Freedom of Information: AIIM's GovSIG GOTS What It Takes to Satisfy - Part 1
Special to the NCC-AIIM newsletter, The Capitol Image
By Owen Ambur, February 3, 1999

Marilyn Vos Savant is listed in the Guinness Book of World Records Hall of Fame for "Highest I.Q." She's best known for her weekly "Ask Marilyn" column in Parade. In her book entitled The Power of Logical Thinking, she makes the following observations:

It doesn't take a genius to figure out that getting an answer, much less a straight answer from our government isn't always as easy as it should be. And that is not to point a finger of accusation at anyone. The same is true of any large organization. The conservative's view is that government should only do for people what they cannot do for themselves. By definition, those are the tough things and dealing with them is seldom simple or straightforward. Moreover, large, tough issues seems invariably to lead to large, intransigent bureaucracies in which plausible deniability is not merely an excuse; it is a reality: When everyone is responsible, no one is responsible. On the other hand, given opportunity and teamwork, there is little that people cannot do and that includes fixing much of what is currently wrong with government.

Opportunity is knocking and the AIIM team is well positioned to answer the call. At the December 10 meeting of NCC-AIIM, Marilyn Wright, Vice President for Standards of AIIM International, announced the formation of a Government Special Interest Group (GovSIG). She noted that members of NCC-AIIM have been clamoring for such a group, and while the focus will be on U.S. federal interests, State, local, and international governmental issues may be addressed as well. Obviously, like all organizations, government agencies (whether they recognize it or not) have a strong interest in many of the activities in which AIIM is already engaged. However, several driving forces suggest that the time may be ripe for AIIM to add very substantial value to the consideration and performance of government functions relative to information technology (IT). Here are a few suggestions.

First and foremost, notwithstanding the hype and confusion engendered by rapid evolution and expansion of use of IT, AIIM can help government officials, employees, and their stakeholders recognize that the document is still the universal human user interface to "captured knowledge." One definition of "document" is "data in context." Without context, data is meaningless. As Ms. Vos Savant points out, not only are errant inferences abundant, they are routine. In the context of "knowledge management," captured knowledge is called "explicit" - as opposed to "tacit" knowledge, which resides only in the heads of people and thus is at risk of walking away at any time. In the context of government (or any other organization), captured knowledge is called a "record" and anything else is an adverse audit or even a lawsuit waiting to happen. Officials who fail to ensure adequate documentation and management of records in the routine course of their business processes are implicitly embracing a policy of misinformation, if not disinformation.

For good or ill, leaders lead. That's their role. That's what they do. Sometimes they're even called to account for their leadership. For example, even Cabinet secretaries are at risk of being found in contempt of court for failing to produce records requested in litigation. [The Washington Post, Jan. 26, 1999, A17.] On the other hand, as deepening scandal engulfed the Salt Lake City sponsors of the upcoming winter Olympics, their predecessors in Nagano, Japan, came up with a highly effective means of retroactively avoiding the same fate: They simply burned their records, as a "courtesy" to those who might be embarrassed by them. [The Washington Post, Jan. 21, 1999, A1 & A22]

As far as the interests of individuals and private organizations are concerned, sometimes the best form of records management is to ensure no record at all. However, service of the public interest demands good records, and in our culture, officials who fail to uphold their obligations in that regard are rightly at risk of embarrassment, if not necessarily action adverse to career enhancement. Sadly, it is often the taxpayers who end up footing the bill not only when the record establishes the malfeasance of public officials, but also when those officials succeed in failing to create and maintain adequate records to justify their actions on behalf of their employer - the U.S. taxpayer.

Is malfeasance too strong a word to use? You be the judge. The Federal Records Act (FRA) requires agencies to make and preserve records of their activities and the important information they possess. The Electronic Freedom of Information Act (E-FOIA) requires that records be made available in electronic form upon request, and the GRS-20 decision makes clear that paper may not be substituted for E-records. Moreover, the new Government Paperwork Elimination Act (G-PEA) requires that the public be given the opportunity to supply information by electronic means as well. Meanwhile, the Government Performance and Results Act (GPRA) requires agencies to establish long-term goals and annual objectives, to link those objectives to budgetary inputs, and to measure outcomes. Implicit is the need for documentation and records management, and those needs are universal not just among government agencies but all business organizations, worldwide.

In 1995, under the auspices of AIIM, the Black Forest Group set forth in a 65-page booklet the "Requirements for an Enterprise Document Management System." With the explosion of the Internet and the growing recognition of the need to manage E-records, now may be the time to update and supplement that document while scoping the "enterprise" to encompass "government" as a logical whole or, better yet, the entire planet. Why not specify the document management requirements for "enterprise earth"? Why not specify the minimum set of metadata elements defining what "freedom of information" really means (or, at least, what it should mean) with respect to public records for any and all government offices, at all levels, world wide? Is it the Dublin Core? A subset of the GILS profile? The ODMA metadata elements? The elements of the X.500 Green Pages? The emerging set of properties that may be queried via DASL? Some other set?

