[
Lists Home |
Date Index |
Thread Index
]
At 12:38 PM 11/6/2002 -0500, Simon St.Laurent wrote:
>The recent Character Entities discussion has been enlightening on a
>number of levels, but what struck me most interesting is that it's the
>first discussion we've had in a while about whether XML was really
>missing something.
>
>Looking at the three original aspects of XML (syntax, linking, styling),
>it looks like they may be close to done.
When you ask if I think XML is complete, my answer depends a great deal on
what is meant by XML. I think you have correctly described the original XML
vision. In this vision, I think of styling as a way of using XML, not as
part of XML itself.
And if you use "XML" to mean something that goes beyond the definition of
XML per se, then at least XQuery and SOAP are activities that may
significantly change the way people use XML. There will be others.
>There is ongoing work on XML, but I think it's fair to characterize both
>XML 1.1 and Namespaces in XML 1.1 primarily as efforts to clean up
>outstanding issues (Unicode tracking, excess namespace declarations)
>while possibly sliding something new in (NEL, IRIs) rather than any kind
>of new structure or significant modification.
Sure.
>XML Schema, so far as I can tell, has exhausted its participants and I
>haven't heard much talk on that front. Critics seem to have moved
>elsewhere, and supporters occasionally talk about trimming it, but
>there's not much activity there.
I don't think the last word has been said here.
Clearly, W3C XML Schema has become widespread, and I don't think we'll have
a world without W3C XML Schema in the foreseeable future.
I would like to see RELAX-NG + XML Schema DataTypes supported as an equal
partner in XML - and yes, I mean that I would like to see the W3C
officially accept it as such. It clearly hits the sweet spot for a lot of
applications, and it is much easier for small companies, individuals, or
research groups to implement.
To me, Schematron and schema extensibility mechanisms point to other
promising areas that have not yet become part of our everyday practice as
they may in the future. Given the fact that it took object oriented
programming from 1967 (the dawn of Simula 67) until roughly 1991 to become
commonplace, and has now become a dominant paradigm, I assume some things
take time to catch on. It's quite possible that we may find some of these
areas make a lot more sense to us after we've become comfortable with the
basics. Of course, it's hard to call the winners until they win.
>XLink is finished, though hardly anyone uses it, and XPointer is
>apparently in a dash to the finish before the WG expires on 31 December.
>Whether XPointer will find more use than XLink thus far remains to be
>seen.
I think the world does need pointers among documents, and I hope that
XPointer achieves more widespread use. If it does not, then we have a
problem when we want to have references among XML documents, and I would
have to conclude that XML is not done.
>XSLT is certainly growing in its 2.0 development process, but it's far
>from clear that there's strong community support for more than a few
>pieces of it. XSLT could use a cleanup/minor addition process, but that
>doesn't seem particularly likely to happen, at least given current W3C
>activity.
I'll let XSLT people comment on this. I generally use XSLT for relatively
simple things that are easily satisfied by XSLT 1.0.
>Are there more issues like Character Entities waiting to surface? Or
>(apart from that issue) can XML declare victory and call itself
>complete?
If you had asked this question a year ago, I suspect many people would have
assumed there were no issues like Character Entities waiting to surface.
Ask us again in a year, and we'll tell you what outstanding issues have
surfaced ;->
So I think I've concluded the following:
1. XML 1.0 by itself is quite useful, and is all many people need. XML 1.0
is long since done.
2. There are ongoing fixes and minor enhancements, and probably will
continue to be.
3. I don't think we're done working on schema languages yet.
4. I hope XPointer is widely adopted; if not, we may need to find something
else.
3. People will be exploring different ways to use or enhance XML for a good
long time. This does not require changes to XML per se. Querying,
stylesheets, SOAP, etc. are examples of this.
Cheers,
Jonathan
|
