discussed in this section. The Integration Backbone project has since been redirected. In our judgment:

Backbone contract changes negotiated in 2003, which altered the timing and nature of remaining work, will likely yield a better value for the state. For example, some of the planned contract deliverables were dropped because the technical specifications supporting these deliverables had not been completed. Some tasks, such as work on the Integration Backbone core functions (the ability of the Backbone itself to store certain information), automated workflow, and the subscription function, were put on hold, to be completed after CriMNet participants make decisions regarding user requirements, data practices, and the types of data that will be stored by the Backbone. Several valuable provisions were also added to the Backbone contract to incorporate better project and system controls, such as (1) discrete tasks and milestones to allow CriMNet officials to make go/ no-go decisions, (2) system security enhancements, and (3) training for state staff to build Backbone components and maintain the Integration Backbone system in the future.

We did not do a technical review of the Backbone functions as part of our evaluation. However, stakeholders in various agencies believe that the Integration Backbone as developed to date is a good technical product. The CriMNet Office developed a "Search Rollout Plan" that includes testing and external validation to

50 CRIMNET The Integration Backbone search function is scheduled fo r statewide release in October 2004.

CriMNet stakeholders believe that the Integration Backbone is a good technical product.

assess the system's performance. The results of this work will help resolve questions and concerns about the work done to date. The rollout plan also details a more realistic work plan and timeline for releasing the search function to users statewide and recognizes risks and limitations in doing so. The statewide release is expected to be completed by October 2004. The decision to test and release the search function rather than pursuing other capabilities reflects the CriMNet Office's priority on providing some integration benefits to criminal justice stakeholders at the earliest possible date.

INTEGRATION OBSTACLES AND LESSONS LEARNED

Overall, the CriMNet projects we reviewed resulted in direct benefits to Minnesota's criminal justice community, but the projects themselves did not always proceed according to the proposed path. Our evaluation showed that agencies encountered similar obstacles and challenges in integrating criminal justice systems. We identified a number of common challenges that help explain discrepancies between early planning estimates and the progress of CriMNet projects. Although not affecting each project to the same degree, these challenges included: (1) inability to adequately resolve issues pertaining to work practices and legal requirements prior to proceeding with technical development; (2) lack of state staff, expertise, or funding to accomplish the established objectives; (3) lack of clear expectations and precise contract language regarding project deliverables; (4) challenges coordinating tasks and activities among various agencies and systems; and (5) opportunities to benefit from new technology or otherwise improve results that required changes to project scope. These factors contributed to the sometimes erratic and occasionally stalled progress of some projects, particularly the Integration Backbone project.

Given the extent to which Minnesota is breaking new ground, criminal justice agencies are their own best resource for developing guidance on how to manage criminal justice integration projects. With this in mind, we found that:

Agencies' experiences managing CriMNet projects demonstrate the importance of making planning decisions up-front, coordinating across jurisdictional lines, and negotiating tighter contracts. The Department of Corrections and State Court Administrator's experiences on two statewide integration projects— Statewide Supervision System and MNCIS— in addition to the CriMNet Office's experience on the Integration Backbone project are of particular benefit in identifying lessons learned.

AGENCY INTEGRATION PROJECTS 51 State agencies experienced similar problems on most criminal justice integration projects.

Planning Several projects ran into difficulties when project team planning did not address longer-term needs for funding and system development. These planning missteps meant the technical development of some systems proceeded even though some prerequisite decisions had not been made. In one instance, system design and development went forward without project stakeholders having made key decisions regarding what they wanted the system to do. On other projects, failure to resolve data practice questions and user needs was a stumbling block.

Several projects were adversely affected by failure to address issues regarding compliance with the Data Practices Act. In March 2003, the CriMNet Office, with the approval of the Policy Group, negotiated a contract with the Minnesota Chiefs of Police Association to integrate, via the Integration Backbone, data from their Multiple Jurisdiction Network Organization— a database containing law enforcement investigation information. Subsequent review revealed that the information system and data structure failed to comply with the Minnesota Data Practices Act. The Policy Group decided to terminate the contract in December 2003 after the state had spent $72,000, not including programming costs for the Backbone. 18 In addition, another expected feature of the Integration Backbone project, called subscription, that will allow users to be notified if a specified event occurs is ready to be used, but it will not be enabled in the near future. According to the CriMNet Executive Director, unresolved data practice questions, which may take several years to resolve, are a primary reason for the delay. And, as we described in Chapter 2, some local jurisdictions advised us they are hesitant to proceed with integrating into the state's Integration Backbone system until the state resolves policy issues pertaining to data practices and technical issues related to system security.

Assessing the needs of system users is another important aspect of project planning. The Integration Backbone project team focused primarily on technical development and did not clearly document prior to starting the project the work practices and data needs that the project was to address. For example, the CriMNet Office and stakeholders have not yet explicitly defined what data will be housed or "indexed" by the Backbone system itself (rather than being held in the local system) or how that data will be managed. Therefore, they do not know precisely what to program or build to connect additional criminal justice systems. This is particularly important since CriMNet plans include integrating other state systems into the Backbone. 19

Project teams also faced challenges accurately reconciling project costs with available funding, project needs, and milestones. For some projects, including the Integration Backbone, project estimates or scope were cut to match available funds, sometimes without making clear to project stakeholders that original expectations regarding system capabilities would likely not be met. According to

52 CRIMNET Technical development of some CriMNet systems proceeded before legal requirements were adequately resolved.

The state must resolve policy issues pertaining to data practices and technical issues related to system security.

18CriMNet Office, MJNO Grant Award and Expenditures, 3/ 1/ 03 through 12/ 31/ 03 , January 8, 2004. 19Two systems that had been slated for connection to the Integration Backbone but are now on hold include the Driver Vehicle Services system and the Automated Pawn System. The first is a system that includes information on driver identification, driving records, and license status. The Automated Pawn System is a database of transactions by pawnshops and is used as a law enforcement investigation tool.

CriMNet officials, the Legislature sometimes places unrealistic funding restrictions on projects. For example, the Legislature set the funding and one-year schedule for the BCA's arrest photo project, and the project proceeded to fit those limits. 20 In other instances, projects required additional funding for such things as technology "refinements" and "enhancements." While modifying a project to better suit users' needs makes sense, better up-front planning may have allowed some of these additions to be built into the project plan from the start.

Contracts Writing tightly constructed contracts is a challenge for many state and local agencies, and our review showed it to be true for CriMNet as well. Common issues among some of the projects we reviewed included (1) failure to clearly describe vendor work products and performance criteria and (2) obligating the state to perform work supporting the project without having suff icient resources to provide needed state products on time.

To ensure that the state is receiving the best value for its money , state contracting guidelines call for contracts to clearly specify technical requirements, work products, performance standards, and roles and responsibilities of the state and vendor. Project managers are also responsible for critically evaluating and documenting whether products delivered by the vendor satisfactorily perform the required function or provide the desired information.

For several CriMNet projects, vendor contracts proved to be poor indicators of eventual project work products and outcomes. For example, in our review of the Integration Backbone project, it was not clear that all vendor products met the state's intent as specified in the contract. We did not independently confirm the adequacy of the vendor's work products; instead, we requested and reviewed documents related to CriMNet Office quality reviews or "acceptance testing" of vendor products before payment went out. Some of this documentation was vague, though we could not ascertain whether the acceptance testing was not rigorous or whether the documentation of that testing was poor. We found at least one instance in which the vendor product clearly did not meet the intent of the contract. The vendor was required to submit a "Return on Investment" report to the state supporting the vendor's technical approach. This document provided minimal information about recommended system products, and the vendor's calculations did not support the findings as written. In this instance, we do not suggest that the vendor was incapable of producing better products, only that the CriMNet Office accepted a weak one. Similarly, the vendor contracts for the Statewide Supervision System sometimes did not clearly articulate the service or deliverable due under the contracts. Rather, the contracts and their subsequent amendments repeated general language about work products or project phases.

AGENCY INTEGRATION PROJECTS 53 It is not clear that all vendor work products met contract specifications.

20Currently, 71 law enforcement agencies submit photos to the database. The BCA expects other jurisdictions to add capacity to submit photos as funding becomes available.

Defining roles and responsibilities is also a critical component of effective contracting, as is following through on those responsibilities. On the Integration Backbone project, both the state and the vendor were responsible for various tasks, deliverables, and deadlines. However, the state failed to deliver some work products, which delayed the project or limited the scope of what the vendor could do. In another instance, the state did not provide the vendor with specifications for developing a complaint form in a timely manner. According to Backbone managers, the state did not meet its contract obligations most often because CriMNet officials underestimated the complexity and scope of the work or they did not have enough staff or the right expertise to complete required tasks. In assessing the vendor's performance, the project manager stated that the vendor's technical staff "performed to the extent that the state allowed them to."

Coordination In a program as inherently collaborative as CriMNet, good coordination is paramount. We found several examples highlighting the pitfalls of missing collaborative opportunities. These problems point to the benefits of identifying and managing dependencies among projects, joint problem solving, and sharing lessons learned.

Because the criminal justice system itself is highly interdependent, progress on one project is sometimes contingent on completion of tasks for a separate system. On most of the projects we reviewed, tasks were rescheduled to wait for completion of tasks on one or more other systems. As a demonstration of MNCIS capabilities, for example, court staff are developing an "Ecomplaint," a document that allows prosecutors to electronically file a criminal complaint via the Integration Backbone that flows from the prosecutor's information system to MNCIS, saving both county attorney and court staff time processing paperwork. The complaint form also interfaces with the state's electronic Statute Table, which ensures accurate data entry on charging and convictions. Developing this prosecution form requires coordinating complaint filing procedures and programming efforts among all agencies to make this workflow function effective. The collaborative effort necessary to resolve this issue took longer than expected. As a result, the state did not provide the Integration Backbone vendor with Ecomplaint form specifications according to deadlines under the contract.

In 2000, the state had funded a pilot project in Ramsey County to develop a similar complaint form, but the state was not able to capitalize on these efforts in the current Ecomplaint project. According to court officials, the courts used the Ramsey County form as a starting point for its Ecomplaint, but the technology for the Ramsey County form was not compatible with MNCIS. Moving forward, improved coordination of integration efforts among agencies could help reduce duplication of effort and mitigate costs for the state. Coordination among agencies is important for identifying dependencies as soon as possible in order to reconcile differences or make schedule adjustments. 21 The CriMNet Office intends to take on a greater role in identifying and coordinating dependent activities among integration projects.

54 CRIMNET Coordinating integration project activities among agencies typically took longer than expected.

21We found several other examples of challenges coordinating project dependencies. One portion of the Statewide Supervision System had to be redone because it was found to be incompatible with another evolving system— the Integration Backbone (see the Department of Corrections Statewide Supervision System contract #A24827). Tasks on the Predatory Offender Registration System were delayed while waiting for adequate completion of tasks on the Department of Public Safety's Standard Interface Project (see Department of Public Safety contract #A16205 Amendment).

 

Meeting the goals and objectives of the CriMNet program is also highly dependent on joint problem solving among state and local agencies. Improving information sharing requires improving day-to-day practices, determining roles and responsibilities, and active participation in resolving issues pertaining to agencies' work practices. The Suspense File program is a case in point. Local law enforcement staff around the state must consistently follow proper f ingerprinting procedures during arrest and booking to ensure that accurate fingerprints are attached to criminal history records, but their efforts only partially resolve the problem. For example, some individuals'initial contact with the criminal justice system is via a warrant to appear in court for a hearing. It is at this point that their fingerprints must be captured, but the state does not have a standard procedure for ensuring that this occurs. Some courts have Livescan machines in their courtrooms, while others will not allow fingerprinting on the premises. 22 This example clearly illustrates the need for reconciling criminal justice processes, agencies' work practices, and legal requirements to meet a basic CriMNet objective— preventing incomplete criminal history records.

In all of the projects we reviewed, state agencies experienced missteps and unanticipated detours from original plans. Those agencies that tackled statewide integration projects, in particular, could share the benefit of their experience. On the Statewide Supervision System project, for example, integrating local jurisdictions'information systems for 87 counties took much longer than expected. According to Department of Corrections officials, delays occurred for two primary reasons. Some counties were hesitant to allow statewide access to their corrections data because of concerns regarding system security and data practices. Second, once these concerns were alleviated and counties agreed to integrate with the state system, the department had to coordinate tasks, work schedules, and funding with each participating county. In doing so, department officials said that coordinating the work of local jurisdictions' vendors was particularly challenging. Integrating local detention facility data, for example, required coordinating the work of 15 different vendors. Other agencies can learn from the department's experiences— particularly as the state begins bringing local systems on to the Integration Backbone. By documenting and sharing insight into obstacles and their resolution, other local and state agencies can improve their planning and implementation strategies, mitigating costs for the state as a whole.

Similarly, long-term experience with operating an integrated statewide court system and familiarity with court personnel needs undoubtedly facilitated the planning and implementation process for the courts. The large staffing complement assigned to the MNCIS project also likely contributes to effective project outcomes. Court administrators could share these experiences with other agencies, such as the CriMNet Office, to help others avoid "reinventing the wheel" and to enlighten stakeholders as to the difficulties associated with implementing a statewide integrated system.

AGENCY INTEGRATION PROJECTS 55 Most of the CriMNet projects we reviewed experienced some missteps or detours.

