Freedom's Just Another Word ... for Metadata:

Knowledge Management and Discovery via DASL, Z39.50, X.500, and the DMA



Owen D. Ambur
Division of Information Resources Management
U.S. Fish and Wildlife Service
Arlington, VA, USA 22203
phone: (703) 358-1729

Owen_Ambur@fws.gov


1.0 PANEL INTRODUCTION

The impetus for this panel was a message posted to the GILS listserv by Ray Denenberg suggesting that if the members of the World Wide Web Consortium (W3C) do not currently recognize the need for the functionality of Z39.50, they will eventually come to understand it.[1] No doubt, the librarians have a great deal of expertise from which we can learn, but they too have much to learn about how best to manage and share knowledge in the information age. Likewise, records managers have expertise from which we can profit, but they themselves must keep abreast of rapidly changing realities as well. None of us is exempt from the need to rethink yesterday's knowledge and logic in the face of rapidly evolving paradigms. Hopefully, this panel can make some small contribution toward reducing the learning curve for all of us.

The reference to "freedom" in the title I have chosen for the panel relates not only to Kris Kristofferson's famous ballad, Me and Bobby McGee, but also to the Freedom of Information Act (FOIA), under which U.S. federal agencies are required to make information available to the public. In my paper published in the proceedings of the 1997 IEEE Metadata Conference, I suggested that metadata specifies what freedom of information truly means.[2] Without it, FOIA is merely a mute slogan ... an emperor bereft of clothing ... a parade without floats, marching bands, balloons, or horses, much less an audience.

Statisticians are familiar with the concept of "degrees of freedom" which relate to the reliability of study results based upon the size of the sample population. Likewise, metadata defines the degree to which information is "freely" available and usable. Incremental benefit/cost analyses should be employed to assess the relative value and cost of various elements of metadata - both to the suppliers as well as the users of metadata.

A good place to start would be to specify the minimum metadata requirements to effectively implement the Electronic Freedom of Information Act Amendments (E-FOIA) - not merely in the U.S. but, at least potentially, for governments worldwide. Is it the Dublin Core?[3] A subset of the GILS profile? [4] The X.500 Green Pages attributes?[5] The ODMA elements?[6] Let us set forth a challenge: Let freedom ring. Let it ring true to the spirit of service to the citizens of the wide world. And, keeping in mind that we must walk before we can run, and that we must not run too far ahead of those whom we would hope to lead, let it begin with those of us who understand the vital role of metadata.

In that vein, I would also note that "metadata" and "management" are synonymous with respect to knowledge as it is represented in information technology (IT). Metadata specifies the parameters by which knowledge can be and is - or is not - in fact being managed. "Knowledge Management (KM)" is among the contemporaneous buzzwords with which we are beset. However, in cutting through the mystical fog that characterizes current KM theory and practice, let one thing be clear: Little or no metadata means little or no management. That may be appropriate for many classes of knowledge and records - those of low quality, little value and, thus, appropriately classified and treated as short-term or even ephemeral records. However, knowledge that is of high value or upon which substantial resources have been expended certainly merits some level of management, which is to say that it warrants the association of some amount of metadata. How best to do so is the essential focus of this panel, whose members and topics are as follows:

WebDAV and DASL endeavor to catch the crest of the tidal wave that is the Internet and particularly the World Wide Web. It is at our peril that any of us choose to ignore that wave.

As Ray pointed out in his posting on the GILS listserv that prompted this panel, the Z39.50 protocol has years of scholarly thought behind it. It's up to those of us who care about knowledge management and discovery technology to determine its future.

In its initial iteration, X.500 suffered from the plague of overspecification and, thus, lack of popular appeal and support. However, it is the international standard for directory services. The requirements that it embodied remain, and its reincarnation as LDAP as well as advancement toward the use of digital signatures have sparked its rebirth. Notably, X.500 provides the potential means to catalog and share "uncaptured" knowledge, that which resides in the minds and skills of the "experts" but which has not been documented.

The DMA recognizes that the universal human interface to recorded knowledge remains the document, and that the word "document" is very broadly defined in IT - as meaning virtually anything that can be given a file name and stored on electronic media. That is, a document is a record of captured knowledge. Subject to appropriate access controls, the DMA aims to share such knowledge widely and openly.

