[
Lists Home |
Date Index |
Thread Index
]
- From: Len Bullard <cbullard@hiwaay.net>
- To: "Simon St.Laurent" <simonstl@simonstl.com>
- Date: Tue, 15 Feb 2000 08:32:49 -0600
At 09:29 PM 2/12/00 -0500, Bill la Forge wrote:
> >Anyway, I strongly believe that as XML grows, we will be dealing with a
> >larger and larger audience. And the W3C specs aren't going to make life
> >easy. Perhaps, in the long run, much of that work will be (largely) ignored.
Simon St.Laurent wrote:
> I hope for and expect to see subsets of functionality - XML 80/20'd the
> complexity of SGML, and the XML family of specs will probably get 80/20'd
> in real use. Hopefully the W3C will write specs that accomodate such
> subsetting.
If history is the guide, it will be simplified by the authors of
articles
and books on the subject. The example code gets cobbled in the context
of the implementations vendors provide. IOW, on any given day, I learn
more about web application building from trolling the MSDN and knowledge
base than I do the specs. Certainly, that may not be the most in depth
way, but it gets the work done. That is one level of the food chain.
At the next level, as I review Eliot's excellent grove examples for
VRML, I attain another level of understanding of
how hard it is to use XML to cleanly define a language
where the implementation and semantics are tied to english text
descriptions and many assumptions are not. Schemas aren't much
help here because XML is just an alternative encoding for
the VRML200x spec, and worse, considered only an extension in
the MPEG plans. It gets murky fast and meanwhile, a battle
of titanic proportions between multinationals for control
of real time 3D is heating up.
We really do need a universal framework of definitions. I don't
think the consortia will provide this. We may have to do it ourselves.
Given that, it could be a good thing to ask ourselves if groves are
the place to begin given that it was designed precisely because
the pre-XML community in markup faced exactly the same problems
and were forced to create a solution. OASIS can be a force for
progress here if it can adapt to the open list free for all.
If it requires a membership to post, while I understand the
need for income, I don't think it a good plan. See Cluetrain.
Otherwise, back to the MSDN.
Good to see the list back up and running. What went wrong?
len
|