Law enforcement and courts personnel must work together to obtain offenders' fingerprints.

22Some courts personnel would like state law amended to explicitly authorize them to proceed with fingerprinting these individuals immediately following their court appearance. Other court administrators believe that fingerprinting is not the role or responsibility of the court and would prefer to direct law enforcement staff to complete this task. In one jurisdiction, according to a local CriMNet representative, judicial authorities have excluded Livescan machines from the courtrooms. In this instance, the local law enforcement agency has placed a Livescan machine in the skyway connecting the judicial building and sheriff's department to allow them to more easily capture fingerprints outside of the courtrooms.

 

CONCLUSIONS Planning, designing, and estimating costs for such large projects can be challenging under the best of circumstances. In spite of these challenges, the state has achieved significant progress in improving criminal justice professionals' access to timely information, as described in Chapter 2. In the process, however, matching evolving systems with complex criminal justice processes, user needs, and agencies' work practices has proven to be more complex than originally assumed.

Agencies' approaches to planning CriMNet projects vary, and on the whole, they have a poor track record for accurately estimating project timeframes. Failure to identify and resolve critical work practice and legal issues resulted in project delays, with insuff icient staffing and funding also being contributing factors. More importantly, future integration of additional state and local systems, such as law enforcement and prosecution systems, will not proceed until CriMNet officials address data practice concerns, state standards, and user requirements.

In the past, state agencies typically approached development and modification of their computer systems and work practices from an individual agency perspective. The goal of CriMNet is to step away from that model and approach criminal justice functions and systems as a unified, statewide service. Missed opportunities for coordinating integration efforts and conflict over agencies' roles and responsibilities are factors impeding CriMNet's progress.

RECOMMENDATION RECOMMENDATION To help facilitate criminal justice integration and mitigate costs for the state, Policy Group members should ensure that CriMNet projects managed by their respective agencies:

Have documented baseline expectations regarding project schedules, budgets, and scope;

Have adequately identified and addressed prerequisite decisions regarding users' requirements, data practices, and other criminal justice practices; and

Coordinate and communicate with stakeholders on other CriMNet projects.

56CRIMNET Missed opportunities fo r coordinating integration efforts have impeded CriMNet's progress.

As we describe in the following chapter, Minnesota has guidelines and manuals for helping agencies plan and manage projects. Using these tools will help CriMNet officials evaluate the progress of CriMNet projects, assess the value of the state's investment, and facilitate progress towards integrating criminal justice information.

Minnesota has invested substantial time and resources to develop and implement two statewide systems, the Statewide Supervision System and MNCIS. Similarly, several counties are working hard to improve their ability to share critical information. Through the CriMNet Office, other state and local agencies could benefit from their expertise to help guide future CriMNet projects.

AGENCY INTEGRATION PROJECTS

Program Management SUMMARY By statute, CriMNet is supposed to be managed according to industry best practices. For the State of Minnesota, these practices are articulated in a state project management methodology created by the Office of Technology. In comparing CriMNet management to these standards, we found that the CriMNet Office and the CriMNet Policy Group failed in the program's early years to complete critical program planning activities, such as clearly defining program objectives and scope. By late 2002, these shortfalls, along with communication problems and staff shortages, contributed to disappointing outcomes, including failure to establish important integration standards and conflicts among some stakeholders. Over the past year, the Policy Group and CriMNet Office have acted to complete needed program planning tasks, boost staffing, and improve day-to-day management. While progress has been slow and there is still work to be done, CriMNet has become more clearly focused and collaboration among jurisdictions is improving. Still, sustained leadership will be needed to shift the focus of CriMNet program activities from planning to active implementation. We recommend several actions to improve CriMNet Office operations and to strengthen oversight and accountability.

I n earlier chapters, we discussed the status of criminal justice information integration thus far and the extent to which agency integration projects proceeded as planned. In this chapter, we take a broader look at management of CriMNet as a whole. Our focus is on two entities— the Policy Group and the CriMNet Office. By law, the CriMNet Policy Group is responsible for (1) setting CriMNet's strategic direction, and (2) ensuring that CriMNet operations follow generally accepted management techniques and meet intended outcomes. 1

This chapter addresses the following questions: To what extent has CriMNet been managed in accordance with accepted program management practices, and what are the consequences of any shortcomings? What factors contributed to any identified program management weaknesses?

To what extent have recent management actions addressed shortcomings? 1Minn. Stat . (2002), 299C. 65, subd. 1b.

To answer these questions, we interviewed CriMNet Office managers and staff, Policy Group and Task Force members, and state and local agency CriMNet project managers. We reviewed reports summarizing the results of two previous program evaluations— one issued by Minnesota's Office of Technology in October 2002 and a second issued in February 2003 by an outside consultant hired to complete a legislatively-mandated CriMNet risk assessment. To avoid duplication of effort, we relied heavily on the Office of Technology report, in particular, to focus on important issues. In addition, we reviewed a wide variety of CriMNet program documents, reports to the Legislature, Policy Group meeting materials and minutes, and various publications used to explain CriMNet to criminal justice professionals, the Legislature, and the public.

The chapter is divided into three main sections. In the first, we briefly describe the generally accepted planning and management standards that served as the framework for our analysis. The second section describes the state of CriMNet program management in 2001 and 2002. In it, we discuss (1) the extent to which CriMNet planning and management practices conformed to standards, (2) the consequences of any weaknesses, and (3) underlying factors that help explain why things happened the way they did. In the third section, we discuss Policy Group and CriMNet Office actions, from early 2003 forward, to redirect CriMNet management and the extent to which these efforts have addressed identified problems. Recommendations for legislative and Policy Group action are discussed at the end of the chapter.

PLANNING AND MANAGEMENT STANDARDS

Minnesota state government has adopted a standard project management methodology, and we used this methodology as a framework for assessing CriMNet program management. 2 Developed by the Office of Technology, the methodology is based on longstanding, nationally-recognized best practices for initiating, planning, managing, and closing information technology projects. As summarized in Figure 4.1, the methodology defines key activities that should take place and documents that should be produced at various stages. Although the state methodology was developed with individual information technology projects in mind, the principles and standards also apply to management of broader programs, such as CriMNet. Given that CriMNet is an ongoing program, our evaluation focused on the planning and management standards rather than those for project initiation and closure.

It is important to keep in mind that standard practices are tools to help a project meet intended outcomes; applying these practices is not, by itself, the goal. It is possible, for example, to have all of the right processes and documents in place and still have a product that does not meet expectations. Similarly, a haphazardly managed project can sometimes result in an effective solution. Throughout the discussion in this chapter, we compare CriMNet planning and management practices to state standards, but in doing so, we also try to show how the presence or absence of standard practices affected the pace, cost, and quality of the work that was completed.

60 CRIMNET The CriMNet Office should f ollo w Minnesota's standard project management methodology.

2Minnesota Office of Technology, State of Minnesota Project Manag ement Methodology, v. 1.8 (St. Paul, 2002).

PROGRAM MANAGEMENT 61 Initiate Major Activities Identify an executive sponsor Document the business case that justifies the project Key Document: project charter Major Activities Establish a governance structure Appoint a project manager Define project objectives and scope Establish work plan, schedule, budget Identify resource needs and ensure they are available

Establish control procedures to manage the project once work is underway

Key Documents: strategic plan, scope statement, and work plan

Plan

Major Activities Track and report on project status Communicate regularly Identify and resolve issues and risks Manage changes to time, scope, budget Make go/ no-go decisions at checkpoints

Key Documents: status reports, updated plans, and project work products

Manage

Major Activities Review and accept final work products Assess whether outcomes met objectives

Key Document: project closeout report

Close

Project Stages Figure 4.1: Major Elements of Minnesota's Project Management Methodology

SOURCE: Minnesota Office of Technology, Quick Reference Project Start-up Guide (St. Paul, 2003) and State of Minnesota Project Management Methodology, v. 1.8( St. Paul, 2002).

 

We consider CriMNet program management to encompass two areas: (1) managing the CriMNet Office's internal operations, and (2) coordinating and overseeing, at a higher level, progress of CriMNet projects implemented by various state and local agencies. CriMNet Office internal operations include such things as developing and maintaining a strategic plan, administering local grants, identifying and documenting user needs, and defining security standards. Program coordination activities include such things as maintaining a programwide budget, tracking the status of state and local projects, and ensuring that these projects conform to the state's integration standards.

We divided our analysis of program management practices into two general time periods. The first covers calendar years 2001 and 2002. This timeframe reflects what we consider to be a start-up period for the CriMNet Office and the beginning of significant state funding for CriMNet. The second time period covers January 2003 to the present. This period reflects a number of events to redirect CriMNet Office operations, including several program reviews that criticized aspects of CriMNet operations and a concurrent change in CriMNet Office executive leadership.

PLANNING AND MANAGEMENT, 2001-02 Although the state had been pursuing criminal justice information integration for a number of years, the CriMNet Office was essentially a start-up organization in 2001. Establishing program operations was complicated by the fact that the office was entering an ongoing endeavor mid-stream. Early CriMNet staff were faced with (1) pulling together integration projects already underway at various state agencies, (2) determining additional needs, (3) clarifying the CriMNet Office's role in meeting those needs, and (4) moving forward under the agreed-upon management model. In this section of the report, we use a framework of standard planning and management practices to discuss the extent to which CriMNet Office operations were structured to meet these challenges and problems that resulted from any shortfalls.

In comparing CriMNet management through 2002 to accepted practices, we found that:

After the CriMNet Office was formally established in 2001, CriMNet Office managers and the Policy Group failed to make crucial planning decisions, such as defining CriMNet's objectives and scope, or to establish standard mechanisms for monitoring progress and resolving problems. These weaknesses had negative effects on priority setting, budgeting, and staffing.

62 CRIMNET The CriMNet Office faced the challenges of a start-up organization in 2001.

Planning For CriMNet to be more than a loosely connected set of projects initiated by various jurisdictions, it must have an effective program planning process. That process should include planning decisions that lead CriMNet from a broad statement of vision (statewide integration), to a narrower set of objectives (statewide search of local criminal history data, for example), to the specific projects and activities that must take place to achieve those objectives (such as, resolving data practice issues or acquiring a technical product), to the order in which those projects and activities will be done given available resources. The purpose of approaching planning this way is to make sure resources are targeted to the highest-priority projects and work gets done in the right order.

Planning decisions regarding CriMNet's scope and direction did not proceed according to this process. Based on our review of CriMNet program documents, the Legislature and CriMNet stakeholders have had a unified sense of CriMNet's vision— integrating criminal justice data to make it easier to identify offenders and know their full criminal histories. However, the Policy Group and CriMNet Office failed to fully translate this vision into an agreed-upon set of objectives and an associated statement of projects and activities to support them. The Policy Group and others were aware that disagreements over CriMNet objectives were hurting the program, and in September 2002, Policy Group and Task Force members, CriMNet Office staff, and local representatives met at a multi-day retreat. The purpose of the meeting was to reach consensus on CriMNet goals and objectives, but according to the facilitators' f inal report, participants were not able to do so.

Absence of a defined CriMNet program scope manifested itself in several ways. For example, defining user needs is an activity that fell under the purview of the CriMNet Office but was never pulled into the scope of CriMNet Office activities. Several of the ongoing projects, notably the Integration Backbone, suffered for lack of clearly stated outcome requirements the system was to achieve. In addition, under standard practices, decisions regarding program scope are used to develop work plans. Work plans are tools to coordinate the timing of interdependent activities; to hold staff accountable for completing assigned work; and to develop staff, schedule, and funding estimates. The CriMNet Office operated through 2002 without a work plan integrating its activities with major work phases of ongoing state and local projects.

Program Controls Accepted program practices also call for setting up various control processes, described in Table 4.1, that should be used to provide a structure for managing the program on a day-to-day basis. The purpose of these processes, individually and collectively, is to help identify problems and find appropriate solutions.

PROGRAM MANAGEMENT 63 Through 2002, the Policy Group and CriMNet stakeholders could not agree on CriMNet program objectives. 7

As with other program planning steps, CriMNet officials did some of the tasks necessary to establish these control procedures, but failed to do so fully. For example:

Status reporting to the Policy Group was sporadic through 2002, and reports were inconsistent in terms of content and quality;

The CriMNet Office did not comprehensively track costs for its own activities, local grant projects, or other agency projects; Communication about CriMNet was not guided by a Policy Group-approved communication strategy or review procedure; The CriMNet Office did not systematically identify and resolve programwide implementation issues or risks; and

64 CRIMNET Table 4.1: Standard Program Controls Control Mechanism Definition Steering Committee Meetings A steering committee is a group of key stakeholders. Its purpose is to provide high-level project oversight, to make key decisions regarding the project's direction, and to help solve problems. To provide consistent direction, meetings should have a standard agenda that includes reviewing project status, reviewing and approving scope changes, formally accepting major work products, assessing accomplishments against expectations set at the last meeting, and making go/ no-go decisions at appropriate points, such as the end of a design phase. Monthly Status Report A report that documents progress, issues, cost, and schedule status. It is used to help identify and resolve problems and to measure progress against original expectations.

Communication Management A process to ensure accurate and sufficient communication about the project such that (1) stakeholders and project team members are informed and aware of project activities and status and (2) messages to external audiences support the project's goals and objectives.

