integrating records management into information resources management in u.s. government agencies

Integrating Records Management into InformationResources Management in U.S. GovernmentAgencies

J. Timothy Sprehe*

This article presents several key issues facing federal agencies with regards to electronicrecords management practices. The article identifies selected management, technology,implementation, and definitional barriers that agencies encounter when considering how tocreate and maintain an electronic records management process. A key question for federalagency technology managers is: If an enterprise creates and manages virtually all of itsinformation in an automated information technology systems environment, does it makeany sense to manage its records in a manual environment? Based on research conducted bythe author, the article reviews various federal agency electronic records managementprocesses and the implications for those practices.

The term “information resources management” (IRM), as it applies to U.S. governmentagencies, was originally coined in the 1970s by the President’s Commission on FederalPaperwork and later embodied in statute in the Paperwork Reduction Act of 1980.1 Sinceits inception, federal IRM has operated with the conceptual metaphor of “life cycle.” Onefinds the concept applied to information itself, as in the original definition of IRM in thePaperwork Reduction Act of 1980 and in the original version of OMB Circular No.A-130.2 The stages of the life cycle are typically described as:

● Design;● Creation or collection;● Analysis;

*Direct all correspondence to: J. Timothy Sprehe, President, Sprehe Information Management Associates, Inc.,4201 Connecticut Ave., N.W., Suite 610, Washington, DC 20008,[email protected].. The author isgrateful to David Plocher, Kenneth Thibodeau, and John Bertot for comments and suggestions on an earlierversion of this paper, and to NEC Corporation for financial assistance in the preparation of the paper.

Government Information Quarterly, Volume 17, Number 1, pages 13–26.Copyright © 2000 by Elsevier Science Inc.All rights of reproduction in any form reserved. ISSN: 0740-624X

● Use;● Dissemination; and● Storage or disposition.

Life cycle is an organic metaphor, an analogy to the birth, life, and death of organisms.The origins of the metaphor are ancient and the notion of life cycle appears commonlythroughout literature. The life cycle metaphor is applied also to information systems.3

Even the process of acquiring information systems has received the life cycle treatment.4

Less focus is devoted to ways in which the life cycle metaphor obscures rather thanelucidates reality. In large-scale government information enterprises—the decennial cen-sus of population and housing is an extreme case—the life cycle concept obscures the factthat each step in the cycle has its own internal life cycle, often lengthy and complex. Thedesign stage of a decennial census itself, for example, entails data collection, analysis, use,dissemination, and storage or disposition stretching over several years. The same can beargued for the other stages of the decennial cycle, and indeed for most large, complexgovernment information programs.

Perhaps no aspect of IRM has suffered more from the drawbacks of the information lifecycle metaphor than records management. Figure 1 shows records management in relationto the rest of the information life cycle. As in the example above, records management hasits own cycle: creation, receipt, maintenance, use, disposition, accessioning, preservation,and access. A principal drawback afflicting records management is that it is thought tooccur at the endof the life cycle. In practice, this has tended to mean that recordsmanagement is the last thing program managers think about because it is viewed as havingno immediate relevance to program execution. Managers tend to view records manage-ment as only entailing preservation and fail to understand that records management isbothday-to-day record keepingand longer term record retention and disposal. In consequence,records management has become information management’s afterthought, and, in thescale of competing priorities for scarce resources, records management all too oftenreceives little attention or resources. Records management has, for too long, been theforgotten stepchild of IRM.

Figure 1Information Management and Records Management

14 GOVERNMENT INFORMATION QUARTERLY Vol. 17/No. 1/2000

THE CHANGING ROLE OF RECORDS MANAGEMENT

Federal agencies, like other organizations, keep records because they need records toconduct their business (see Figure 2). They keep records because they need to beaccountable to themselves and to external entities (e.g., the Congress, other agencies, andthe public) for how they have spent public funds in discharging their missions. Businessesmust keep records for tax and auditing purposes and for legal reasons. Records are thecritical element in the “institutional memory” of any enterprise, its knowledge of how andwhy things were done in the past, whether the past is last week or last decade. Thebusiness imperative to keep records is so strong that it is codified in statute as the FederalRecords Act,5 the legal requirements for federal record keeping.

