[no subject]

From: Chris Maden <crism@xxxxxxxxxxx>
Date: Thu, 4 Mar 1999 18:40:14 -0500 (EST)
[John McClure]
> Microsoft is one of those that implements XSL as (just!) a
> transformation language, with the apparent idea that the formatting
> objects in HTML are quite adequate for the time being. "For the time
> being", I agree with this approach also, and would like to see the
> specs track to that, renaming XSL into something more descriptive of
> its current transformation function.

Renaming a draft specification because of how it is currently being
used is silly.  XSL has a very distinct, clear purpose.  If it is not
meeting that purpose, then the spec should be changed, not the
purpose.

> I am waiting to see what CSS3 is to offer, expecting stylesheets to
> apply to XML elements in addition to HTML elements, so that when
> fully implemented (particularly all the "display" types) then we can
> dispense with the HTML transformation altogether.

CSS2 is applicable to XML, and both MSIE and Mozilla (as well as other
browsers, like MultiDocZilla) have implemented it (though not all of
CSS2).  I'm not particularly eager to see CSS3 until at least one of
the Big Two finishes implementing CSS1.

But I think you'll find that for any non-trivial document, you need
<term id="xform">transformation</term>.  There have been proposals to
do this with CSS syntax, but whatever the solution, <xref
linkend="xform"/> is necessary.

> In the meantime, we'll continue with the messy business of
> converting edited HTML (back) into XML when it is to be pushed to a
> server.

Ew!  Why are you modifying the HTML when you have XML source?  That's
a process problem, not a technology one.

> FYI, ISO's DSSSL spec splits manipulation apart from formatting
> objects, and Scheme stood apart already.

Not really.  There is a DSSSL tree transformation language, but the
basic expression language includes node lists as primitives and
defines operations on them.  Without that, styling would be nearly
impossible.  What XSL lacks as of the most recent draft is an
expression language.

The very idea of the XSL transformation syntax came from the
observation that the so-called styling portion of DSSSL, together with
simple 'element flow objects', could perform very powerful
transformations.  Since transformation of this sort is necessary for
styling, if the styling operation is expressed as transformation into
a formatting document type, then a potentially powerful transformation
language falls out of the process for free.

> IMHO, XSL as it stands is really a dumbed down Scheme for the
> unwashed non-IEEE masses, and that's OK, it's just that the DSSSL
> purists out there (for whom I have nothing but the greatest respect)
> I sense are feeling a bit betrayed by the process they're in with
> the W3C and its primary members, and are trying to force the issues
> with this poll.

Which DSSSL purists would those be?  The three editors of DSSSL (the
co-chair of the XSL WG, her alternate on the WG, or the XSL spec
co-editor)?  Jon Bosak, editor of dsssl-o, DSSSL evangelist, and
initial convenor of the XSL WG?  Or me?

I like DSSSL.  I like Lisp; I like parentheses.  But I've been
convinced of the political argument that nearly no one on the Web
would be willing to accept Lisp syntax, and the technical argument
that a completely new parser isn't necessary if XML syntax is used
(you still need a tokenizer for patterns and such, but note that the
original submission didn't even have those).

> Anyway, my vote is actually to simply second James Clark's view of
> the matter, whatever that may be, since he has to-date been right on
> target.  The entire community owes James and his compatriots great
> gratitude for sticking in there, getting the crucial elements of
> DSSSL through to the W3C knuckleheads.

Statements like this confuse me.  Exactly which knuckleheads are
those?  I'm not sure against whose authority the "compatriots" are
rebelling; as far as I can tell, the WG discusses issues, and either
reaches consensus or takes a vote.  I haven't noticed any spy cameras,
cattle prodes, or mind probes; maybe they're just very, very good...

-Chris
 definitely not speaking for the XSL WG on this one
-- 
<!NOTATION SGML.Geek PUBLIC "-//Anonymous//NOTATION SGML Geek//EN">
<!ENTITY crism PUBLIC "-//O'Reilly//NONSGML Christopher R. Maden//EN"
"<URL>http://www.oreilly.com/people/staff/crism/ <TEL>+1.617.499.7487
<USMAIL>90 Sherman Street, Cambridge, MA 02140 USA" NDATA SGML.Geek>


 DSSSList info and archive:  http://www.mulberrytech.com/dsssl/dssslist


Current Thread