Risk Management A process to assess factors, such as insufficient staffing, that may lead to project failure and to develop strategies to avoid or mitigate them. Issue Management A mechanism to document, track, and resolve obstacles that arise during project implementation and require a coordinated effort to fix. Change Management A process to ensure that changes to scope, cost, schedule, and work products are formally reviewed, approved, and documented so that project teams, reviewers, and other stakeholders share a common set of expectations.

SOURCE: Minnesota Office of Technology, State of Minnesota Project Management Methodology, v. 1.8( St. Paul, 2002) and Quick Reference Project Start-up Guide (St. Paul, 2003).

CriMNet lacked some standard policies and procedures that w ould have helped with day-to-day management.

The CriMNet Office did not have a procedure in place to ensure that major program changes, such as adding to a project's scope, were reviewed, approved, and communicated. Some of these management practices would have been difficult to fully implement because prerequisite work had not been completed. For example, meaningful status reporting requires baseline expectations against which to measure progress. Absent time, scope, and budget planning estimates for CriMNet Office work, a status report on progress was of limited value. As we discuss in more detail later, f inancial tracking for CriMNet is complicated by a number of factors, including (1) limitations in the state's accounting system that make it difficult to isolate CriMNet spending and (2) questions regarding the extent to which agencies are obligated to provide project cost and status information to the CriMNet Office and Policy Group.

Consequences As we said earlier, compliance with standard practices is important insofar as it affected the ability of the Policy Group and CriMNet Office to manage the program. In that regard, we found that:

By late 2002, there were clear signs of CriMNet management problems and lost momentum.

Program management weaknesses made it more difficult to manage CriMNet on a day-to-day basis and to identify and resolve problems. This resulted in incomplete information regarding CriMNet's costs and future investments, unrealistic expectations regarding the extent and nature of ongoing integration activities, and conflicts among some stakeholders. In the end, important tasks did not get completed, and CriMNet lost momentum.

As discussed above, CriMNet did not have several needed tools that would have helped CriMNet managers, the Policy Group, and others outside of the program understand CriMNet spending and progress. For example, ideally, the CriMNet Office would maintain (1) a CriMNet Office budget and project-specific budget information linked to major milestones (such as a project design phase), and (2) data on spending compared to budget. However, state and local agencies that are implementing integration projects have not produced consistent statements of spending to date or reliable cost projections.

In reviewing CriMNet documentation, we found few examples of detailed budget proposals or spending reports used internally by the CriMNet Office or by the Policy Group. According to current CriMNet officials, staff created one of the f irst comprehensive spending reports in January 2003, though they also reported that it was difficult to compile and not as detailed as they would have liked. Absence of a program plan has also undermined efforts to discuss the specifics of future investments. Without comprehensive financial data and a plan for CriMNet's next steps, CriMNet officials were not able to answer basic questions, such as ours, regarding how much has been spent to date or what investments are expected in the next several years.

PROGRAM MANAGEMENT 65 CriMNet policymakers and managers did not have reliable spending data.

Throughout our evaluation, we noted many instances in which CriMNet stakeholders voiced expectations for CriMNet that had not been met— in some cases because expectations were unrealistic and in others because work products were not delivered as expected. For example, the CriMNet Office asked pilot counties in early 2002 to focus their efforts on intra-county system enhancements and integration rather than statewide integration until the state developed work process, data, and technical standards. These standards, in large part, have not been forthcoming. In general, stakeholders said that unmet expectations were a f actor that eroded trust. CriMNet's success requires participation and collaboration at all levels, so lack of trust can have serious consequences.

In a broader sense, aggressive marketing of the CriMNet vision has contributed to gaps between stakeholders' beliefs regarding the extent of integration to date and actual progress achieved. Many of the CriMNet materials that we reviewed— literature, videos, and other presentations— did not clearly distinguish between actual progress to date and integration capabilities that are planned for the future. For example, the CD-ROM multimedia product depicting search, subscription, and workflow models for the Integration Backbone project (discussed in Chapter 3) accurately presents the vision for criminal justice information sharing in Minnesota. But, this presentation does not make clear that many of the functions for moving information among systems are actually years away from being implemented. Clearly articulating the end-state vision for CriMNet is important, but we think it is equally important to establish clear and realistic expectations for the incremental steps required to move from the current state of integration to the vision.

Some local government representatives reported to us that their participation in CriMNet and overall confidence in state CriMNet administration had declined by varying degrees through 2002. Dissatisfaction with the 2001-02 local grant award process was a primary source of frustration. According to CriMNet documents, the grant criteria changed fundamentally between the first grant announcement issued in the fall of 2001 and the final grant request for proposals issued in February 2002. The first solicited grant proposals related to inter-jurisdictional integration; the final criteria gave strong preference to local system enhancements that would facilitate integration at a later date. In addition to this change in direction, local representatives were particularly frustrated by grant requirements that explicitly prohibited communication and coordination among grant applicants, a requirement that appears contrary to the notion that CriMNet is inherently a cooperative endeavor. Overall, local governments were discouraged by the time, effort, and uncertainty involved in preparing multiple grant proposals, and Minnesota's two largest counties, Hennepin and Ramsey, eventually dropped out of the state's grant process at that time. 3 Hennepin County continued with its integration projects using county funds and grants received directly from the federal government and later reentered the state's process, receiving a CriMNet grant in November 2003 for $420,000. According to the Ramsey County official that led its 2001-02 grant proposal, the county's implementation of its integration plan has largely stalled.

66CRIMNET Aggressive marketing of CriMNet created expectations that could not realistically be met.

Some local stakeholders lost conf idence in CriMNet.

3County representatives also reported that the grant application was long and cumbersome and that the grant criteria were applied inconsistently, with the final terms being determined through a negotiation process with CriMNet Office officials.

City and county representatives voiced other misgivings, including (1) concerns over the accuracy of reports to the public and the Legislature regarding the actual status of the Integration Backbone project and statewide integration as whole, and (2) failure of the state to provide technical specifications for how local systems should be configured. Several of the implementation grant counties reported that they have had little interaction with the CriMNet Office as their grant projects are proceeding. Open communication with the Policy Group and CriMNet Office and state leadership on standards are particularly important issues to cities and counties because the state integration model requires that local governments make significant financial investments.

Factors Contributing to Program Management Weaknesses

A number of systemic factors contributed to the problems just described. CriMNet officials struggled to establish a focused program management structure in 2001-02 in part because of the inherent complexity of the task at hand and an understandable learning curve. We found that:

A variety of factors contributed to program management shortcomings, including CriMNet Office staffing issues, insufficient attention to important program management responsibilities, and governance weaknesses.

We discuss each of these issues in turn. Staffing Problems The CriMNet Office has had staffing problems since its inception. Based on our analysis of CriMNet Office staff records and interviews with CriMNet officials, the CriMNet Office through calendar year 2002 lacked both a suff icient number of staff and an appropriate skill mix for an effort as large and complex as CriMNet. When first created in early 2001, the CriMNet Office had about three full-time staff. As shown in Table 4.2, the office in fiscal year 2002 had approximately ten full-time staff. During these years, CriMNet Office managers relied heavily on volunteers from other state and local agencies that donated time to CriMNet. While this donated staff time is an important demonstration of shared responsibility for criminal justice integration, CriMNet cannot function effectively over the long-term without a sufficient core of full-time staff. A shortage of full-time business and technical analysts in the CriMNet Office most certainly slowed efforts to complete key program management tasks and technical work to support the Integration Backbone project. Staffing increases in fiscal year 2003 were a direct attempt to address the staffing problem. But, of the approximately 17 full-time staff shown in Table 4.2, only 7 were with the office for all or nearly all of the fiscal year. Most of the remaining positions were added between October 2002 and March 2003, and two were filled at the start of the f iscal year but vacant by March 2003. In comparison, the courts'MNCIS project team included over 20 administrative, business, and technical staff in fiscal year 2003, and the CriMNet Office staffing plan approved by the Policy Group in 2003 (discussed below) authorizes 26 CriMNet Office positions.

PROGRAM MANAGEMENT 67 The CriMNet Office has been understaffed from the start.

It is not clear to us why staffing problems persisted for so long. The CriMNet Office deputy directors pointed to a number of inhibiting factors— lack of funding for new positions and lack of a clear directive from the Policy Group to initiate hiring. While these are certainly barriers, neither appears to be insurmountable. We found evidence that CriMNet Office officials repeatedly reported to the Policy Group in 2002 that staff shortages were slowing progress, and Policy Group members acknowledge that understaffing was, and still is, a problem. It also is not clear to us that funding for staff positions was unavailable, but assuming that state CriMNet funding was insuff icient to support additional staff, federal funding was an option. As we discuss below, current CriMNet Office officials sought and received federal funding in 2003 to support new staff positions. Overall, we found understaffing to be a clear case of a pressing need with a systemic failure to address it.

Administrative Problems We also found that former CriMNet Office managers did not pay suff icient attention to the full range of their program management responsibilities. 4 This is particularly true of the director who was in place from late 2001 to late 2002. He was described by many of the officials we interviewed as a "poor fit" for CriMNet. Our review of program documents produced during his tenure supports officials' views that the former Executive Director focused his attention primarily on the technical aspects of the Integration Backbone project and placed little emphasis on broader CriMNet concerns. He did not pay adequate attention to documenting the needs of prospective system users, establishing technical standards, conducting long-term planning, or building collaborative relationships.

68CRIMNET Table 4.2: Estimated Number of CriMNet Office Staff Positions Filled, FY 2002-04 Estimated Number of Positions Filled Primary Area of Responsibility FY 2002 FY 2003 a FY 2004 Program Management 5.0 6. 5 7.5 Integration Backbone Project 4.5 10. 0 9. 0 CriMNet Office Total 9.5 16. 5 16. 5

NOTES: Position counts are estimates and include administrative, professional, and supervisory full-time positions filled at some point in the fiscal year. Counts include contract employees hired to do CriMNet Office work and staff from other state agencies working for the CriMNet Office under formal interagency agreements. Counts exclude student interns and state and local government employees doing CriMNet work through informal agreements. a Totals for fiscal year 2003 overstate the number of staff on board for the full fiscal year. Of the total positions shown, only seven of the staff worked at the CriMNet Office for all or nearly all of the fiscal year.

SOURCE: Office of the Legislative Auditor analysis of CriMNet Office staff lists and CriMNet Office estimates of full-time equivalent positions.

The Policy Group and CriMNet Office let staffing problems persist too long.

4The CriMNet Office had an Executive Director for a short time in early 2001, followed by an interim Director, and finally a third Director who served from late 2001 through 2002.

To monitor overall progress of the CriMNet program, the Policy Group expected to get programwide information— including the status of agency and local projects— from the CriMNet Office. But according to officials we interviewed, some state and local agency project teams were reluctant to provide such information. It was not clear to us, though, if they were hesitant because they mistrusted the CriMNet Office or because they did not find reporting to the CriMNet Office appropriate. In any case, it is imperative for the CriMNet Office to serve as a conduit of programwide information to ensure compliance with the state's integration model and to help facilitate the entire process. From an accountability perspective, CriMNet projects led by state and local agencies receive state funds, and these agencies should report to the CriMNet Office as needed.

Beginning in the summer of 2002, the Policy Group started making stronger and more specific requests that the Executive Director meet his program management responsibilities. Two primary vehicles through which they did this were requests for regular, programwide status reports and written documentation regarding CriMNet program scope. According to CriMNet documents, the former Executive Director cited a number of reasons for delays in meeting these expectations, including staffing shortages, pressure to complete the Integration Backbone project, poor cooperation from agencies running CriMNet projects, and lack of clear direction from the Policy Group regarding program scope. While each of these factors undoubtedly played a part, the Policy Group was dissatisfied with his attempts to rectify these problems and voted in January 2003 to terminate his contract.

Governance Weaknesses Because of the systemic and long-standing nature of many of the management problems we have discussed, we think governance must also be considered a contributing factor. The Policy Group is, by law, accountable for CriMNet program outcomes, and its role encompasses both setting strategic direction and ensuring that the strategy is implemented via the CriMNet Office. As we discuss later in the chapter, the Policy Group's membership brings perspectives that are needed to discuss and set CriMNet strategy. But, the extent and duration of CriMNet Office performance problems experienced through 2002 highlight limitations in the Policy Group's ability to manage day-to-day CriMNet Office operations.

Documents we reviewed show that the Policy Group met frequently throughout 2001 and 2002. Policy Group members were clearly committed to CriMNet and made good-faith efforts to guide the program. But, after identifying CriMNet Office performance problems in the spring of 2002, the Policy Group allowed the problems to continue without measurable improvement for the remainder of the year. In our view, the Policy Group did not provide the level of timely oversight and action needed in a poor-performance situation. The Policy Group met and discussed performance-related issues, but did not provide the follow-through required to rectify problems.

PROGRAM MANAGEMENT 69 Former CriMNet Executive Directors did not complete key tasks, such as assessing CriMNet user needs.

The Policy Group did not provide the day-to-day oversight required to rectify management problems.

Day-to-day monitoring and direction for CriMNet Office operations might have come from the Department of Public Safety where the CriMNet Office is an organizational unit. But through 2002, the CriMNet Office was not substantively integrated into Public Safety's management structure. Public Safety provided support services, such as payroll, personnel, and contracting assistance, but we found little evidence that the department provided authoritative, day-to-day supervision, or that the Policy Group encouraged Public Safety to play this stronger role.