Like other enterprises, agencies do not keep records simply because of statutorymandates or because they would be “nice to have” from an historical viewpoint, eventhough the historical value of records is part of the legal requirement in the FederalRecords Act. They keep records because they believe they may need to access the records.The raison d’etreof record keeping is access to the information in records, not simplepreservation. If agency officials believe they will not need to access the information asevidence of the agency’s past collective behavior, then the information should not qualifyas a record in the first place. Records are valuable because the agency places value on theinformation contained therein for its relevance to agency mission and program perfor-mance; they are not just the detritus of yesterday’s activities. To put the matter anotherway, when one speaks of agency records, it is assumed that one speaks of information thata responsible agency official has determined is valuable for executing agency programsand hence worth keeping for future access.

If records management has been a perennial backroom and backstage activity, thatsituation began to change radically in the late 1980s. The Iran-Contra scandal in the WhiteHouse came to light because of the idiosyncrasies of PROFs, the IBM electronic com-munications system used in the White House, a forerunner to what today is universallycalled e-mail. Unbeknownst to the PROFS users on the National Security Council,messages intended for deletion were still captured by the computer backup system. Afterthe Tower Commission’s report on the scandal, the White House proposed to destroy thecomputer backup tapes. Scott Armstrong and the National Security Archives, a nonprofitorganization, sued the White House alleging that the tapes were federal records that shouldnot be destroyed. Thus, began a series of federal court cases that has decisively altered theways in which federal agencies think about records management and treat their records.

Figure 2Why Enterprises Keep Records

15Integrating Records Management

The White House rationale for destroying the PROFS tapes was that they were notfederal records. This contention was based on the existing White House records manage-ment policy, namely, that staff were instructed to print out in paper form any informationthat rose to the level of a record. InArmstrong v. Executive Office of the President,Armstrong and the National Security Archives argued that paper printouts were notadequate substitutes for the electronic versions on the computer tapes. Federal JudgeCharles Richey held for the plaintiffs in late 1992, deciding that the paper record was notequivalent to the electronic record.

What is generalizable from Judge Richey’s opinion was the finding that the WhiteHouse “record keeping procedures are arbitrary and capricious because there is noadequate management program or supervision by record keeping personnel of the staff’sdetermination of record or nonrecord status of computer material.” It was not enough thatagency staff were charged with preserving record material appearing on the e-mail system;the agency records officers were not involved in the decision making.

What the decision meant for federal agencies was that they must establish a well definedrelationship between e-mail and federal records, and must be able to show that agencyrecords officers were supervising how staff decided which e-mail messages were recordsand which were not. What was true for e-mail records was sure to be true for all otherfederal records in electronic formats: word processing documents, spreadsheets, andadministrative and scientific databases. Richey’s 1992 decision was one of the first in aseries of federal court decisions pertaining to the legal status of records “born digital.” Itsignaled that agencies would have to reevaluate their traditional practice of printing outelectronic files and saving the paper versions and be prepared to take their recordsmanagement in an entirely new direction.

An even stronger blow to traditional practice came in October 1997 when Judge PaulFriedman ruled that the National Archives and Records Administration’s (NARA’s)General Records Schedule (GRS) 20, Electronic Records, was arbitrary and capricious inits contents and judged GRS 20 null and void.6 General records schedules cover dispo-sition of records that are common to all agencies (e.g., travel and payroll). NARA issuedGRS 20 in response to Judge Richey’s decision concerning the PROFS tapes. In itspertinent part, GRS 20 said that, when agencies have records in electronic form, therecords must be transmitted to a NARA-approved record keeping system and that such asystem could be paper, microform, or electronic. The paper loophole—the option ofprinting electronic files to paper—was the target of Friedman’s opinion. The decision heldthat electronic files have unique characteristics not captured in paper printouts. Thosecharacteristics are the capabilities for electronic records to be manipulated, searched,indexed, and transmitted via telecommunications in ways that are not mirrored in thepaper world. Paper copies, in the court’s view, simply do not capture the full legal,administrative, research, and historical value of the electronic record. Friedman also heldthat, in issuing GRS 20, NARA had exceeded its authority by improperly lumpingtogether in a general records schedule administrative and program records. He declaredGRS 20 null and void.

