November 11, 2006
Anticipatory Web discussionà [home]
Anticipatory Web Challenge Problem à [link]
Community Centric Service Methodology Glass Bead Games
Lobbying Congress on the Merits of the National Project
Dear Friends,
I write only to say that reports of the demise of Topic Maps are greatly exaggerated. The U.S. and U.K. are not the only countries in the world.
I know of four annual conferences partly or entirely devoted to Topic Mapping in Canada, Norway, Germany, and Korea. There are literally countless projects, now on every continent (I just learned about one in South Africa). I recently put together some slides that I gleaned from the Emnekart (that's Norwegian for "Topic Map") 2006 conference:
http://www.coolheads.com/SRNPUBS/lita061027/lita025.htm
, and that I showed at a well-attended ALA technical meeting recently. The library community in America is not so insulated from foreign practices as the various communities of the U.S. government are. They knew topic mapping was out there and they wanted to know about it!
I'm not at all worried about topic mapping going away. Here are three reasons why:
1) Subject identity cuts across all ontologies, past, present, and future. And it's the only thing that does! Every subject has multiple ways to identify it, and multiple ways to express every allegation about it, in terms of any number of ontologies.
You probably think that Topic Mapping is just another ontology. It's not, even though RDF bigots wish that it were, and even claim that it is. Instead, it's simply a set of practices that ensure that no subject, and no ontology, is less privileged than any other. See
http://www.isotopicmaps.org/TMRM/TMRM-6.0/TMRM-6.0.pdf .
2) Human nature dictates that languages (including ontologies, vocabularies, and taxonomies, both implicit and explicit) are constantly being invented. Human communications systems, like human communities, are fractal, and monolithic approaches to the problem of communication (knowledge management) are doomed. Every single way of saying anything at all is exceptional. Anytime you have to say something that hasn't been said before, you have to invent a way of saying it. Moreover, when faced with any kind of communications challenge, humans would generally *rather* invent a language than learn one; that's much of what human dialogue is about. The more ontologies/taxonomies/vocabularies fall out of such processes, and the more diverse they are, the more topic mapping is needed to knit the social fabric together.
3) It's a fact that every language, and every feature of every language, is useless baggage in at least some contexts. In the end, every formal language (including OWL) is mission-specific. OWL is falling far short of its lofty, world-beating goals. Those of us who cut our teeth on SGML (now called "XML") have long since moved on to more practical goals. That's why there's an ISO Topic Maps standard; it's made by the same people who brought you SGML. SGML didn't go away, and neither will Topic Maps.
As far as I can see, with respect to topic mapping, the US is falling farther and farther behind.
I see little evidence of malevolence here.
True, there certainly has been a concerted effort among *academics* to make whatever comes out of Tim Berners-Lee's mouth the basis of all grants for knowledge management research -- but ultimately that's pretty much irrelevant. What matters is real solutions to real problems that real people and institutions can live with.
I now believe the failure of U.S. institutions to adopt topic mapping is mostly structural. For competitive reasons, other societies need enhanced transparency in order to make the economic alliances that they need in order to enjoy the economies of scale that the U.S. has enjoyed for most of two centuries.
I agree with Paul to the extent that large U.S.-based enterprises whose business models are predicated on the existing inefficiencies in the ways we handle knowledge and "intelligence" don't want things to get more efficient.
And let's not lose sight of the fact that the World Wide Web Consortium -- the source of RDF and OWL -- is a vendor consortium. Contrary to its public image, disruptive change is exactly what the WWW seems to work to *forestall*.
With respect to the U.S. government, which ideally should be a bastion of transparency, here are two observations:
1) Real intelligence, and real transparency, is exactly what the Bush administration doesn't want. (Insert your favorite piece of evidence here; there are so many to choose from!)
2) The E-Gov Act of 2002 is basically a list of the design requirements that, realistically speaking, only topic mapping can meet. The E-Gov Act of 2002 remains unfunded -- an unwanted stepchild if ever there was one.
Maybe the U.S. economy will enhance its own transparency -- making it cheaper and easier for all of us to know stuff and to exploit our knowledge of it. Or, maybe the U.S. will be overtaken by nimbler economies whose participants know more, know it sooner, and are not burdened by the economically parasitic activities of those who make a profession of squabbling over intellectual property and engaging in similar rent-seeking behaviors of dubious value to the productivity of society as a whole.
If, as an American, you want to frighten yourself, consider China's de-facto laissez-faire approach to intellectual property. Then consider Friedrich Hayek's thinking about what makes an economy robust and adaptable -- namely, doing everything possible to minimize the role of ignorance in the structure of every economic transaction. Hayek showed that, when everybody wins -- when all deals make sense in terms of reality -- everybody wins much more. In his famous "The Road to Serfdom" -- a great favorite of Maggie Thatcher and of many other conservative leaders -- he showed that the reverse is also true. America and its trading partners should take note!
-- Steve
Steven R. Newcomb, Consultant Coolheads Consulting