Our findings regarding program management problems and governance weaknesses through 2002 should not be construed to mean that CriMNet had gone completely awry. As we discussed in Chapter 2, the state made important progress in integrating criminal justice information in spite of these problems. In addition, the Integration Backbone project continued to move forward. Rather, our conclusion is that the program was not managed with the rigor and discipline needed for such a complex and important venture. Some important tasks, such as resolving data practice issues, setting integration standards, and planning for local integration, were not adequately addressed. As a result, at the end of 2002, the CriMNet Office was ill-positioned to move forward with new integration initiatives.

ACTIONS TAKEN TO IMPROVE CRIMNET MANAGEMENT, 2003-04

The Policy Group came to a similar conclusion about CriMNet Office performance problems, and in September 2002, it directed Office of Technology staff to review CriMNet operations. In its report of October 2002, the office made numerous recommendations, including that the CriMNet Office identify integration requirements, define the scope of CriMNet Office operations, and develop an overall program plan. They also recommended changes to staffing, communication, and program controls. In essence, the Office of Technology recommended that CriMNet be managed according to the standards set forth in the state's program management methodology. 5

Following this review, the Policy Group initiated a series of events that, by January 2003, resulted in the CriMNet program management function essentially being restarted. We assessed the nature and extent of these efforts to reinvigorate CriMNet program management and found that:

Beginning in early 2003, the Policy Group and a new CriMNet Executive Director initiated a mid-course correction that demonstrates a strong commitment to strengthening program management. Actions to date have been aimed primarily at rectifying planning shortfalls. These efforts are an essential investment to get and keep CriMNet on track, but we also found that:

70 CRIMNET The Policy Group and CriMNet Office initiated a mid-course correction in early 2003.

5Minnesota Office of Technology, CriMNet Program Review Report , Presentation to the CriMNet Policy Group, October 25, 2002.

Despite recent corrective actions, progress is still slower than expected, in large part because insufficient staffing remains a critical problem. Sustained leadership and resources will be needed to shift the focus of CriMNet Office activities from planning and capacity building to active implementation of projects and activities that directly further statewide integration.

In the last year, CriMNet officials have made progress in several areas: (1) staffing, (2) defining CriMNet's scope and direction, (3) governance, and (4) program controls. We discuss each of these areas in turn, focusing on the extent to which CriMNet management is getting closer to meeting state standards.

Staffing CriMNet has made limited progress in rectifying persistent staffing problems, as summarized in Table 4.3. Overall:

A change in executive leadership and approval of a new staffing plan are steps in the right direction, but the CriMNet Office needs to get permanent staff on board.

One of the most important efforts to improve CriMNet management was a change in executive leadership at the CriMNet Office. In December 2002, the Policy Group named the Office of Technology Director as interim Executive Director, and in January 2003, voted to terminate the former Executive Director's contract. In February 2003, the Policy Group named the Director of Criminal Justice Information Systems (CJIS) at the Bureau of Criminal Apprehension as interim director. In October 2003, after a national search, the policy group named the interim director as permanent Executive Director.

PROGRAM MANAGEMENT 71 Table 4.3: Progress Addressing CriMNet Office Staffing Needs, as of January 2004 Action Needed Status Accomplishments Work Remaining Appoint a Program Director Implemented New CriMNet Executive Director appointed in October 2003. Provide sufficient assistance to balance CriMNet and Bureau of Criminal Apprehension responsibilities.

Ensure Sufficient Staff Resources Partially Implemented Staffing plan approved in June 2003. Necessary federal funding received in September 2003. Hiring process initiated for most positions.

Get needed staff on board. As of January 2004, three of the new staff positions had been filled.

SOURCE: Office of the Legislative Auditor analysis of CriMNet Office planning documents and staffing data.

The Policy Group has a plan to reorganize and enlarge the CriMNet Office.

As a temporary measure beginning in late 2002, the CriMNet Office also received an infusion of staff on loan from the State Court Administrator's Office, the departments of Corrections, Public Safety, and Administration, and counties. These staff were assigned to complete some critical tasks (for example, the 2002 report to the Legislature), to assess the then-current status of CriMNet operations, and to add technical resources to the Integration Backbone project.

But, relying on volunteer staffing is insuff icient for a program like CriMNet. With this in mind, the interim Executive Director developed a CriMNet Office organization and staffing plan, which was presented to the Policy Group and approved in June 2003. The plan calls for (1) changes in the organizational structure to better focus CriMNet Office resources on statewide issues, such as common work practices, user requirements, and integration standards and (2) an increase in CriMNet Office staff resources.

At the Department of Public Safety's suggestion, the Policy Group has discussed separating the Integration Backbone project team from the CriMNet Office, making it into a stand-alone entity within the Bureau of Criminal Apprehension's CJIS division. This would treat the Integration Backbone project like other CriMNet system projects— managed within a state agency but with reporting obligations to the CriMNet Office. Because CJIS maintains other statewide criminal justice information systems, we agree that this placement would make sense. For now, the Integration Backbone project team remains in the CriMNet Office and will be organized into two staff groups: (1) system development and maintenance and (2) operations support.

According to the organization plan, the remaining CriMNet Office staff will be divided into two main groups more clearly focused on program management, as shown in Figure 4.2. One group is to focus on criminal justice practices and user needs and the other on technical standards and assistance. Again, based on our discussion in Chapters 2 and 3 regarding the impact of shortfalls in these areas on specific projects, we agree with the plan's emphasis on addressing these issues.

We also agree with plans to increase the number of CriMNet Office staff. According to the staffing plan, the CriMNet Office would have a total of 14 full-time program management positions with the Integration Backbone team having another 12. To better ensure that the CriMNet Office and Integration Backbone project team have an appropriate mix of staff skills, the Policy Group also agreed that all positions would be staffed through a competitive process, meaning that existing CriMNet staff would have to apply for a position in the new organization.

Insufficient staffing, however, remains a critical problem. Although various personnel-related processes are underway, as of January 2004, few of the program management or Integration Backbone positions had been filled. The Executive Director said two factors underlie the slower than expected progress. First, state funding for the CriMNet Office was not suff icient to support planned staffing levels. As a result, the CriMNet Office had to obtain federal grant funding for over half of the planned positions; this funding was not secured until September 2003. Second, navigating the state's personnel procedures required to create and f ill new positions has taken longer than expected.

72 CRIMNET Staffing remains a critical concern because, to date, few of the authorized CriMNet Office staff have been hired.

PROGRAM MANAGEMENT 73 CriMNet Program Management

Business process modeling Data model definition User requirements definition Data integrity and quality Integration planning and implementation

Change control

Technology standards Data model implementation Technology tools Security model Integration infrastructure

Overall program guidance Program management Strategic planning Communication

Business Services Center

Technical Services Center

Figure 4.2: CriMNet Office Organization Plan for Program Management Staff, as Approved June 2003

NOTE: Along with program management staff organized as shown, the CriMNet Office includes an Integration Backbone project team, organized into two groups: (1) system development and maintenance and (2) operations support. SOURCE: CriMNet Office planning documents.

Program Direction a nd Scope In 2003, CriMNet officials made noteworthy progress in defining program objectives and setting CriMNet Office near-term priorities. However, some projects, particularly those related to defining user needs and technical system requirements, need to be completed to support credible planning for future integration. As shown in Table 4.4, the Policy Group formally adopted a CriMNet strategic plan in September 2003. The plan's goals, objectives, and approach are described in Chapter 1. Defining CriMNet objectives has been a divisive issue among CriMNet stakeholders, so Policy Group approval of the strategic plan was an important step toward resolving questions of program direction.

As we discussed earlier, it is important to translate these objectives into an agreed-upon set of projects and activities with accountability and resources assigned. CriMNet officials made progress in this regard on two different levels: (1) implementing a plan to bring program management policies and procedures up to standards and (2) setting an agenda for CriMNet Office projects to support integration, such as gathering user requirements and defining security standards. Although the scope of CriMNet activities in these areas had been the subject of intense discussion throughout 2003, the specifics have not been formally agreed upon. The CriMNet Office has drafted a document (called a scope statement) that communicates the details of the agenda for the next several years, and it will be debated by the Task Force and presented to the Policy Group for approval. 6 Given CriMNet's history, we think that completing this document is important, in that it will help set baseline expectations for products, time frames, and costs and establish control mechanisms to support effective oversight.

74 CRIMNET Table 4.4: Progress Defining Program Direction and Scope, as of January 2004 Action Needed Status Accomplishments Work Remaining Define Vision and Strategic Objectives Strategic Plan Completed Plan approved in September 2003. Develop a long-range strategic plan. Define Scope of Projects and Activities Partially Implemented Discussions underway and draft document prepared regarding specific projects and activities that will be led or coordinated by the CriMNet Office.

Task Force review and Policy Group approval (target date March 2004).

SOURCE: Office of the Legislative Auditor analysis of CriMNet Office documents and interviews.

The CriMNet strategic plan, approved in 2003, has not yet been translated into a more detailed work plan.

6The document is scheduled to go to the Policy Group for approval in March 2004.

As yet, CriMNet officials have not defined longer-term integration priorities. To better position itself to do so, the CriMNet Office needs to complete a number of important prerequisite activities, such as assessing user needs, assessing local jurisdictions'capacities to integrate, and resolving data practice issues. As discussed above, it is imperative that the CriMNet Office get needed staff resources on board to direct these efforts, which will, by necessity, continue to involve extensive collaboration with staff in state and local agencies.

Governance Changes The Policy Group, Task Force, and CriMNet Office made several changes in CriMNet governance to improve collaboration, more clearly define a CriMNet Office chain of accountability, and facilitate timely decisionmaking. Changes in 2003 included assigning more responsibility to the Task Force and embedding the CriMNet Office more fully within the Department of Public Safety management structure. At this point:

Governance changes initiated in mid-2003 make sense, but it is too early to assess their impact on accountability and decisionmaking.

In June 2003, the Policy Group approved a revamped charter for the Criminal and Juvenile Justice Task Force. The new charter modified the Task Force's roles and responsibilities to more closely align with those of a traditional technology project steering committee and to have its members, via subcommittees (called delivery teams), investigate specific problems and develop recommendations. For example, a Task Force subcommittee investigated data practice issues and, in December 2003, recommended to the Policy Group a list of legislative proposals. The Policy Group members we interviewed commented on the added value of this Task Force work, citing the strength of the background work that they do and the importance of obtaining input from the diverse perspectives presented by Task Force members. Using the Task Force in this way, rather than as simply a discussion forum, better reflects the collaborative nature of CriMNet and brings needed resources to the program.

Transition to the Task Force's new role has not been without bumps. According to some stakeholders, group members are at times struggling to shift their perspectives from advocacy for their respective jurisdictions to a more collaborative, problem-solving role. Several of the stakeholders we interviewed commented that trust is an issue in this transition. They added that trust among stakeholders at all levels eroded during 2002, and that melding the CriMNet Office, Task Force, and Policy Group into a unified force will take time.

As we said earlier, the CriMNet Office has always been an organizational unit within the Department of Public Safety, but through 2002, it operated largely outside of the department's day-to-day management structure. In 2003, the Commissioner of Public Safety changed that relationship and made the CriMNet Office an organizational unit within the Bureau of Criminal Apprehension, with the CriMNet Executive Director reporting to the bureau's superintendent. At its December 2003 meeting, the Policy Group tacitly supported this arrangement, although members raised several questions regarding the line between day-to-day supervision of CriMNet Office operations and the Policy Group's statutory authority. The Policy Group clearly has authority in setting the strategic direction of CriMNet Office work by, for example, approving the strategic plan. It also has broad oversight authority over CriMNet and the responsibility to ensure that expected outcomes are met. But, by more clearly placing the CriMNet Office within the Department of Public Safety, the department will have more control over day-to-day decisionmaking. 7

PROGRAM MANAGEMENT 75 The Policy Group's advisory task fo rce has been given greater responsibility fo r CriMNet projects.

Supporting work needs to be completed before CriMNet policymakers can set a longer-term integration agenda.

 

Not all stakeholders are comfortable with this arrangement. Some are concerned that Public Safety's influence on the CriMNet program will be too great, undermining the collaborative nature of CriMNet at a time when improving collaborative relationships is paramount. In contrast, others suggest that CriMNet be integrated fully into the Department of Public Safety, with the Policy Group functioning as an advisory body to the Commissioner of Public Safety and with no direct authority over the CriMNet Office. Based on our work, we have concluded that:

Keeping the CriMNet Office under the Policy Group's strategic direction serves important policy goals, but the office also needs direct day-to-day support and direction from the Department of Public Safety.

We agree that CriMNet's governance structure and operations should reflect CriMNet's multi-jurisdictional nature, so we see the merits of the Legislature's governance design and the importance of the Policy Group. But, Policy Group oversight of CriMNet Office operations has its limits, as demonstrated in the performance problems experienced through 2002. Given the complexity of the CriMNet Office's mission and the level of effort needed to bring CriMNet program management practices up to par, we think it is important to have the CriMNet Office embedded in a clear, day-to-day accountability structure, which the Policy Group simply cannot provide. The Bureau of Criminal Apprehension, given its existing responsibility for the state's criminal justice data network, is a logical choice. Where strategic direction ends and day-to-day supervision begins is, of course, a gray area. Minimizing conflicting direction to the CriMNet Office will require close communication among the Executive Director, the Commissioner of Public Safety, and the Policy Group.