The Friedman decision was a major shock to the federal information technology (IT)community. Federal IT managers now considered it inevitable that the courts would orderthe agencies to manage electronic records in an electronic environment and cease thepractice of paper printouts. The Clinton administration appealed the Friedman decision.


The government did not contest the court’s argument about the unique characteristics ofelectronic records. Rather, it contested the court’s conclusions about whether the Archivisthad exceeded his authority in issuing GRS 20. It also argued that the court’s decision, ifpractically implemented, would have a paralyzing effect on the government because thesoftware tools to implement the decision did not exist and/or were not installed ongovernment computers. On August 6, 1999, to the surprise of the IT community, the Courtof Appeals for the District of Columbia overturned the Friedman decision (http://www.cadc.uscourts.gov/). The Court of Appeals held that the Archivist had not exceededhis authority and reinstated GRS 20. The decision effectively threw the problem ofelectronic records management policy back into the lap of NARA.7 The plaintiffs in thecase appealed to the Supreme Court on November 4, 1999.8

The series of court cases also had the effect of spotlighting the growing business casefor electronic records management. IT managers were asking themselves: if an enterprisecreates and manages virtually all of its information in an automated IT systems environ-ment, does it make any sense to manage its records in a manual environment? The answerseemed obviously to be no.9

The effect of the court decisions and the business case logic on federal agencies was toconvince the more forward thinking IT managers that they must begin actively exploringacquisition of electronic records management capability. Whereas up until now agencieshad progressed toward an ever more electronic environment for creating, managing, anddisseminating government information, they had continued the practice of printing outpaper copies of documents and treating the paper copy as the record copy. The agenciespracticed increasingly automated information management except for manual recordsmanagement. By the mid-1990s progressive IT managers adopted the conviction thatmandatory electronic records management was simply a matter of time.

STANDARDS AND SOFTWARE DEVELOPMENT

This conviction grew particularly strong at the Department of Defense (DoD). In the wakeof the Gulf War, DoD experienced unprecedented external and internal demands forinformation. Internally, the department had enormous need to share information amongthe military services before, during, and after the conflict. DoD encountered a myriad ofincompatible IT systems so that, internal to the department, major components could notcommunicate with one another. Externally, phenomena such as the Gulf War Syndromegenerated voluminous requests that DoD publicly divulge any information that couldaccount for the combination of medical symptoms combat veterans were experiencing.The Secretary of Defense appointed a Special Assistant for Gulf War Illness and devoteda special web site (http://www.gulflink.osd.mil/index.html) to making publicall unclas-sified and declassified documents pertaining to the illness.

In 1995, DoD formed an internal task force on electronic records management, directedby Kenneth Thibodeau, then director of the Center for Electronic Records at NARA.Defense records management officials recognized that, even if DoD decided to initiatewidespread electronic records management forthwith, neither custom-designed nor com-mercial off-the-shelf (COTS) software systems existed for implementing such a decision.Hence, the job set for the task force was to create functional, system, and data require-ments for managing records electronically. What the task force did, in essence, was to take


the corpus of federal records management laws and regulations and extract from it abaseline set of minimal requirements meeting two criteria. The criteria were that therequirements must be satisfied to manage records and must be capable of implementationacross all of DoD’s components.

The work of the task force led to the issuance of a DoD standard, DoD 5015.2-STD, theDesign Criteria Standard for Electronic Records Management Software Applications.10 InNovember 1997, DoD published 5015.2 as a standard applicable to all DoD units andstated that the standard would become mandatory as of November 1999. In effect, thestandard said: any software system acquired by a DoD component that purports to carryout records management must comply with 5015.2. Next, DoD established a testingprogram at the Joint Interoperability Testing Command (JITC) at Fort Huachuca, AZ. AnyCOTS vendor putting forward an electronic records management application (RMA)product was encouraged to have the product tested at JITC. Products that successfullypassed the JITC test would be awarded a certification that could be used in the vendor’ssoftware sales program. In December 1998, the Archivist of the United States issued astatement saying that 5015.2 could usefully be applied by any agency in the executivebranch of government.

