xml-dev - RE: Begging the Question (the novel)

[Mike.Champion@SoftwareAG-USA.com]

Title: RE: Begging the Question (the novel)

> -----Original Message-----
> From: Lisa Rein [mailto:lisarein@finetuning.com]
> Sent: Saturday, December 30, 2000 9:04 PM
>
> This kind of attitude kinda frustrates me because it's really an
> outdated perspective to just "not care" about other technologies that
> aren't on your immediate radar in your own little corner of the web
> world (especially when they really are on radar and you are just not
> aware of it).  The whole point of the semantic web (and actually the
> good old fashioned traditional web too) is to start
> connecting all this
> stuff together and to eventually create systems to facilitate these
> connections automatically.

Welll ... there is a school of thought that says that the whole point of the Web is to make our businesses (or research, or whatever) work better, not to build it for its own sake :~)

Some of this stuff we've been debating will help, some of it won't, and different people are laying their bets on different bits and pieces. The Semantic Web may come together, or it may fade away into obscurity.  What one person sees as an "outdated perspective" another might see as "pragmatic conservatism."  So far we have been phenomenally lucky in that URIs, HTTP, HTML, and SGML/XML have proven to be useful for far more purposes than they were originally designed.  There have been relatively few "gotchas" that keep these things from working together, and with other technologies, in the synergy we know as the Internet Boom, and there really has not been that much of a tension between the visionaries and the pragmatists, since everyone has been pretty visionary and things have come together in a nice practical way.

It would be NICE if we can, by inventing new bits of glue (such as Namespaces, XSD and RDF) keep the synergy going so that we create systems to facilitate "semantic" connections automatically.  But there is simply not the track record of success here yet, and not everyone here is ready to bet the farm that it will happen.  We see a lot of people (Paul Tchistopolskii, Simon St. Laurent, Clark Evans, and myself come to mind) who worry about the XML Boom collapsing under the weight of the complexity/confusion/uncertainty of the interactions among the underlying specs and want to pull back a bit. We also see an articulate and visionary group of Semantic Web advocates who want to push forward ever faster.  This Yet Another Namespace Thread has certainly raised frustration levels on both sides, but I think it is very important to recognize that the debates here reflect some fundamental differences in outlook and priority that will only be resolved when we look back on this in 5 or 10 years.

To give one of my favorite bits of Horse Tartare another whack, the W3C (and this list, and the XML "community" as a whole) are basically trying to do two things at once: distill what we do know about how "all this" works into standards and best practices, and to aggressively experiment to learn more about what we don't know.  Both are perfectly reasonable and useful exercises ... but we must work harder to keep in mind which we're discussing at any given time.  Paul started the "don't use URLs in namespaces unless you mean them to be dereferenced" thread in hopes of getting some agreement on "best practices," and much of the criticism has come from those fearing that this would limit their freedom to aggressively experiment.  It is indeed frustrating, but we  pragmatists need to be reminded of the vision thing now and then, and you visionaries need to make sure that your cloud castles rest on a reasonably solid foundation.