Another governance issue raised in 2003 is whether the Policy Group should include one or more local government representatives as voting members. Some criminal justice functions, such as prosecution, also are not represented. In our view:

Expanding the Policy Group to include local representatives would provide a perspective that is currently absent and would better reflect the collaborative nature of CriMNet.

76CRIMNET Not all stakeholders agree with the decision to give the Department of Public Safety greater authority over the CriMNet Office.

7In December 2003, the Policy Group voted to recommend to the Legislature that the law be changed to state specifically that the CriMNet program manager serves at the pleasure of the Policy Group. According to Policy Group members, the recommended statutory language regarding the relationship between the CriMNet Office and the Policy Group simply states more clearly a relationship already established in the law— that the CriMNet Office was created to implement the work of the Policy Group. Since the Policy Group has always selected the CriMNet program manager, we do not take a position on this issue.

The Policy Group chose not make such a recommendation in its proposals for the 2004 legislative session, but this remains a high profile issue for cities and counties. 8 Those opposed to changing the Policy Group's composition said that they understood the philosophical importance of having local representation on the Policy Group, but were uncomfortable with the logistics of choosing who that might be. They also said that the majority of the Task Force is comprised of local representatives, and as restructured, the Task Force has a great deal of influence on CriMNet strategy and operations. Those in favor of local participation on the Policy Group argue that (1) cooperation from local jurisdictions is vital to CriMNet's success; (2) local criminal justice professionals have a unique, front-line perspective that should be reflected in CriMNet's strategic direction; and (3) a voting seat on the Policy Group is more influential than an advisory role on the Task Force. We find the rationale for adding local representation to be more compelling.

Program Controls CriMNet has made progress in implementing standard program control procedures, as summarized in Table 4.5. To date, efforts that are farthest along include formal program status reporting and strengthening program review and decisionmaking by the Task Force and Policy Group. Other controls are in various stages of design. Overall, we found that:

CriMNet officials have designed structured processes to help manage and oversee CriMNet operations, but it is important to shift from planning these controls to actively using them to manage CriMNet work.

Status Reporting The new CriMNet Office team took immediate steps in mid-2003 to improve program status reporting. At first, these reports were filled with lists of program management and Integration Backbone project tasks that had gone undone in previous years. Later reports focused on the status of efforts to complete this make-up work. Improvements in status reporting worked in tandem with the substantive work being done by the Task Force to better focus Policy Group meetings and to improve the quantity and quality of information available about CriMNet's status. Still, work remains in this area. As of December 2003, neither the Task Force nor Policy Group had consistently reviewed program status during their respective meetings. Instead, both groups'meetings focused on other issues, such as legislative proposals for data practices and other statutory changes. While these are, without a doubt, important issues, it is imperative from a program recommendation, a key sticking point for Policy Group members was that the Task Force chair might not necessarily be a local employee because the Task Force includes members from, for example, the state and private sector. At the December 2003 Policy Group meeting, the Commissioner of Public Safety suggested that the proposal be modified such that the Task Force chair would appoint to the Policy Group a Task Force member who was also a local jurisdiction employee. The Task Force Chair did not support this suggestion, and the motion failed.

PROGRAM MANAGEMENT 77 Local jurisdictions should have a place on the Policy Group.

8The Task Force twice put forward recommended statutory language that would add Task Force leaders to the Policy Group— once recommending that the Task Force chair and two vice-chairs be added and the second time recommending only the Task Force chair. In discussing the merits of the control perspective that both groups routinely scrutinize CriMNet progress against baseline expectations.

Funding and Spending CriMNet financial data are spread among various state entities and local units of government, and we had a great deal of difficulty collating data into a comprehensive picture of CriMNet funding and spending to date. In addition, planning shortfalls have made it difficult to realistically forecast investments needed in the next biennium. The CriMNet Office in 2003 started compiling high-level data from the courts and departments of Corrections and Public Safety to develop estimates for 2002-03 program expenditures and a 2004-05 program budget. But, CriMNet funding and spending should be monitored in more detail, with budget detail aligned with major project phases and products. Our office's CriMNet financial audit report reached a similar conclusion, finding that CriMNet has not been established in the state's accounting system in a way that allows agencies and project managers to have reliable accounting information to monitor, analyze, and control program costs. 9

A related financial issue that needs to be resolved is how integration costs will be shared by the state and local entities. This means aligning local integration grant criteria with cost-sharing principles and with the CriMNet integration strategy. In the past, the Policy Group approved grants to counties for acquiring or enhancing county systems and required counties to provide matching funds equal to at least

78CRIMNET Table 4.5: Progress Establishing Program Controls, as of January 2004 Action Needed Status Accomplishments Work Remaining Hold Regular Steering Committee Meetings Implemented Task Force meets monthly and the Policy Group quarterly. Increase frequency of meetings and adjust agendas as needed to ensure timely decisionmaking and thorough review of program status on a regular basis. Report on Program Status Implemented CriMNet Office started issuing regular monthly status reports in levels. Draft action plan through spring 2004 prepared by Department of Public Safety.

June 2003. Strengthen the quality of budget data and add high-level status of state and locally-led integration projects. Monitor Spending Partially Implemented CriMNet Office started compiling programwide spending data on a snapshot basis. Develop a more consistent program accounting structure among agencies receiving CriMNet funds. Manage Communication Partially Implemented Active steps among CriMNet officials to improve communication at all percent of the grant amount. There are nuances to the local funding issue that the Policy Group is considering and that may require legislative action. They include: (1) the extent to which counties can use "in kind" matching funds (such as staff time) to meet the local match requirement and (2) the extent to which some smaller counties will be able to meet a 50-percent match requirement at all. The law requires that local grant criteria be consistent for all counties, but adhering to this requirement may affect the pace of integration to the extent that some local jurisdictions are unwilling or unable to meet the standard grant criteria. 10

 

Complete a full communication plan, including procedure for review of messages to external audiences, and obtain Policy Group approval.

Implement Change Control, Risk Management, and Issue Management Processes

Partially Implemented Change control implemented for strategic plan. Risk and issue management procedures defined in planning documents. Move these procedures from plans to active use in managing the program.

SOURCE: Office of the Legislative Auditor analysis of CriMNet Office documents and interviews with CriMNet officials.

Changes in the way CriMNet funding and spending are tracked in the state's accounting system would improve program oversight.

9Minnesota Office of the Legislative Auditor, CriMNet Financial Audit( St. Paul, 2004).

 

Communication In the past year, CriMNet officials have made a concerted effort to refocus their communication strategy to emphasize improving the quality and consistency of communication with project teams and local stakeholders. For example, making the Task Force a more integral element of CriMNet operations along with regular status presentations from the CriMNet Office to the Task Force and Policy Group are important steps toward improving communication among CriMNet stakeholders.

Through Public Safety's communications director, the CriMNet Office drafted a "Communications Action Plan" covering September 2003 through the spring of 2004 that states as its goals: strengthening CriMNet's asset base, educating key stakeholders, institutionalizing CriMNet as an operational system, and leveraging accomplishments. The action plan, however, does not establish routine mechanisms for communicating with stakeholders and criminal justice entities at all levels or include an explicit policy for formally reviewing and approving major communication products (such as newsletters). Also, the action plan shows a private sector company as a CriMNet spokesperson for many of the plan's action items. 11 Given past concerns regarding the accuracy and extent of communication about CriMNet, we think it is important for the Policy Group to engage in a discussion regarding CriMNet's communication strategy and to document that strategy as programwide guidance.

Other Program Controls Last, CriMNet Office staff have made some progress in designing and implementing project management processes for (1) reviewing and approving changes to time, scope, and budget; (2) identifying and mitigating project risks; and (3) identifying and resolving business or technical implementation issues that cannot be resolved at the project team level. The office has implemented change control techniques, starting with the strategic plan approved in September 2003. We assume that other key documents and program decisions will fall under the same process, which collectively will provide the CriMNet program with an authoritative set of documents and information that will serve as a baseline against which to measure progress. To date, procedures to identify and manage risks and implementation issues are being planned but have not been uniformly implemented.

PROGRAM MANAGEMENT 79 The CriMNet Office has made a concerted effort to improve communication with local governments and other stakeholders.

10As of January 2004, CriMNet officials were revising the criteria that will be used to award local grants (and the application process itself) and said that the grant criteria would focus on integration investments rather than enhancing local systems.

11In 2003, this private company produced and distributed a video about CriMNet and the need for criminal justice information integration.

 

Priorities f o r 2004 The CriMNet Office is poised to shift its focus from planning and capacity building to active implementation of projects and activities that directly further statewide criminal justice information integration. A number of key activities will be pivotal in 2004. While not an exhaustive list, the priorities listed in Table 4.6 include outcomes that need to be achieved to support integration. Legislative policy and appropriation committees should look for progress in these areas.

Meeting these priorities will require action by the CriMNet Office, Policy Group, and Legislature, and all require sustained leadership and resources. Legislative action in the 2004 and 2005 sessions to resolve key data practice issues is particularly critical. As we discussed in Chapters 2 and 3, uncertainty regarding the treatment of certain data has stalled progress on some integration efforts and will likely continue to do so.

80 CRIMNET

Table 4.6: CriMNet Priorities for 2004 Program Management Hire CriMNet Office and Integration Backbone project staff in accordance with the current staffing plan.

Define and implement a CriMNet Office scope statement and a work plan showing how the office will support work on the integration blueprint, data practices, user requirements, and security. Integration Successfully roll out the Integration Backbone search function. Attain legislative action on data practice questions presented to the Legislature in the

2004 session; develop more specific data practice recommendations for action in the 2005 legislative session. Assess and document user needs and make material progress translating these needs into integration requirements. Make material progress developing the integration blueprint— the collection of integration practices, standards, requirements, and plans that together describe how further integration will proceed.

SOURCE: Office of the Legislative Auditor analysis of CriMNet Office documents and interviews with CriMNet officials.

Getting staff on board, assessing user needs, and completing key planning tasks are priorities for 2004.

CONCLUSIONS Although various integration projects made notable progress, 2002 was a difficult year for CriMNet from a program management perspective. Inattention to program management basics, insuff icient staffing, and other issues spiraled into an unfortunate set of interrelated problems. These problems included eroding trust among various stakeholders, little progress on important program management projects, and confusion regarding CriMNet's direction and integration priorities. Absent institutionalized use of standard practices, the Policy Group lacked some important tools, such as a CriMNet Office work plan, that might have helped it pinpoint and address these problems earlier. Implementing standard practices does not, by itself, guarantee that CriMNet will be managed eff iciently and effectively. But, the bumps and missteps that CriMNet has experienced thus far demonstrate that more rigorous application of standard methods is needed.

The Policy Group and CriMNet Office staff took important steps in 2003 toward getting CriMNet's program management house in order. Appointing a new Executive Director and implementing a structured program management approach were key to establishing a more disciplined tone for CriMNet Office activities. In addition, Policy Group and Task Force efforts to address some long-standing issues, such as data practices, were equally important. Through sustained leadership, the Policy Group and Commissioner of Public Safety need to ensure that staffing problems are rectified and that basic planning tasks are completed. Action in both areas is needed to establish realistic baseline expectations for time, scope, and budget against which CriMNet progress can be measured.

In sum, CriMNet is at a critical juncture. Although some legislators and others may be disappointed with the extent of integration that has been achieved thus far, the problems driving the need for integration will not go away. Individuals who commit crimes cross jurisdictions, and criminal justice professionals must be able, on a statewide basis, to identify offenders and know their full criminal histories. The state, as a whole, needs to put forth its best effort to solve this problem. The integration process needs to take place through incremental steps, and both policymakers and the various state and local agencies involved need to set aside parochial interests and take responsibility for developing a shared sense of direction and setting priorities. These priorities should reflect the information sharing capabilities that front-line professionals need to do their jobs better. Moving forward, the Legislature should look to the Policy Group for a plan that clearly identifies the next steps.

RECOMMENDATIONS Fully Staff the CriMNet Office RECOMMENDATION To provide appropriate staffing for the CriMNet Office, the Commissioner of Public Safety should expedite filling open CriMNet Office staff positions.

Understaffing at the CriMNet Office is a long-standing problem that needs to be addressed immediately. The CriMNet Office has initiated the personnel actions required to hire staff, but six months after the Policy Group approved the staffing plan, few of the office's staff positions had been filled. We realize that state procedures constrain the speed with which new employees may be hired, but we urge the Commissioner of Public Safety, with support from other Policy Group members as needed, to expedite the hiring process. The CriMNet Office's ability to make material progress on its priorities for 2004 requires that it have sufficient staff.

Complete the Planning Process RECOMMENDATIONS To fully articulate the state's incremental approach to integrating criminal justice information, the Policy Group should:

Ensure that the CriMNet scope statement outlining projects and program controls is completed and approved as soon as possible;

Ensure that the agreed-upon scope of CriMNet projects in 2004 and 2005 adequately addresses the need to complete work practice, data, and technical requirements as quickly as possible;

Require the CriMNet Office to provide a comprehensive work plan showing time, budget, and outcome milestones for key activities and projects; and

Initiate, as soon as practical, the process of setting integration priorities for 2006-07.

