Australian Computer SocietyTom Worthington, will talk about work on standards for indexing and accessing Commonwealth Government documents on-line. He will describes the frustrations and successes of chairing and being a member of multi-disciplinary committees of IT, library, archives and records management professionals. Tom argues that the Information professions will need to work more closely to delivery information in the on-line age.
This document is: http://www.acs.org.au/president/1997/vala98.htmDraft of 15 December 1997:
The content of this talk will be developed here. The printed version (15 December 1997) for the proceedings is available and "slides" will also be available. Suggestions and comments welcome: tom.worthington@tomw.net.au
In March 1997 I visited the USS Blue Ridge by helicopter (6) off the coast of Queensland, during Exercise Tandem Thrust 97.
The Blue Ridge (LCC 19) is the flagship of the U.S. Seventh Fleet; a 620-foot 18,500 ton, 1550 crew purpose built Command and Control ship.
The "Joint Maritime Command Information System" consists of computers distributed throughout the ship, with data from world-wide sources presenting a tactical picture of air, surface and subsurface contacts. According to the Blue Ridge Home Page: "...enabling the Fleet Commander to quickly assess and concentrate on any situation which might arise. This ability to access information from military and civilian sources throughout the world gives Blue Ridge a global command and control capability unparalleled in Naval history." (7)
The Blue Ridge is essentially a floating web server. Military personnel sit at PCs and laptops, using mostly ordinary office automation and internet software. Equipment is tied down using webbing straps, rack mounting and in some cases adhesive tape.
1
2
3
4
We are at the beginning of the end of a revolution in the way organisations operate. That revolution is the Internet. The effects of that revolution are rippling through government and private organisations, effecting the way they work. Last year it became fashionable for your organisation to have a web page. By the end of this year it will be fashionable to run your organisation via the Internet.
As the "mission critical" information for an organisation is put on-line, it becomes critical for the information professions to work together. Not only will individual organisation's records and operations be at risk from demarcation disputes between previously separate disciplines, but the whole cultural heritage of the world.
Computer and telecommunications professionals, librarians, archivists and records managers previously did their work largely independently in different parts of the same organisation. My experience of the last four years has been that the Internet forces these disciplines, and others such as journalism, together. We need to respect the role played by the different disciplines, learn each other's language and techniques. This must be done in the next few months, to keep up with the pace of technological development and the great enthusiasm of people in taking up the Internet.
Organisations and whole communities will become dependant on the Internet in the same way we now depend on the telephone, electricity and water supply. A loss of on-line records would wipe out an on-line organisation very quickly. A lack of appropriate archiving might wipe out most written records of our culture in a few years.
Canberra, where I live, is a planned city. The competition for the design of Canberra was a controversial and slow process, partly because it coincided with a turf war between design professionals (1).
At the time of the competition for the design of Canberra there was no separate town planning profession. Those who entered the competition were architects, surveyors or engineers. Each discipline saw its own as the natural one to extend to town planning. The merits of the site and designs for the city were argued over using the language and concerns of separate disciplines. The result was that a demarcation dispute was played out in public, to the confusion of the professions, the clients and the public.
In 1994 I chaired an Federal interdepartmental committee on electronic document management (2). The report of the committee discussed possible impediments to the use of electronic documents by Federal Government agencies and the implications of the Internet.
On the committee were IT professionals, librarians, archivists and records managers. We quickly discovered that we used different language, had different ideas as to what was important and how to work. An example was the word "archive", which had very different meanings to IT people and the others: IT people might "archive" something in a matter of minutes, archivist in decades.
That committee and report were concerned with internal record keeping by Commonwealth Agencies. This is a relatively dry topic, usually only of interest to professionals and bureaucrats. The only time record keeping methods are of public interest is when an important record is lost, or a public figure wished it was lost. ;-)
About eighteen months after the committee I chaired, I was asked to be a member of a further one on on-line indexing of electronic documents in federal agencies (3). In that time the Internet had come to public attention and agencies were starting to put material on-line for the community. The Internet had gone from something for academics and anarchists, to a serious computer network for Government and business. The Internet was starting to provide the standards, infrastructure and customer base for serious electronic document systems. The issue of the format of electronic documents and how to find them became more of a community concern. The keeping of records in agencies and public release of material by agencies was starting to converge.
As a member of the second committee I argued that we could use meta-data tags embedded in web documents (5) for indexing Government information. The idea was that meta-data, the sort of information usually stored in a records management system, could be simply typed into the top of each document using a text editor. The same technology used to "crawl" the web indexing documents could crawl an agency's records collecting the meta-data.
The idea that meta-data would be scattered around the system in individual documents was too radical for many of my colleagues and still is for some. They are used to nice neat collections in central databases. However, my experience as Defence Web Master was that organisation units would not surrender control of information to any central authority. The monopoly which central records areas had would not last in the electronic age.
In November 1997 I was nominated by the Department of Defence as a member of the "Search Engine Working Group" (4). This is another Federal interdepartmental committee looking at the details of implementing what was proposed by the previous committee (with several of the people from previous committees).
The Search Engine Working Group (SEWG) was established in October 1997 by the Chief Government Information Officer to consider proposals for the development of whole-of-government entry point services for, including both browse and search services. SEWG will make recommendations to the Commonwealth/State Navigation Working Party (8).
By now the use of the Internet and web has gone from being perhaps acceptable, to being essential. This was no longer an In the next year the Internet will go from essential, to being the core of many businesses and industries. However, the different points of view as to how to use technologies from different professions has not gone away.
Exactly how do you build a system to search all public government information in Australia? What technical architecture will be scaleable to work on such a large collection of information? Will free text searching do the job? Can we rely on agency staff putting in meta-data consistently? Will a centralised or distributed harvest and search strategy be more workable? What services will the public expect from such a service? What will professional information users expect? What is it reasonable for thegovernment provide? These questions are being debated by the working group right now (December 1997) on-line. I hope to be able to share some of the results with you at the VALA conference in January 1998 and on the 'net.
The rate of progress with the 'net has emboldened us. Half jokingly I have said that SEWG were specifying a system in November 1997 which was beyond the state of the art, on the assumption that the technology would have been invented by the time it was needed in 1998. That assumption may be proven correct.
As I was about to finish this paper by the deadline of 15 December (so it could be e-mailed from Perth to Melbourne and printed on old fashioned paper), I received an announcement about the Australian MetaWeb Project (9). This is a joint initiative of Australian Universities, researchers and agencies to develop these tools for meta-data. This would allow metadata to be stored in a local repository rather than embedded within HTML pages and this gathered by the harvesting engines. The Project will utilise Dublin Core metadata and will run from September 1997 to July 1998.