From the perspective of U.S. federal agencies and service to our citizens, several other issues and initiatives warrant mention: First is the GRS-20 decision, which exploded the myth that paper may lawfully be substituted for electronic records.[12] Second is the fact that the Department of Defense has specified the requirements for electronic records management and is in the process of certifying commercial products meeting those specifications.[13] Third are the X.500 Blue, White, and Green Pages, which will provide directory services for government functions, employees, and documents, respectively.[14] Fourth is the newly enacted Government Paperwork Elimination Act (G-PEA), which requires agencies to give the public the opportunity to submit information in electronic form, including the use of digital signatures.[15]

Many government officials will view these requirements as excessively burdensome upon the agencies, a set of self-inflicted "unfunded mandates". However, those officials should remember for whom they work. In our nation, "under God, the people rule." And while it may sound trite to say, in truth, without metadata, there is indeed nothing left to lose ... except the potential to avoid reliving the mistakes of the past in not capturing and managing the appropriate elements of metadata. Those who fail to do so are not only condemning themselves to relive the mistakes of the past, but they are also relegating their knowledge and life's work to the dustbins of irrelevancy. If they choose to do that on their own time, that's one thing. However, if they choose to do it on the taxpayer's dime, that is entirely another. Under U.S. law, it becomes a matter of malfeasance - by the literal meaning of the word - and there are signs that the wisdom of such laws is becoming more apparent to our friends in Europe as well.[16]

Feelin' good may have been good enuf for Kris and Bobby, but it ain't good enough for government work in the information age. Feelin' good may be good enuf, indeed it may be the essential outcome of social engagements, entertainment events, avocations and vacations. But it ain't good enuf for any organization whose aim is to conduct business, serve an economic purpose, make a profit, or survive in a competitive marketplace ... much less to serve the interests of the public. Nowhere is the need for business to be conducted in a business-like way more apparent than in the conduct of the people's business - since government takes its revenues from the people by force of law, rather than by the voluntary exchange of value. By definition, business requires the creation, management, and maintenance of business-quality records. And, make no mistake ... that means metadata.

2.0 REFERENCES

[1] Global/Government Information Locator Service (GILS) FAQ. Available at: http://www.gils.net/faq.html

[2] Ambur, O. Metadata or Malfeasance: Which Will It Be? In Proceedings of the 1997 IEEE Metadata Conference. September 1997. Available at: http://computer.org/conferen/proceed/meta97/papers/oambur/malfea1.html Panel introduction at: http://computer.org/conferen/proceed/meta97/papers/oambur/panel.html

[3] Dublin Core Metadata Initiative. Home page: http://purl.oclc.org/dc/

[4] GILS Core. Available at: http://www.dtic.mil/gils/documents/naradoc/ and http://oseda.missouri.edu/mogils/gilscore.html

[5] X.500 Green Pages. Attributes listed at: http://www.fws.gov/laws/greenpgs.html

[6] Open Document Management (ODMA) Attributes. See http://www.gils.net/odm/ and http://www.aiim.org/industry/standards/index.html

[7] WebDAV. Home page: http://www.ics.uci.edu/pub/ietf/webdav/

[8] DASL. Home page: http://www.ics.uci.edu/pub/ietf/dasl/

[9] Z39.50. Home page: http://lcweb.loc.gov/z3950/agency/

[10] X.500 and LDAP. White Paper by Data Connections Ltd. Available at: http://www.datcon.co.uk/press/mdwhite1.htm

[11] Document Management Alliance (DMA). Home page: http://www.aiim.org/industry/standards/index.html and White Paper at: http://www.erols.com/ambur/DMA.html

[12] Electronic Records Information/GRS 20, National Archives and Records Administration (NARA). Available at: http://www.nara.gov/records/grs20/index.html

[13] Records Management Application (RMA) Certification Testing, DISA, Joint Interoperability Test Command. Home page: http://jitc-emh.army.mil/recmgt/

[14] U.S. Government On-Line Directory (USGOLD). Available at: http://www.usgold.gov/

[15] Government Paperwork Elimination Act (G-PEA). Text available at: http://www.fws.gov/laws/G-PEA.html

[16] Public Sector Information: A Key Resource for Europe. A green paper by the European Commission.  January 20, 1999.  Available at: http://www2.echo.lu/info2000/en/publicsector/gp-index.html



Copyright 1999 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.