The advent of 5015.2 and its associated testing and certification process had a galva-nizing effect on the software industry. COTS vendors now had the opportunity to receivea “good housekeeping seal of approval” from the largest federal purchaser of software, anagency that annually spends billions of dollars to purchase software systems. The JITCcertification effectively granted successful vendors a hunting license to sell to DoD andany other federal agency. The software industry responded with a miniexplosion of newRMA COTS products.

As of November 1999, JITC had certified 15 COTS software products sold by 10companies. The products are listed in Figure 3.

Two aspects of these software systems deserve comment;

● Provenance Systems and PSSoftware Solutions Ltd. are Canadian companies;Tower Software Corp. is an Australian company. In general, the Canadians and

Figure 3JTIC Certified List as of November 1999


Australians have come to electronic records management several years earlier thanthe Americans and the United States is indebted to these nations for the advancesthey have made.

● PCDOCS is primarily a vendor of electronic document management systems andFileNet is primarily a vendor of workflow systems. These two companies haverecognized the importance of electronic records management and have bolted onelectronic RMAs to their products through strategic alliance with an RMA vendor.

In general, workflow software systems are systems that track and manage the way thatwork flows through an organization. They are especially useful where work processes aresequential. Electronic document management systems (EDMS) manage the creation andprocessing of documents throughout an organization. What appears to be happening is aconvergence between workflow and EDMS, on the one hand, and electronic RMAs, on theother. In the next few years, workflow/EDMS vendors will add RMA functionality to theirproducts and RMA vendors will add workflow/EDMS functionality11 for the basic reasonthat customers want integrated software systems. As this trend matures, organizationsacquiring enterprise-wide information management systems will be simultaneously ac-quiring electronic records management functionality as part of the package. Over the nextfive years, it is reasonable to expect that electronic records management will become astandard feature of the automated office environment.12 The argument advanced by thefederal government in appealing the 1997 decision by Judge Friedman will be renderedobsolete by developments in the software industry as electronic RMAs spread throughoutfederal, state, and local government agencies, and the business world as well.

NEXT STEPS

Operationally, how does an electronic RMA work in the office environment? An instruc-tive example is found in the experience of the Office of Thrift Supervision, Departmentof the Treasury (OTS).13 OTS purchased Tower Software Corporation’s TRIM product in1998 and began gradually introducing it within the agency. In pilot tests, OTS recordsmanagers developed software such that, when desktop computer users acted to save adocument in a word processing application, a pop-up box would appear on the screen andpose three questions, as shown in Figure 4.

In practice, the OTS records management personnel found that the pop-up box encoun-tered user resistance and also generated confusion. The confusion arose because, whenasked whether the document was a record, the reaction of many OTS personnel was thatthey did not know what constituted a record and what did not. The reaction is under-standable in light of the fact that most federal employees have never had the most basic

Figure 4OTS “Pop-Up” Box


training in records management. If nothing else, the advent of electronic RMAs to thefederal office environment will force agencies to address their chronic neglect of recordsmanagement and provide basic training to all personnel. As agencies explore options forelectronic RMAs, they are encountering the fundamental reality that you cannot do goodelectronic records management unless you are doing good records management in the firstplace.14

OTS recognized that the combination of confusion and resistance to declaring adocument a record could lead to substantial under-reporting of materials that a qualifiedrecords manager would judge to be records. Under-reporting or nonreporting of recordsbecause of user resistance is a problem records managers believe may afflict all RMAs.The ideal solution to user resistance and confusion in the face of an electronic RMA, fromthe standpoint of records management, would be a system that was transparent andunobtrusive to the user. In the traditional world of nonautomated, paper-dependent offices,records management occurred in the file room in interactions between secretaries andrecords officers. Secretaries typed documents and were trained to include the propernumber of carbon copies: copies for the “official” file, for the chronological file, and forcross-filing categories. The records manager made sure that copies were filed according toan approved file plan. These practices presumably were transparent and unobtrusive to thedocument author in the sense that they generally happened behind the scenes and out ofmanagement’s sight.