CriMNet Office staff worked hard in 2003 to rectify past planning shortfalls. Completing the strategic plan was an important step toward focusing CriMNet resources and creating a common understanding among CriMNet stakeholders regarding CriMNet's direction over the next several years. But, CriMNet operations should be guided by a narrower set of incremental outcomes to be achieved (for example, accomplishments expected in 6-to 12-month increments), and the Policy Group should ensure that the plan for these incremental steps is well laid out. As of January 2004, some of this detailed planning work was underway; the Policy Group needs to ensure that it is completed as quickly as possible.

82 CRIMNET The Policy Group needs to ensure that program planning is completed and performance benchmarks are set.

 

We think having a comprehensive plan showing how various CriMNet Office, state agency , and local projects fit together would also be useful from a management and oversight perspective. In addition, the Policy Group needs to present to the 2005 Legislature a CriMNet funding plan for 2006-07. To make a credible presentation, the plan should include an explanation of projects to be funded and how they are linked to the CriMNet strategic plan.

Improve Communication and Accountability RECOMMENDATIONS To improve the timeliness and quality of information on CriMNet program status:

The Policy Group should require state agencies implementing CriMNet projects to establish a common coding structure in the state's accounting system that will allow for overall analysis, monitoring, and reporting of CriMNet financial activity.

Policy Group members should ensure (1) that state agency CriMNet project teams report all needed status information, including budget, schedule, and outcome data, to the CriMNet Office and (2) that local grant contracts include the same requirement.

The Policy Group should implement a communication plan that (1) balances external publicity with information sharing and communication among CriMNet project teams and stakeholder groups; (2) defines the parameters of private sector involvement in promoting CriMNet; and (3) includes a requirement for review and approval of key messages and communication materials.

To provide more meaningful oversight, the Policy Group and Task Force, in collaboration with the CriMNet Office, should adjust meeting agendas and schedules as needed to ensure that CriMNet program status is regularly and rigorously reviewed.

To improve its ability to evaluate CriMNet budget requests, the Legislature should require the Policy Group to present during the biennial appropriation process a comprehensive spending plan showing the status of current projects, the amount of continued funding requested, high-level outcomes expected in the coming biennium, and priorities among new project proposals. In addition, legislative finance and policy committees of jurisdiction should consider specific agency criminal justice information system requests in light of this overall CriMNet plan.

PROGRAM MANAGEMENT 83 The Legislature should require more detailed and prioritized spending plans for CriMNet projects.

CriMNet is a high profile and important initiative for the state. Based on how difficult it was for agencies to compile CriMNet spending data for their own use and for our evaluation, the program would benefit from a clearer financial accounting structure. Creating a common structure will require cooperation among those agencies receiving CriMNet funds and the Department of Finance to (1) establish a consistent definition of the types of criminal justice information system spending that will be included in CriMNet, and (2) implement a coding structure to track CriMNet spending, including CriMNet costs funded with general operating or general information system funds.

If CriMNet is to function as a collaborative effort among jurisdictions, state and local agencies need to do their part by providing basic project planning and status information to the CriMNet Office where the information can be compiled for a programwide perspective. In addition, to ensure active evaluation of programwide progress against baseline expectations, the Task Force and Policy Group need to reach the program review portions of their respective meeting agendas. From mid-2003 to date, both groups devoted most of their meeting time to discussion of long-standing issues, such as data practices. We certainly do not dispute the importance of resolving these issues, but for accountability, adequate attention must also be paid to the progress of integration projects.

CriMNet's current management team emphasized to us that good communication among CriMNet stakeholders is an important management tool, and past experience has shown how poor communication can damage relationships and contribute to unrealistic expectations. We think it is important to set clear expectations for internal and external communication in a formal plan. The plan should be used as a means to explicitly articulate some important communication decisions, such as (1) how to educate the public about the vision for CriMNet while providing realistic expectations regarding incremental progress; (2) the role of private sector companies in serving as formal spokespersons for the state; and (3) how communication materials, such as speeches and newsletters, should be reviewed to ensure that the message fits with the communication plan.

Address Data Practice Issues RECOMMENDATION To address deficient and conflicting Minnesota statutes pertaining to criminal justice data practices and to ensure that individuals have an appropriate level of access to data about themselves:

The Legislature should, during the 2004 legislative session, consider and act on the Policy Group's December 2003 recommendations to modify the Data Practices Act; and

In future years, the Policy Group should ensure that it makes timely recommendations to the Legislature regarding additional data practice issues requiring legislative action.

84 CRIMNET The Legislature needs to act on Policy Group recommendations regarding changes to Minnesota's Data Practices Act.

In its December 2003 annual report to the Legislature, the Policy Group recommended a number of statutory changes to address conflicting data classifications and rules of access. The Legislature needs to consider and act on these and future recommendations to better ensure that, as CriMNet moves forward, it does so in compliance with the Data Practices Act. Uncertainties regarding compliance with the Data Practices Act have slowed some aspects of CriMNet system design and have affected some local jurisdictions' willingness to share data.

Modify CriMNet's Governance Structure RECOMMENDATIONS To better ensure that the perspectives of local jurisdictions are considered in setting CriMNet policy, the Legislature should amend the law to add local representatives to the Policy Group.

To provide stronger day-to-day support and oversight of CriMNet Office operations, the Commissioner of Public Safety and the Policy Group as a whole should clarify for the CriMNet Executive Director distinctions between the Policy Group's strategic authority and the Department of Public Safety's day-to-day management authority.

The Policy Group should evolve as CriMNet does. Several of the largest statewide integration projects managed as part of CriMNet— MNCIS, the Statewide Supervision System, and Integration Backbone— are well underway or completed. As CriMNet's focus turns toward integrating city and county jurisdictions, the Policy Group's membership should be rebalanced to reflect that shift. Adding local jurisdiction representatives to the Policy Group reinforces the notion that CriMNet is a collaborative effort among branches and levels of government. The Legislature can use this opportunity to consider adding criminal justice functions not currently represented on the Policy Group, including prosecution, public defense, and local law enforcement. It may make sense to choose these representatives from among Task Force members, as they are likely to be well versed in CriMNet issues.

In considering how local representatives may be added to the Policy Group, we suggest that the Legislature engage in a broader discussion regarding the number of executive and judicial branch representatives as well. For example, the Commissioner of Finance has requested that her department no longer serve on the Policy Group because of concern that the department has a role in both crafting CriMNet budget proposals and reviewing them. If the Legislature chooses to act on her recommendation, it may wish to reduce the number of judicial representatives from four to three as well. In general, we think this is a good time for legislators from the House and Senate committees of jurisdiction to meet with Policy Group members and local representatives from the Task Force to discuss how the Policy Group can be configured to best meet the program's needs moving forward.

PROGRAM MANAGEMENT 85 This is a good time for the Legislature to review CriMNet's governance structure.

We support the decision to more closely align CriMNet Office operations with management of other state criminal justice information systems by the Bureau of Criminal Apprehension. But, we recognize the complications that this dual reporting relationship can present for the CriMNet Executive Director. Our recommendation is intended to urge the Commissioner of Public Safety and other Policy Group members to proactively address and resolve any differences of opinion regarding the bounds of their authority.

 

Summary of Recommendations Strengthen CriMNet Project Management (p. 56) To help facilitate criminal justice integration and mitigate costs for the state, Policy Group members should ensure that CriMNet projects managed by their respective agencies:

Have documented baseline expectations regarding project schedules, budgets, and scope;

Have adequately identified and addressed prerequisite decisions regarding users' requirements, data practices, and other criminal justice practices; and

Coordinate and communicate with stakeholders on other CriMNet projects.

Fully Staff the CriMNet Office (p. 82) To provide appropriate staffing for the CriMNet Office, the Commissioner of Public Safety should expedite filling open CriMNet Office staff positions.

Complete the Planning Process (pp. 39, 82) The CriMNet Office, in coordination with local governments, should inventory local governments' criminal justice information systems and integration plans and estimate the resources that will be required to bring needed information into CriMNet.

To fully articulate the state's incremental approach to integrating criminal justice information, the Policy Group should:

Ensure that the CriMNet scope statement outlining projects and program controls is completed and approved as soon as possible;

Ensure that the agreed-upon scope of CriMNet projects in 2004 and 2005 adequately addresses the need to complete work practice, data, and technical requirements as quickly as possible;

Require the CriMNet Office to provide a comprehensive work plan showing time, budget, and outcome milestones for key activities and projects; and

Initiate, as soon as practical, the process of setting integration priorities for 2006-07.

Improve Communication and Accountability (p. 83) To improve the timeliness and quality of information on CriMNet program status: The Policy Group should require state agencies implementing CriMNet projects to establish a common coding structure in the state's accounting system that will allow for overall analysis, monitoring, and reporting of CriMNet financial activity.

Policy Group members should ensure (1) that state agency CriMNet project teams report all needed status information, including budget, schedule, and outcome data, to the CriMNet Office and (2) that local grant contracts include the same requirement.

The Policy Group should implement a communication plan that (1) balances external publicity with information sharing and communication among CriMNet project teams and stakeholder groups; (2) defines the parameters of private sector involvement in promoting CriMNet; and (3) includes a requirement for review and approval of key messages and communication materials.

To provide more meaningful oversight: The Policy Group and Task Force, in collaboration with the CriMNet Office, should adjust meeting agendas and schedules as needed to ensure that CriMNet program status is regularly and rigorously reviewed.

To improve its ability to evaluate CriMNet budget requests: The Legislature should require the Policy Group to present during the biennial appropriation process a comprehensive spending plan showing the status of current projects, the amount of continued funding requested, high-level outcomes expected in the coming biennium, and priorities among new project proposals. In addition, legislative finance and policy committees of jurisdiction should consider specific agency criminal justice information system requests in light of this overall CriMNet plan.

Address Data Practice Issues (p. 84) To address deficient and conflicting Minnesota statutes pertaining to criminal justice data practices and to ensure that individuals have an appropriate level of access to data about themselves:

The Legislature should, during the 2004 legislative session, consider and act on the Policy Group's December 2003 recommendations to modify the Data Practices Act; and

In future years, the Policy Group should ensure that it makes timely recommendations to the Legislature regarding additional data practice issues requiring legislative action.

 

Modify CriMNet's Governance Structure (p. 85) To better ensure that the perspectives of local jurisdictions are considered in setting CriMNet policy, the Legislature should amend the law to add local representatives to the Policy Group.

To provide stronger day-to-day support and oversight of CriMNet Office operations, the Commissioner of Public Safety and the Policy Group as a whole should clarify for the CriMNet Executive Director distinctions between the Policy Group's strategic authority and the Department of Public Safety's day-to-day management authority.

SUMMARY OF RECOMMENDATIONS Further Reading Aeritae Consulting Group, Ltd. CriMNet Risk Assessment and Mitigation. Roseville, MN: Aeritae Consulting, 2003.

The Markle Foundation. Creating a Trusted Information Network for Homeland Security. New York: The Markle Foundation, 2003.

Minnesota Criminal and Juvenile Justice Information Policy Group. CriMNet Strategic Plan. St. Paul, 2003.

Minnesota Office of the Legislative Auditor. CriMNet Financial Audit. St. Paul, 2004.

Morton, Heather. Integrated Criminal Justice Information Systems. Denver, CO: National Conference of State Legislatures, 2001; http://www.ncsl.org/Home/tabid/118/Default.aspx?programs/lis/intjust/report01.htm.

National Association of State Chief Information Officers. Concept of Operations for Integrated Justice Information Sharing, Version 1.0. Lexington, KY: NASCIO, 2003.

SEARCH. Integration in the Context of Justice Information Systems: A Common Understanding. Sacramento, CA: SEARCH/ The National Consortium for Justice Information and Statistics, 2001.

Planning the Integration of Justice Information Systems: Developing the Justice Information Exchange Model. Sacramento, CA: SEARCH/ The National Consortium for Justice Information and Statistics, 2002. Office of the Commissioner 445 Minnesota Street, Suite 1000, North Central Life Tower, St. Paul, Minnesota 55101-5000 Phone: 651/ 296-6642 Fax: 651.297.5728 TTY: 651/ 282-6555 Internet: http://www.dps.state.mn.us

February 20, 2004 James R. Nobles Legislative Auditor Centennial Building 658 Cedar Street St. Paul, MN 55155

Dear Mr. Nobles: On behalf of the Department of Public Safety, I want to thank you and your staff for all your efforts to produce such a thorough and comprehensive report on such a complex program as CriMNet and related criminal justice projects. We feel the report represents a fair and accurate assessment of the criminal justice projects and the CriMNet program .

Your staff did an excellent job of crafting this report and identifying the issues that agencies face when developing complex, multi-disciplined projects. We recognize that many of the early projects, while delivering additional information and infrastructure, did not address integration. That is precisely the reason the legislature originally adopted the CriMNet plan in 2001. Since my appointment as Commissioner of Public Safety in February, 2003, we have made a concerted effort to focus our attention on many of the areas that were identified in the report. We acknowledge the issues that CriMNet faced in its early years but believe that significant progress has been made and that many of the issues are continuing to be addressed.