Is the notion of a worldwide standard for freedom of information too bold? Too much to bite off? Bound to fail? ... Perhaps ... On the other hand, why not give it a try? In this world, there are no perfect answers but the only unforgivable failure is the failure to try, particularly when the need is so clear. As the cliches say, nothing ventured, nothing gained and better to fail in service to a noble cause than not to strive at all.

In large measure, the Document Management Alliance (DMA) is already addressing many of the requirements. Moreover, the greatest number of errors in IT systems development are made in requirements analysis, rather than in programming. Thus, it is clear that the greatest error that can be committed with respect to public information is the failure to specify the minimum cross-cutting requirements for managing it - once and for all - and then to continue to improve and build upon the standard. Let's face it; it may be complex but it's not rocket science.

The DMA provides a solid basis on which to build. The Open Document Management API (ODMA) is of the essence of records management; it's where the rubber meets the road. In effect, public agencies that fail to implement and use ODMA-compliant applications are choosing not to manage E-records, which is to say they are flaunting the GRS-20 decision and their statutory obligations to the public under the FRA and E-FOIA. The classic way to commit fraud is to maintain two (or more) sets of books. And that is precisely the effect of processing business documentation outside of the enterprise electronic document/records management system (EDMS/ERMS).

On October 25, 1996, former OMB Director Raines issued a memorandum setting forth guidance under the Information Technology Management Reform Act (also known as the Clinger-Cohen Act). Among other things, those policies (commonly known as "Raines' Rules") require that Federal IT investments should:

These "rules" are actionable imperatives, not only for Federal agencies but also for AIIM. In addition to promoting awareness and acceptance of these principles among agency decision-makers, the GovSIG can add real value to the process by helping to: These principles set forth in Raines' Rule cut across all agencies. So too do the basic requirements for "capturing" and managing public records. There is no reason that each agency should be forced (or allowed) to reinvent the underlying requirements analysis over and over again, while foisting upon the taxpayers the cost of the shortcomings and mistakes that are inevitable in each instance. Indeed, it should be noted that the CIO Council is endeavoring to craft a "framework" for the Federal information technology architecture (ITA). However, the pace as well as the "high" (abstract) level at which the ITA is being compiled seem likely to render it meaningless as agency decision-makers go about the business of acquiring and developing systems - systems which may or may not: a) support the way the work is actually done, b) be appropriately "chunked,"c) be interoperable, and/or d) take the public interest into account, including their right to ready access to public records.

A somewhat more hopeful initiative, one in which NCC-AIIM has already taken keen interest, is the specification of Federal electronic records management requirements by the Department of Defense, the so-called 5015.2 standard. The information "architects" can devise all the "frameworks" they desire, but where the rubber meets the road, people will continue to conduct their business via documents. At the instant and in whatever form or format those documents are "saved," they become records.

Rightfully so, the persistence (what records managers call "retention") of many of those records will be short or even ephemeral. However, it is incumbent upon agencies to classify and "schedule" all of their records for retention or destruction. That is, agency leaders should give at least passing consideration to the treatment of their agency's most important records, and all other records should be given similar consideration at the appropriate level within the organization. Legalese and the wishes of records managers notwithstanding, there ain't no such animal as a "nonrecord" record. Nowhere is the need for the President's Plain Language initiative more apparent than in the twisted, bureaucratic double-speak that has been applied to the realm of records management.

Reportedly, Australia is seeking the establishment an international standard for records management. AIIM's GovSIG should jump in with both feet and help make it happen, quick and soon. Time, money, knowledge, and records are awastin'. An incomplete standard that is open to enhancement is better than one delayed in search of perfection. We're creating more records now many times faster than at anytime in the history of humankind. It's high time that we start managing them at least in a semi-rational manner. At its root, that is exactly what records management is (or should be) about. The optimal outcome of records management is not merely the destruction of records on a preordained schedule, as many old-line records managers would have us believe. Rather, the desired outcome is access to knowledge exactly when and where it is needed. Records retention/destruction schedules are meaningful only to the degree that they rationally facilitate that outcome.

To make sense of the title of this article, check out Part 2 in the next edition of the Capitol Image.


Owen is a personal member of AIIM and a masters candidate an the University of Maryland University College. He worked on Capitol Hill for 14 years (8 in the House and 6 in the Senate) and was Congressional liaison for the U.S. Fish and Wildlife Service for 7 years. Currently, he is on assignment to redraft the agency's records management policies. His personal home page is at http://www.erols.com/ambur and he may be contacted at ambur@erols.com.

On April 6, at the third IEEE Metadata Conference, at the NIH auditorium in Bethesda, Owen is scheduled to chair a panel entitled "Freedom's Just Another Word ... for Metadata: Knowledge Management and Discovery via DASL, Z39.50, X.500 & the DMA." Chuck Fay of FileNet will represent the DMA. Alex Hoppmann of Microsoft will discuss DASL. Ray Denenberg of the Library of Congress will address Z39.50, and Graeme MacArthur of Data Connections Ltd. will cover X.500 and LDAP. The conference Web site is at http://www.llnl.gov/liv_comp/metadata/md99/.