In today’s automated environment, the ideal solution, one that is transparent andunobtrusive to the computer user, would be an “expert system.” In principle, it should bepossible to design an expert system that would examine the contents of all an office’selectronic files—word processing, spreadsheets, e-mail, and so forth—and employ a set ofdecision rules to determine which files qualify as records; and then to associate recordmaterials with the correct schedule for retention and disposal. The process would operateindependent of the file author. Records management experts would design the decisionrules in collaboration with software engineers and could monitor and adjust the operationof the expert system to make certain that it is functioning properly. Without interventionfrom document authors, the expert system would decide which files were records andcause them to be associated with the correct file in the approved file plan.

This train of thought has not been lost on the software industry. As one example, inOctober 1999, Provenance Systems announced a partnership with a knowledge manage-ment firm, Autonomy Inc., to market AutoRecords, characterized as “a solution for theautomatic classification of electronic records.” In the words of Provenance’s promotionalliterature: “With AutoRecords, all electronic information is automatically captured andprocessed, not by the end user, but by an automatic process dramatically increasing thespeed and accuracy of classification for official records contained within large volumes ofinformation.”15

It is far too early to say whether this or any similar system automated approach canbe successful in identifying and classifying records. Many doubt that it can, but thenfew have ever tried an automated approach. After all, the definition of what is a recordand what is not continues to be debated year in and year out. Outside of certaindocuments that are “automatically” classified as records,16 government personnel havelittle guidance beyond the statutory definition of a record (see footnote 5) and theirown judgment.


CONSEQUENCES FOR INFORMATION RESOURCES MANAGEMENT

The advent of electronic RMAs and the incorporation of records management into theautomated office environment has potentially profound consequences for federal IRM.The statutory definition of IRM found in the original Paperwork Reduction Act of 1980and echoed in the 1985 OMB Circular No. A-130, Management of Federal InformationResources,17 was a bifurcated concept. IRM was two things:

● The management of information considered as a resource; and● The management of those resources associated with information, such as computers,

telecommunications, personnel, and funds.

Federal IRM has lived throughout its brief history with the curse of this bifurcation.Although agency managers might give lip service to the primacy of information manage-ment as the essence of IRM, resources have flowed overwhelmingly to IT management.Typically, the senior-most IRM official, today called the Chief Information Officer (CIO),came from a background of IT management and spent most of his or her career with theacquisition and management of IT. The vast majority of IRM dollars went to IT acqui-sition and management. Information management typically referred to the duties of anagency’s Privacy Act officer, Freedom of Information Act officer, paperwork controlofficial, and records officer, four sets of duties usually relegated to lower rankingpersonnel.

As the demand for greater information dissemination to the public grew in the 1980sand hyper-accelerated with the advent of the Internet in the 1990s, information manage-ment began to occupy more of the time and attention of agency CIOs, intruding on theprimacy of IT management. The 1995 amendments to the Paperwork Reduction Actparticularly placed emphasis on information dissemination, adding responsibilities in thisarea both to the duties of the Director of the Office of Management and Budget (OMB)and to agency heads. Congress, in effect, widened the missions of all federal agencies toinclude attention to information dissemination. Initiatives such as the Vice President’sNational Performance Review accentuated the message.

Electronic records management’s arrival onstage has occurred somewhat independentfrom the new emphasis on dissemination, driven more by Iran-Contra and the courtdecisions. Whatever the historical causality, the integration of information managementand IT management in federal agencies is finally being forced. Once electronic RMAs areincorporated into an agency’s IRM environment, the following propositions are truer thanever before:

● It is no longer possible to plan and manage the life cycle of information withouttaking into account the life cycle of the IT system in which the information resides.

● It is no longer possible to plan and manage the life cycle of an IT system withouttaking into account the life cycle of the information residing in the system.

What evidence presents itself for these assertions? As an example, consider the studyof digital imaging standards completed in 1999 by Lockheed Martin on behalf of NARAand DoD.18 Following are excerpts from the executive summary of the study;


The use of computers is changing the way government documents are created, accessed andmanaged. Electronic records, the Internet and E-mail have become an increasingly large partof the everyday work environment. To improve access, distribution, and interoperability,Federal agencies are converting large numbers of documents from paper to electronic digitalimages. . . .

Long-term preservation of digitally imaged records has become problematic for Federalrecords requiring permanent retention. While the advantages of digitally imaged documents aretremendous, due to the relatively short life cycle of digital image technology (both hardwareand software), it is commonly accepted that all formats used today will eventually becomeobsolete.