CriMNet is a program that requi res collaboration from stakeholders at all levels of government. Not everyone agrees what constitutes CriMNet or what should be included in the scope of CriMNet. At times, it is difficult to get consensus among all these stakeholders, but we are confident that CriMNet is now moving in the right direction. It is important to remember that no other state has achieved successful integration in this area. This is new territory; Minnesota is a pioneer state and is seen as a national leader in the area of criminal justice information integration. While there is no national model, there are a number of national groups and other states who are working diligently in this area and recognize that successful criminal justice information integration is critical to improving public safety . CriMNet is a major player in this national effort and will continue to participate at the national level with groups such as: Search, Global, National Governor's Association and National Association of Chief Inform ation Officers.

MINNESOTA DEPARTMENT OF PUBLIC SAFETY Alcohol and Gambling Enforcement Bureau of Criminal Apprehension Capital Security

Division of Homeland Security and

Emergency Management

Office of Justice Programs Driver and Vehicle Services

State Fire Marshal and Office of Pipeline Safety Minnesota State Patrol

Office of Traffic Safety

Your office identified a number of findings and recommendations for the CriMNet program. We generally agree with these findings a nd recommendations, and in many cases, have already taken action to rectify problem areas. One major conclusion of the report is that the CriMNet program is inadequately staffed. Signifi ant progress has recently been made in this area and a permanent staff should be in place in the very near future. Once that hurdle is behind us, CriMNet can focus on the additional recommendations that need to be addressed. The report included a list of summary recommendations at the end, and we would like to take the opportunity to respond to these:

1. Strengthen CriMNet Project Management. The CriMNet Office fully agrees with implementing program/project management best practices as defined by the state Office of Technology Project Management Office. Over the past year, much progress has been made to implement these guidelines and the CriMNet Office regularly consults with the Office of Technology on processes and related tools. The goal is for all projects to follow consistent guidelines for scope statements, phase reviews and decisions points. CriMNet has also implemented a consistent status reporting system for the program and associated projects to make sure that key stakeholders are kept informed of program/project status, issues and changes. We are currently tracking budgets at the CriMNet program level but recognize the need to address tracking budgets at the project level . 2. Fully Staff the CriMNet Office. Over the last three years, the CriMNet Office has not produced the desired results and outcomes largely in part due to insufficient staffing. Currently, CriMNet is comprised of a number of temporary Rule 10 positions and a number of contractors that have worked on different projects at various times. The program has recently hired five permanent staff, including the executive director, and is in the process of hiring an additional 21 permanent positions . The majority of these positions are technical positions and business analyst positions. We feel very positive about the difference having a full-time, permanent staff will make to strengthen the future of the program.

3. Complete the Planning Process. The major hurdle to complete the planning process has been the lack of a defined scope statement. The CriMNet Office currently has a stakeholder delivery team that is working on the scope statement and should have a draft available for the Task Force and Policy Group to review in early 2004. And, for the past nine months, the program has been operating under a work plan and will cont inue to do so. The work plan will coincide with the approved scope statement and will include current projects and key milestones . A long-term strategic plan will be developed once the scope statement is complete and will cover a three to five year period.

Improve Communication and Accountability. We feel the utilization of the status reporting system, as described in recommendation number one, will dramatically improve both communication and accountability to the Task Force, Policy Group and other stakeholders. A formal communication plan, which includes communication with not only stake holders, but with the legislature and the public, is currently being developed. 5. Address Data Practice Issues. A delivery team, made up of a number of interested parties and stakeholders, has been meeting for the past year and a half to discuss and seek some resolution on data practice issues. With a strong majority, the group was able to reach some consensus and made a number of recommendations to the Task Force, Policy Group and ultimately the legislature. A bill has be en introduced in the House and Senate, which addresses the initial issues relating to data practices as CriMNet moves forward. There is still much work to be done in this area, and this delivery team plans to continue to meet regularly and plan for the next steps. 6. Modify CriMNet's Governance Structure. The Departme nt of Public Safety and the CriMNet Office appreciate the input and recommendations offered by the Legislative Auditor but would defer to the Task Force and Policy Group to address these recommendations . The Policy Group is looking forward to the opportunity to provide a formal response and will be convening in early March to discuss your findings. Again, I would like to express my appreciation on behalf of the depart ment and the CriMNet Office for the efforts of the auditor's staff to produce an objective look at the CriMNet program . The road has not been easy over the past three years, but CriMNet is alive and well, and we feel very confident with the direction CriMNet is now moving. The people involved with CriMNet are dedicated and committed people who believe in the value of criminal justice information integration and what that means to the safety of citizens across the state of Minnesota. We look forward to a continued positive working relationship with the legislature to ensure the success of CriMNet.

Sincerely, /s/ Rich Stanek Rich Stanek Commissioner
State of Minnesota Minnes ot a Department of Corrections Office of the Commissioner

February 18, 2004 James R. Nobles Office of the Legislative Auditor 658 Cedar, Room 140 St. Paul, MN 55155

Dear Mr. Nobles: The Department of Corrections would like to thank you for the opportunity to review the final draft of the CriMNet Program Evaluation. We understand that this has been a sizeable endeavor involving multiple agen cies and commend you and your staff for their work. Your staff has been receptive to our concerns and we have appreciated the opportunity to share our efforts in successfu lly implementing the Statewide Supervision System.

As you indicated, Minnesota has indeed made significant progress in integrating parts of its criminal justice system. The Statewide S upervision System is one example noted as a key improvement by providing probation and dete ntion data previously available only at each individual agency.

In addition, as mentioned, the experience of the Department of Corrections in working with local agencies and various vendors can be of significant value for future CriMNet efforts in addressing remaining gaps in Minnesota's integration efforts. We would like to thank Deborah Junod as proj ec t manager and her staff for their work on this report. We look forward to continued coll aboration with our state and local partners.

Sincerely, /s/ Joan Fabian Joan Fabian Commissioner

1 450 Energy Park Drive, Suite 200 St. Paul, Minnesota 55108 -5219 Phone 651/ 642 -028 2 Fax 651/ 642 -0 414 TDD 651/ 643 -3589 An Equal opportunity employer
THE SUPREME COURT OF MINNESOTA MINNESOTA JUDICIAL CENTER 25 REV. DR. MARTIN LUTHER KING JR. BLVD. SAINT PAUL, MINNESOTA 55155

SUE K. DOSAL (651) 296-2474 STATE COURT ADMINISTRATOR Fax (651) 215-6004 E-mail: sue.dosal@courts.state.mn.us

February 20, 2004 Mr. Jim Nobles Office of the Legislative Auditor Centennial Building 658 Cedar St St. Paul, MN 55155

Dear Mr. Nobles: Thank you for the opportunity to review the Legislative Auditor's program evaluation of the CriMNet Project. We acknowledge the professional work of the staff in preparing the report and identifying recommendations for program improvement. As you know, the Policy Group, which represents more than a single agency perspective, did not have an opportunity to consider the report or its recommendations in advance. The recommendations identified by the report are worthy of careful consideration by the full Policy Group, which will next meet on March 8, 2004.

In general, the report recognizes the complexity of the CriMNet project, the diversity of perspectives and components, the significant work that has been accomplished to date, and the commitment of the organizations involved to work toward a satisfactory solution. You have the Judiciary's commitment to carefully consider the report and to work expeditiously through the Policy Group process toward resolution of the issues that you have identified. I expect that a formal response will be provided by the Policy Group after it has had the opportunity to meet as a body and review this report as a body.

Sincerely yours, /s/ Sue K. Dosal Sue K. Dosal State Court Administrator

April 6, 2004

Mr. James Nobles
Legislative Auditor
Centennial Building
658 Cedar Street
St. Paul, MN 55155
Dear Mr. Nobles:

On behalf of the Criminal and Juvenile Justice Policy Group (Policy Group), this letter is a follow-up response to the letter we sent to you on March 10, 2004. Since then, the Policy Group met on March 17, 2004 to hear personally from you and your staff regarding the findings and recommendations from the report as they specifically related to the Policy Group, and we met again at our regularly scheduled Policy Group meeting on March 24, 2004 to discuss the actions that will be taken. We appreciate the input and comments from your office and thought the discussion brought about some thoughtful dialogue.

While our discussions as a Policy Group have only begun, we did want to take the opportunity to outline the recommendations submitted in the report and provide a response to them.

Improve Communications and Accountability

Improve timeliness and quality of information on CriMNet program status:

The Policy Group should develop a policy about CriMNet project costs funded with general operating funds or with general information system funds, and direct agencies to identify those costs in the accounting system or other records in a way that allows them to be included in the overall analysis, monitoring, and reporting of CriMNet program financial activity.

The Policy Group should work with the agencies to identify resources currently available for the CriMNet program and to define which projects and costs are CriMNet program costs.

The Policy Group should require state agencies implementing CriMNet projects to establish a common coding structure in the state’s accounting system that will allow for overall analysis, monitoring, and reporting of CriMNet financial activity.

Policy Group members should ensure (1) that state agency CriMNet project teams report all needed status information, including budget, schedule, and outcome data, to the CriMNet Office and (2) that local grant contracts include the same requirement.

The Policy Group has not come to any resolution regarding these four recommendations. They are all related to the definition of CriMNet, CriMNet projects and agency criminal justice projects and initiatives. Each agency has dealt with their accounting and reporting requirements according to their internal agency requirements. The Policy Group has agreed that a working group made up of the Department of Finance, finance representatives from the major criminal justice agencies and the appropriate systems’ business representatives from those agencies would begin to meet and develop options for recording criminal justice expenditures. There is still an issue of what a CriMNet project is and what an agency project is. The Policy Group, Task Force and CriMNet Office are working on a scope statement that will address some of theses issues. Once that has been determined, decisions can be made regarding accounting and status reporting practices. The Policy Group expects that thoughtful discussion will continue on these recommendations. Over the next few months, the Policy Group plans on holding several working sessions focused on specific topics.
Scope Statement April – May
Governance/Roles & Responsibilities May - June

The Policy Group should implement a communications plan that (1) balances external publicity with information sharing and communication among CriMNet project teams and stakeholder groups; (2) defines the parameters of private sector involvement in promoting CriMNet; and (3) includes a requirement for review and approval of key messages and communication materials.

The Policy Group agrees that communication at all levels is a major issue that needs to be addressed. Effective internal communication between the CriMNet Program Office, the Task Force and the Policy Group is an immediate need that has been lacking since CriMNet’s inception. The CriMNet Office has implemented a consistent status reporting system for the program and associated projects to make sure that key stakeholders (primarily the Task Force and Policy Group) are kept informed of program/project status, issues and changes. CriMNet is also in the process of hiring a new communications position. The focus of that position will be gathering input from CriMNet stakeholders and building consensus regarding what CriMNet’s message will be, what the best way is to communicate that message, and how the Task Force and Policy Group will provide the necessary review and approval of key messages and communication materials. The goal is to develop an effective process for internal communication and an overall, formal communication plan. One of the first tasks for the new communications position will be to develop a quarterly newsletter aimed at the Policy Group, Task Force and stakeholders. The purpose of the newsletter will be to highlight all of the key initiatives within criminal justice agencies and CriMNet.


Provide more meaningful oversight:

The Policy Group and Task Force, in collaboration with the CriMNet Office, should adjust meeting agendas and schedules as needed to ensure that CriMNet program status is regularly and rigorously reviewed.

As stated above, in the last year, the CriMNet Office has made a diligent effort to make sure that Task Force and Policy Group members receive a written status report of the CriMNet program at their monthly/quarterly meetings. This status report includes status on budget, schedule and scope and details if those areas are green, yellow or red (controlled, caution, critical) and what the next steps are to rectify problem areas identified. On a fairly regular basis, there are so many agenda items that need to be covered at both the Policy Group and Task Force meetings that the status report is not given adequate time to be discussed; however, members always receive a written report that they are able to review. The utilization of this status reporting system should dramatically improve both communication and accountability to the Policy Group, the Task Force and other stakeholders. The Policy Group has changed the agenda order to ensure that program and project status updates are the first order of business to be reviewed. The Task Force will review the order of agenda items to determine if any changes are required. There are times when it has been determined that the status reports do not have to be reviewed and that other items are of a higher priority.

Improve its ability to evaluate CriMNet budget requests:

The Legislature should require the Policy Group to present during the biennial appropriation process a comprehensive spending plan showing the status of current projects, the amount of continued funding requested, high-level outcomes expected in the coming biennium, and priorities among new project proposals. In addition, legislative finance and policy committees of jurisdiction should consider specific agency criminal justice information systems requests in light of this overall CriMNet plan.

The Policy Group agrees that CriMNet should participate in the legislative biennial appropriation process as long as the program continues to receive state funds. The Courts, the Department of Corrections and the Department of Public Safety may request appropriations for CriMNet-related activities for each of their agencies within their own base budget requests, but the Policy Group will provide an overall spending and project plan for CriMNet for the fiscal year 2006/2007 process.

Address Data Practice Issues

To address deficient and conflicting Minnesota statutes pertaining to criminal justice data practices and to ensure that individuals have an appropriate level of access to data about themselves:

The Legislature should during the 2004 legislative session, consider and act on the Policy Group’s December recommendation to modify the Data Practices Act.

The Policy Group did approve language to revise the data privacy statutes as they relate to CriMNet and forwarded those recommendations to the legislature. This CriMNet data privacy bill was introduced in the House and Senate. We will know more by the end of the legislative session as to how the legislature would like to see CriMNet proceed with related data practices issues. The Policy Group remains committed to addressing data practices issues as they arise and fully supports the Task Force Data Practices Delivery Team as they continue to study and make recommendations on data practices.

