OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.


Help: OASIS Mailing Lists Help | MarkMail Help



   RE: RE: [xml-dev] Painful USA Today article (was RE: [xml-dev] AN N: RES

[ Lists Home | Date Index | Thread Index ]

It is the one way to ensure less bloat while enabling 
interoperability by choice rather than fiat.  X3D is working 
out a component profile mechanism for browsers and authors 
to use.  It's a hassle but the experience was that 
optionality by profile is a requirement.  It takes more 
than a schema to spec that.

I hate the thought of tweaking those declarations probably 
only less than those who would have to implement them.  But 
on occasion, it was very good to be able to.  I expect that 
as Rick Jeliffe suggested jovially, at some point, someone or some 
group will go back to SGML and do the subset work again. 

But not today.

My point was, XML is bought based on specific applications 
and their real benefits.  Some of these are small 
system to system communications.  No committees, no requirements 
to scale to the universe; just the qualified parties.  Others 
really are (eg, HTML, SVG) pieces that solve a problem 
everyone has, and for these, expect slower progress, interminable 
haggling, and possibly, that at the other end, only one or 
two vendors can support it (eg, SVG) so no matter what one 
says about the spec process, open systems, and so on, 
the result is that a limited number of sources survive.

I don't know if it has to be that way, but the patterns 
say it is a probable outcome.  So again, it is a question 
of opportunities deferred and having the right reasons.


From: Simon St.Laurent [mailto:simonstl@simonstl.com]

Lean sounds good.  Perhaps it makes sense to require the specifications
themselves to come with an explicit set of checkboxes this way?

That'd mean developers could:
a) see what the options are
b) specify what options they use in a way that will be easily understood
and shared

Hmmm.... maybe we'll get back to the SGML declaration eventually.  Us
open systems ranters could just take the core set minus all the extra
pieces so we didn't have to worry about inconvenient expectations.

I suspect that Working Groups will shudder at the thought of any of
their precious features being listed as expendable this way, but it's
about time for clarity if they aren't willing to throw more things out.


News | XML in Industry | Calendar | XML Registry
Marketplace | Resources | MyXML.org | Sponsors | Privacy Statement

Copyright 2001 XML.org. This site is hosted by OASIS