. . . For an electronic record long term preservation requires that as the technology changes that therecord be migrated from one format to another and then verified to ensure no loss of data. . . .

Plan and budget for migration of digital images every 3–5 years with cost equivalent to50–100% of the costs associated with original imaging project.[emphasis added]

These quotations were selected to make a point. The point is that the study tells federalIT managers to change their thinking aboutinformation system planningbecause of thenature of theirinformation managementresponsibilities. The logic and economies oftoday’s IT capabilities drive agencies toward digitization of information, but the nature ofrecords management forges an indissoluble link between information management and ITmanagement. Yes, an agency realizes economies and efficiencies by automating itsinformation management. But the agency also incurs new and perhaps unanticipated ITmanagement costs by the same token. To restate the propositions: information manage-ment and IT management must occur in an integrated framework in today’s officeenvironment. Records management requirements are only a reflection of the need for anagency to manage its information holdings efficiently and effectively.

This realization gives new weight to dicta records managers have been uttering foryears. Archivists have long preached that the time to plan for records management is atthe point when IT systems are being designed; that is,at the beginningof the system lifecycle, not at the end. Failure to plan and budget for proper information (read: records)management at the outset of the system life cycle leads to far greater costs at the end ofthe cycle and/or the failure to carry out statutory records management responsibilitiesaltogether.

THE INTEGRATED IM/IT ENVIRONMENT

Electronic records management has, by no means, become established on the federalscene. Agencies still have to gain much experience before electronic RMAs are afunctioning reality in their IRM environments. Among other things, introducing anelectronic RMA into an agency will certainly require extensive business process reengi-neering. The few federal agencies that have served as pioneers learned that they must firstacquire the RMA and attempt several pilot experiments before reaching an understandingof how and where their business processes need reengineering. These agencies wouldgenerally concede they were naı¨ve about how much process and policy change would benecessary before they achieved operational success at electronic records management.And, government-wide guidance from NARA is essential as the agencies move forward.19


At the same time, the promise of electronic RMAs may bring records management toits rightful position in federal IRM. The eyes of management officials tend to glaze overwhen one begins speaking of records management. Henceforth, it should be possible tosell records management internally as one key element in an enterprise-wide knowledgeand information environment, something that managers find much more attractive.

Figure 5 presents one picture of a federal agency office environment in which infor-mation management and IT management are integrated to encompass the full life cycle ofboth technology and information. The desktop computing capabilities should enable thefederal knowledge worker to accomplish all of the following;

● Manage records electronically, including e-mail, word processing documents,spreadsheets, Web site content creation, multimedia files, and even voicemail.

● Manage responses to Freedom of Information Act requests, both paper-based andelectronic, and provide files needed for agency litigation support.

● Manage word-processing, spreadsheets, and databases in an integrated manner, acapability present in most desktop systems today.

● Manage various document versions, perform group editing of documents, and carryout electronic document clearance. In government agencies, and probably in alllarge organizations, many different offices must concur in the content and form ofmajor policy and procedural documents before they are deemed final and official.Traditionally, sequential pen-and-paper signatures constituted the clearances, a timeconsuming and frustrating process. The integrated IM/IT environment, includingdigital signatures, should permit electronic performance of this function.

● Manage information in a networked environment that includes e-mail and aninternal agency network or intranet.

● Manage the external network, the agency Internet presence, within the same system.● Manage variegated information dissemination programs via the same system.

Agency publishers should be able, from a single integrated system, to dispatchinformation products via fax-on-demand, e-mail, Web site publication, CD-ROM, orprint publication.

The Government Information Locator Service

The Paperwork Reduction Act Amendments of 1995 established the GovernmentInformation Locator Service (GILS)20 to replace the previous Federal Information Locator

Figure 5Integrated IM/IT Environment


System. In order to assist the public in finding government information, agencies were toestablish and maintain an electronic information locator service. Although considerationof GILS is tangential to the principal thrust of the present article, several points are worthnoting;

● By 1999, GILS was either fully integrated into agency Web sites or fully forgotten.● In the author’s opinion, GILS was and is a highly useful metadata standard for

making agency public information resources accessible electronically. The deficien-cies of GILS arose from the manner in which it was implemented through theauthority of the Office of Management and Budget. Had OMB insisted that GILS bedemonstrably integrated with then-current agency Web site development GILScould have been a substantial success. In practice, however, OMB instructedagencies to carry out the GILS mandate, and agencies, perceiving the mandate as anonerous “unfunded mandate,” accomplished the minimum necessary to signifycompliance and then turned their attention elsewhere. Where GILS was successfullyimplemented, it became indistinguishable from other Web site content.