In future years, the Policy Group should ensure that it makes timely recommendations to the Legislature regarding additional data practices issues requiring legislative action.

The legislative language approved by the Policy Group this year includes a requirement that the Task Force report to the Policy Group by December 1st on any recommendations that need to be considered for future legislative action.

Modify CriMNet’s Governance Structure

To better ensure that the perspectives of local jurisdictions are considered in setting CriMNet policy, the Legislature should amend the law to add local representatives to the Policy Group.

The Policy Group has discussed the merits of this issue on more than one occasion. There are Policy Group members who see value to adding local representation to the Policy Group. Particularly the Chairs of the Task Force, who currently serve as non-voting members on the Policy Group, have expressed their endorsement of adding Task Force representation to the Policy Group. However, there are also concerns expressed by Policy Group members to add local representation. One concern is that Task Force members on the Policy Group will be voting on their own recommendations made to the Policy Group. Another concern is that local representatives will be making decisions regarding state funding that will impact them as locals. The Policy Group expects that thoughtful discussion will continue on this recommendation. As discussed earlier, the Policy Group expects to hold a series of working meetings in the May – June timeframe to address governance and roles and responsibilities.

To provide stronger day-to-day support and oversight of CriMNet Office operations, the Commissioner of Public Safety and the Policy Group as a whole should clarify for the CriMNet Executive Director distinctions between the Policy Group’s strategic authority and the Department of Public Safety’s day-to-day management authority.

In November 2003, the CriMNet Office moved its operations to the new Bureau of Criminal Apprehension (BCA) building under the Department of Public Safety (DPS). DPS has given CriMNet the operational support needed as well as the perception that CriMNet is accountable as part of a state agency to improve the credibility CriMNet was lacking. The Policy Group still maintains the strategic oversight of CriMNet and is involved in major decisions that affect the program. The Policy Group will work with the Executive Director to clarify roles and authority by distinguishing strategic oversight decisions from day-to-day management decisions. This should be an outcome from the discussion on roles and responsibilities that the Policy Group plans to hold.

Strengthen CriMNet Project Management

To help facilitate criminal justice integration and mitigate costs for the state, Policy Group members should ensure that CriMNet projects managed by their respective agencies:

Have documented baseline expectations regarding project schedules, budgets, and scope.

The Policy Group agrees that baseline expectations regarding these items have been inadequate in the past. The Policy Group does support the steps the CriMNet Office has taken to implement program/project management best practices as defined by the Office of Technology Project Management Office. Over the past year, much progress has been made to implement these guidelines and the CriMNet Office regularly consults with the Office of Technology on processes and related tools. The goal is for all projects to follow consistent guidelines for scope statements, phase reviews and decision points. The CriMNet Office currently tracks budgets at the program level but recognizes the need to track budgets at the project level. To that end, the CriMNet Office will work with the Policy Group and the Task Force to develop project level budgets and reports. The CriMNet Office is committed to having specific project budget information available by the June 2004 Policy Group meeting.

Have adequately identified and addressed prerequisite decisions regarding users’ requirements, data practices, and other criminal justice practices.

As stated above, data practices are key issues for the implementation of CriMNet. The Task Force has a delivery team, which developed the 2004 legislative recommendations, that is continuing to review and analyze data practice requirements for integration and information sharing. This group will be developing its recommendations by December 1, 2004. That delivery team is specifically tasked to address: (1) providing web-access to CriMNet data by data subjects; (2) use of CriMNet data for non-criminal justice purposes; (3) advisability of public access;
(4) standards for dissemination of CriMNet data to entities that are not subject to chapter 13; (5) effects of federal requirements on the rights of individuals under chapter 13; (6) implementing the Minnesota Government Data Practices Act and court rules of access requirements regarding disclosure of disputed data held by CriMNet. The CriMNet Office recognizes the need to work on user requirements and other criminal justice practices. The business requirements should be the key drivers to future CriMNet initiatives. One recently hired business analyst has already begun meeting with counties to develop user requirements. Another project will be started in April that focuses on developing the standards for data and document exchanges. A detailed work plan addressing these areas and several more is under development and is expected to be completed within the next of couple of months as staff is hired.

Coordinate and communicate with stakeholders on other CriMNet projects.

As stated previously, the CriMNet Office has developed a consistent status reporting system to improve communication with both the Task Force and Policy Group. However, the Policy Group agrees that CriMNet could improve its communication regarding all current projects. The Policy Group, as well as other stakeholders, has expressed frustration that there is not a comprehensive list of current CriMNet projects and their status. Once the scope of CriMNet is defined, there should be more clarity of which projects fall within CriMNet’s scope. As discussed under the communications section, a quarterly newsletter will be developed as one of the methods of communications. Another is to post all Policy Group and Task Force meeting information on a web site accessible to stakeholders.

Fully Staff the CriMNet Office

To provide appropriate staffing for the CriMNet Office, the Commissioner of Public Safety should expedite filling open CriMNet staff positions.

The Policy Group approved an organizational chart of staffing positions in June of 2003. Since that time, the current staff has worked on drafting position descriptions and vetting them through the state human resources process. To date, five permanent positions have been filled, 17 positions are beginning the interview process and four positions are still being created or audited. Many of CriMNet’s difficulties have stemmed from the lack of staff to get the work done. The Policy Group recognizes the need to expedite the staffing process and supports the efforts of DPS Human Resources and the CriMNet Office to complete the current organizational structure. It is expected that the majority of the positions will be filled by mid-May 2004.

Complete the Planning Process

The CriMNet Office, in coordination with local governments, should inventory local governments’ criminal justice information systems and integration plans and estimate the resources that will be required to bring needed information into CriMNet.

The Policy Group recently approved a resolution to support the development of a state and local user implementation plan. The resolution requested that the Task Force work on recommendations to identify a plan, timetable and funding mechanism to achieve statewide state and local participation in CriMNet. The plan will assess local governments’ criminal justice information systems and provide the cost estimates to achieve integration. The input for the implementation plan will consist of the technical analysis of state and local agencies capabilities and data systems, business requirements analysis of agency needs, gap analysis regarding information, business process redesign requirements, standards development for connectivity, event and exchange points standards, and document standards for those events and exchanges. A detailed work plan addressing these areas is under development and is expected to be completed within the next couple of months as staff is hired.

To fully articulate the state’s incremental approach to integrating criminal justice information, the Policy Group should:

Ensure that the CriMNet scope statement outlining projects and program controls is completed and approved as soon as possible.

The Policy Group recognizes that the scope statement is the fundamental document to guide CriMNet as the program moves forward; therefore, finalizing the scope statement is the number one priority for the Policy Group. A delivery team has been working on a draft scope document for the Task Force to review at its April 2004 meeting. Once the Task Force reaches consensus, the Policy Group will convene as a working group to review and finalize the scope statement. This is expected to be done in the April – May timeframe depending on legislative activities and the approval of the scope statement by the Task Force.

Ensure that the agreed-upon scope of CriMNet projects in 2004 and 2005 adequately address the need to complete work practice, data, and technical requirements as quickly as possible.

The business and technical practices piece of the program is critical to the success of CriMNet. Until 2004, there has not been staff specifically assigned to business and technical processes and standards. As part of the new organizational structure, the CriMNet Office has hired one business analyst and is in the process of hiring an additional six business and technical positions that will primarily focus on these requirements.

Require the CriMNet Office to provide a comprehensive work plan showing time, budget, and outcome milestones for key activities and projects.

The status reports referred to previously do provide milestones and deliverables for some of the key activities and projects; however, the Policy Group agrees that there is some confusion as to what is included in the overall scope of CriMNet projects and activities. Policy Group members have expressed a lack of clarity of what CriMNet is currently involved in due to the lack of a comprehensive work plan. The CriMNet Office is confident that these issues will be resolved once the scope statement is approved and finalized. A work plan is under development and as staff are hired, more definitive timelines should be available. For existing initiatives, CriMNet will begin to develop and track the project budget components and expects to provide the Policy Group this information at the regularly scheduled June Policy Group meeting

Initiate, as soon as practical, the process of setting integration priorities for 2006/2007.

Again, once the scope statement is finalized, a comprehensive work plan can be completed and the priorities established for the next biennium. As work is performed to develop the implementation plan, it is expected that various initiatives will be identified. CriMNet expects to complete the comprehensive Implementation Plan by December 2005.

As Chair of the Policy Group, you have my commitment that the Policy Group will continue our discussions on these findings and recommendations and will work to resolve these issues. I would like to take the opportunity again to thank you and your staff for all your efforts, to not only write a fair and accurate report, but to provide sincere and insightful recommendations that will help us, as policymakers, to guide CriMNet and ensure CriMNet’s success.

Sincerely,

/s/ Rich Stanek

Rich Stanek, Chair
On behalf of the Criminal and Juvenile Justice Policy Group

cc: Policy Group Members
Task Force Members
Governor Tim Pawlenty
Chief Justice Kathleen Blatz
Senator Jane Ranum
Senator Leo Foley
Representative Steve Smith

Recent Program Evaluations
Residential Facilities for Juvenile Offenders, February 1995 95-01
Health Care Administrative Costs, February 1995 95-02
Guardians Ad Litem, February 1995 95-03 Early Retirement Incentives, March 1995 95-04
State Employee Training: A Best Practices Review, April 1995 95-05
Snow and Ice Control: A Best Practices Review, May 1995 95-06
Pollution Control Agency's Use of Administrative Penalty Orders, Update July 1995 95-07
Development and Use of the 1994 Agency Performance Reports, July 1995 PR95-22
State Agency Use of Customer Satisfaction Surveys, October 1995 PR95-23
Funding for Probation Services, January 1996 96-01 Department of Human Rights, January 1996 96-02
Trends in State and Local Government Spending, February 1996 96-03
State Grant and Loan Programs for Businesses February 1996 96-04
Post-Secondary Enrollment Options Program, March 1996 96-05
Tax Increment Financing, March 1996 96-06 Property Assessments: Structure and Appeals,
A Best Practices Review, May 1996 96-07 Recidivism of Adult Felons, January 1997 97-01
Nursing Home Rates in the Upper Midwest, January 1997 97-02
Special Education, January 1997 97-03 Ethanol Programs, February 1997 97-04
Statewide Systems Project, February 1997 97-05 Highway Spending, March 1997 97-06
Non-Felony Prosecution, A Best Practices Review, April 1997 97-07
Social Service Mandates Reform, July 1997 97-08 Child Protective Services, January 1998 98-01
Remedial Education, January 1998 98-02 Transit Services, February 1998 98-03
State Building Maintenance, February 1998 98-04 School Trust Land, March 1998 98-05
9-1-1 Dispatching: A Best Practices Review, March 1998 98-06
Minnesota State High School League, June 1998 98-07
State Building Code, January 1999 99-01 Juvenile Out-of-Home Placement, January 1999 99-02
Metropolitan Mosquito Control District, January 1999 99-03
Animal Feedlot Regulation, January 1999 99-04 Occupational Regulation, February 1999 99-05
Directory of Regulated Occupations in Minnesota, February 1999 99-05b
Counties' Use of Administrative Penalties for Violations of Solid and Hazardous
Waste Ordinances, February 1999 99-06 Fire Services: A Best Practices
Review, April 1999 99-07 State Mandates on Local Governments,
January 2000 00-01 State Park Management, January 2000 00-02
Welfare Reform, January 2000 00-03 School District Finances, February 2000 00-04
State Employee Compensation, February 2000 00-05 Preventive Maintenance for Local Government
Buildings: A Best Practices Review, April 2000 00-06
The MnSCU Merger, August 2000 00-07 Early Childhood Education Programs,
January 2001 01-01 District Courts, January 2001 01-02
Affordable Housing, January 2001 01-03 Insurance for Behavioral Health Care,
February 2001 01-04 Chronic Offenders, February 2001 01-05
State Archaeologist, April 2001 01-06 Recycling and Waste Reduction, January 2002 02-01
Minnesota Pollution Control Agency Funding, January 2002 02-02
Water Quality: Permitting and Compliance Monitoring, January 2002 02-03
Financing Unemployment Insurance, January 2002 02-04
Economic Status of Welfare Recipients, January 2002 02-05
State Employee Health Insurance, February 2002 02-06 Teacher Recruitment and Retention: Summary
of Major Studies, March 2002 02-07 Local E-Government: A Best Practices Review,
April 2002 02-08 Managing Local Government Computer Systems:
A Best Practices Review, April 2002 02-09 State-Funded Trails for Motorized Recreation,
January 2003 03-01 Professional/ Technical Contracting,
January 2003 03-02 MinnesotaCare, January 2003 03-03
Metropolitan Airports Commission, January 2003 03-04 Preserving Housing: A Best Practices Review,
April 2003 03-05 Charter School Financial Accountability,
June 2003 03-06 Controlling Improper Payments in the Medical
Assistance Program, August 2003 03-07 Higher Education Tuition Reciprocity,
September 2003 03-08

Evaluation reports can be obtained free of charge from the Legislative Auditor's Office, Program Evaluation Division, Room 140, 658 Cedar Street, Saint Paul, Minnesota 55155, 651/ 296-4708. Full text versions of recent reports are also available at the OLA web site: http://www.auditor.leg.state.mn.us