Re: The Peace Process: DOM and namespaces...

Subject: Re: The Peace Process: DOM and namespaces...
From: Tyler Baker <tyler@xxxxxxxxxxx>
Date: Wed, 10 Feb 1999 18:53:24 -0500
Tim Bray wrote:

> At 05:32 PM 2/10/99 -0500, Rick Ross wrote:
>
> >This is definitely the problem, but the solution might then lie in not
> >defining an intrinsic mismatch between the first official XSL spec and the
> >existing XML 1.0 and DOM Level 1?
>
> The problem won't go away.  The W3C specs are built in parallel
> process by a bunch of different but overlapping groups of people.  A
> bit of thought shows that if everything was put on hold until
> everything else was finished, you get dead-lock, there are
> circular dependencies.  As a consequence, things lurch forward
> without (so far) getting too far out of sync.  There will be a stable
> DOM spec that does namespaces some number of months after namespaces
> was frozen. This gap is unfortunate but not IMHO avoidable.
>
> Now that I think about it, there is a good chance that there will
> be a stable DOM level 2 *before* there's a stable XSL 1.0!  It's
> hardly consistent to complain that unfinished-standard-A can't
> be used because of the unfished state of standard-B, on which
> it depends.
>
> >As was mentioned on the list the other day, XML 1.0 doesn't support
> >namespaces either - and it is abundantly clear that there is huge
> >controversy within the community about the namespaces in XML recommendation.
>
> I don't buy that.  There is tons of controversy on this mailing list.
> All the leading implementers - I repeat, *all* the leading implementers -
> in the world of XML have either already completed their namespace
> implementation or will very soon.  (Hint: it's not hard.)

It is not hard to support namespaces in an XML parser, but it is far harder to deal with them
at the application level.  Just because all the tools support namespaces, does not mean that
anyone is using them outside of a few niche applications.  Like another list member recently
said "where is the content"?

> >Perhaps it is simply too early to require namespace support in XSL at all?
> >Perhaps, instead, the initial XSL spec should deliberately be designed to
> >mesh well with the existing and prevailing implementations of XML and DOM.
>
> That's a suggestion that is perfectly sane on the face of it; why
> don't you make it formally to the XLS committee?
>
> >DOM Level 2 is a Pandora's box
>
> Really?  I didn't know that.

I think he means that it is huge and still a working draft.  I have been told a lot of the
stuff that makes it huge are optional features, so in that sense it may indeed be a "Pandora's
Box".  OpenDoc largely failed because it was so huge and no one could justify the development
dollars to spend on supporting it.  When you create specs that are huge and take tons of time
to implement, you are indirectly costing a lot of people a lot of money.  After all, someone
has to foot the bill for commercial software unless you expect XML to solely thrive on
freeware in the future.

Tyler


 XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list


Current Thread