In principle, the advent of widespread electronic records management might be seen asa boon to GILS. When records are captured into an electronic record keeping system, itshould be much easier and less expensive to make them available to the public.

Information Management and Records Management

Finally, an important difference obtains between information management and recordsmanagement. Despite its name, the Freedom of Information Act (FOIA) does not entitlethe public to gain access to government information. Rather, FOIA entitles individuals togain access to government records. The same is true of NARA’s enabling legislation.NARA does not preserve information; it preserves records.

This distinction may sound like a trifling quiddity but it is not. Records presumablycontain information. However, laws define and pertain to records, but not to the infor-mation contained therein. (Indeed, the legal definition of “information” is “accusation ofa crime.”) Records contain information in a particular physical or electronic form; they arethe setting down of information in the context of the business operations of an agency. Itis the agency context, its mission and programs that gives reality to the records.

Records preserved in electronic form make information perhaps more accessible thaninformation on paper records, although not necessarily so. The entire purpose of bringingrecords under the discipline of records management is to render the information containedin the records more accessible because the information is needed for agency business.Keeping records may be praiseworthy, but only if the records are kept systematically sothey are reliably connected to the agency’s business. NARA’s regulations require thatrecords be associated with arecord keeping system.The point of the record keepingsystem is to bring the records under the discipline of a file plan and a scheme for retentionand disposal. The file plan classifies the records so that present and future seekers candiscover where to find them. The retention and disposal schedule tells how long to keepthe records (one year, five years, 50 years, permanently?) and what to do with them(destroy them or keep them in perpetuity). Without question, sound records management


practice—whether paper or electronic records are involved—promotes public access togovernment information, because public access is, after all, one important element in theagency’s business. Because of the inherent advantages of electronic formats pointed to byJudge Friedman, electronic records ought to be more easily accessible than paper, but onlyif associated with record keeping systems.

These distinctions may sound elemental but they bear repetition. Many IT aficionadosespouse the view that one should simply saveeverythingelectronically because computermemory is increasingly cheap. That is, these individuals see no benefit to selecting out thematerials worth saving from those that are not, and discarding the worthless. They fail tounderstand that one only saves in order to access, and that history shows individuals andinstitutions get into trouble just as often from saving too much information as they do fromsaving too little.

The prudent information manager saves only those records that will be needed fortoday, tomorrow, and the distant future, no more and no less. Even so, the packratmentality dies hard. If one cannot in the present discover a good reason for keeping aparticular set of records, it is no more likely that one will discover a good reason in thefuture. True, unforeseen future events may lead to the wish that one had saved certainrecords. Yet, if one could not foresee the events, one could also not be certain the rightinformation was being saved.

With all of this said, will widespread electronic records management lead to greaterpublic access to government information? The answer is that better recordsmanagement—both paper and electronic records—will lead to greater public access togovernment information. Certainly, responses to Freedom of Information Act requestsbecome easier when an agency practices good records management, and electronic FOIAis aided by electronic records management. Improved public access to governmentinformation is driven primarily by an increased awareness in federal agencies that publicaccess is an integral part of agency missions.

NOTES AND REFERENCES

1. For the 1980 version see Title 44, Section 3502,United States Code. “The term ’information resourcesmanagement’ means the planning, budgeting, organizing, directing, training, promoting, controlling, andmanagement activities associated with the burden, collection, creation, use, and dissemination of infor-mation by agencies, and includes the management of information and related resources such as automaticdata processing equipment.” Available: http://www.access.gpo.gov/congress/cong013.html. In the 1995amendments to the Paperwork Reduction Act, the term was redefined in Title 44, Section 3502,UnitedStates Code: “the term ’information resources management’ means the process of managing informationresources to accomplish agency missions and to improve agency performance, including through thereduction of information collection burdens on the public.” Available: http://law2.house.gov/uscode-cgi/.The purpose of the redefinition was to link IRM closely with program performance. See David Plocher,“The Paperwork Reduction Act of 1995: A Second Chance for Information Resources Management,”Government Information Quarterly,13 (1996): 35–50.

