> [Joe English]
> [...]
> > * Build a keywords-based search index, giving higher weight
> > to keywords, emphasized phrases, and stuff in headings
>
> This last one is dubious. I have no way of saying, find me all
> occurences of "Sprint" (as a proper noun, ie. name) in a document
> or set of documents, skipping "sprint" the verb or noun. Obiously,
> "... winning the men's 100m sprint." does not pertain to
> telecommunications or American corporate culture.
This is true. Yet things like Lycos manage to do a decent job
of finding information in spite of that.
> A lot of the things you mention are superficial. They still don't
> scratch the surface of *semantic* tagging of information.
It would be great if the entire Web were fully semantically
tagged, with a rich set of CApH-style topic maps. I keep
thinking of Star Trek, where Captain Picard can say "Computer,
tell me about twentieth century North American telecommunications
networks and track and field events" and the computer gives
him an answer. If the Web were semantically tagged, that
would actually be possible!
But it's a tremendous amount of effort to encode the information
in the first place.
HTML does a not-too-bad job for very little investment.
> > And let's not forget:
> > * Render it on just about any output device, with reasonably
> > good results.
> Whoopie.
> > This last is something that few other text markup languages
> > have been able to accomplish.
>
> To be honest there aren't that many.
?!?!?
Do you consider two or three hundred "not that many"?
--Joe English