2. Note that the original 1980 definition of IRM contains the life cycle metaphor: burden, collection, creation,use, and dissemination of information. Significantly, the definition leaves out “retention and disposal,” thearchival and records management components of IRM. This definition was transported verbatim into the1985 publication of OMB Circular No. A-130,The Management of Federal Information Resources(Washington, D.C.: Office of Management and Budget).

3. See “Life-Cycle Management (LCM) of Automated Information Systems (AISs),” Department of Defense


Directive No. 8120.1, January 14, 1993, and associated publication such as DoD Directive 7920.1. Thisdirective has now been superseded.

4. See Office of Government-wide Policy, General Services Administration,A Guide to Planning, Acquiring,and Managing Information Technology Systems, Version 1, (December, 1998).

5. The statutory definition of a record is: “Information, regardless of medium, detailing the transaction ofbusiness. . . made or received by an Agency of the United States Government under Federal law or inconnection with the transaction of public business and preserved or appropriate for preservation by thatAgency or its legitimate successor as evidence of the organization, functions, policies, decisions, proce-dures, operations, or other activities of the Government or because of the value of data in the record.” (Title44 United States Code, Section 3301).

6. SeePublic Citizen et alv. John Carlin et al., U.S. District Court for the District of Columbia, October, 22,1997. Available: http://www.cadc.uscourts.gov/.

7. Curiously, the Court of Appeals decision did not address or elaborate on District Court Judge Friedman’stheory that electronic records possess unique characteristics not found in paper records, but ratherproceeded to argue in an entirely different direction, namely, whether the Archivist had exceeded hisauthority in promulgating GRS 20. The two decisions, viewed side by side, constitute a gap in legal theory,one that will doubtless occupy both theorists and subsequent court proceedings.

8. Public Citizen et al. vJohn Carlin, Archivist of the United States, et al.,Petition for a Writ of Certiorari,November 5, 1999.

9. In one important sense the argument for paper printouts has not yielded to the inexorable march ofautomation. That sense is the preservation issue. Outside of microform (microfilm, microfiche, etc.), nopreservation medium is yet the equal of paper for long term preservation of records. Even the mostsanguine technophoria does not lead to the assertion that today’s electronic or optical media will be readas easily 50 years from now as will paper.

10. The standard and associated documentation can be found at http://jitc-emh.army.mil/recmgt.11. Corporate mergers and acquisitions will accomplish the same end results.12. For example, Tower Software Corp. and Provenance Systems have both announced products that combined

EDMS and electronic RMA capabilities. Available: http://www.provsys.com and http://www.tower-usa.com.

13. This account is based on personal interviews and other communications between the author and the recordsofficer at the Office of Thrift Supervision in 1998 and 1999.

14. For example, all RMAs require as input the agency’s records schedules. If schedules do not exist or are notup to date due to agency neglect of records management, the agency must first devote effort to its schedulesbefore it can fully utilize an RMA’s capabilities.

15. Available: http://www.provsys.com. At this writing, the author has no information as to how effectiveProvenance’s AutoRecords is at classifying records or even whether it functions as an expert system withdecision rules approximating federal records management laws and regulations.

16. Official government actions such as laws and regulations are obviously records. Any official publicationof an agency is considered a record also, frequently a permanent record to be retained in perpetuity.

17. Office of Management and Budget.Management of Federal Information Resources, Final Publication ofOMB Circular No. A-130,Federal Register, 50 (December 24, 1985): 52730–52751

18. Susanne H. MacTavish & Michael R. Pickard,Electronic Digital Imaging Standards for ArchivingRecords, Report No. GA22F042 (Falls Church, VA: Lockheed Martin Technology Services, June 1, 1999).

19. See General Accounting Office,National Archives. Preserving Electronic Records in an Era of RapidlyChanging Technology, Report No. GAO/GGD-099–94 (July 19, 1999).

20. See Title 44, Section 3511,United States Code.


integrating records management into information resources management in u.s. government agencies

Documents