RE: XSL as a better XPointer

Subject: RE: XSL as a better XPointer
From: "Didier PH Martin" <martind@xxxxxxxxxxxxx>
Date: Sun, 11 Apr 1999 13:34:08 -0400
Hi Micah,

I think that the problem could be more political than practical. The two
specs are developed by two groups and there is probably no mechanism in
place to re-unite these worlds. I may be wrong, and I wish I am.

This said, Yes both (Xpointers and XSL patterns) should be at least
identical or one a subset of the other. In fact, you are right to say that
there is a lot of similarities. Both organize their world with a name space.
In that case, a hierarchical name space. This would make sense that we have,
for documents organization a common name space convention. This is simple
practical application of good knowledge management. We re-use the knowledge
learned. I'll repeat to be clear, the main advantage of a common name space
convention is knowledge re-use. Thus, if someone learned how to create a XSL
query (i.e. a match in XSL) he or she learned at the same time a Xpointer
convention. XPointers could express more complex relationship like document
to document linkage but the basic rules would be the same.

Now let's go a bit further. Why can we not use also the full XPointer name
space in XSL? A XSL script would then be able to process multiple documents
and build an output document form multiple source. Why not? I have even some
concrete needs for this.

So why these world with so much similarities are not unified? I have also
the same question on some Win32 API. Why do I have different API for similar
things. Simple answer, because different group created them. And each group
creates its own world. If W3C resolve this political issue, we may have some
chances to have some uniformity. I do not say that both group do not do
their job, This is in fact totally the contrary. I am saying that if there
is no mechanism in place to get uniformity of name spaces, there won't be
any.

When working on the design of XScripts I had to create a scheme for name
space convention I turned to W3C specs and discovered a certain
schizophrenic pattern here. If I take XPointers as name space I am loosing
XSL name space convention. If I take XSL name space convention I am loosing
XPointer convention. If I unify both world I get excommunicated from W3C as
not respecting the specs. Is this what we call a catch 22? How to resolve
it? Simple decision made by W3C: to have a group or have both group work
with the same name space convention. Am I talking of utopia here maybe? I
just hope that some people in W3C look at the problem and try to resolve it.

Regards
Didier PH Martin
mailto:martind@xxxxxxxxxxxxx
http://www.netfolder.com

-----Original Message-----
From: owner-xsl-list@xxxxxxxxxxxxxxxx
[mailto:owner-xsl-list@xxxxxxxxxxxxxxxx]On Behalf Of Micah Dubinko
Sent: Sunday, April 11, 1999 12:25 PM
To: xsl-list@xxxxxxxxxxxxxxxx
Cc: Steven_DeRose@xxxxxxxxx
Subject: XSL as a better XPointer


Hello all,

Let me jump in to point out that there are many conceptual similarities
between XSL patterns and XPointer. Also, the sky is blue. :-)

My point is, it would be a shame to have two completely different specs
with so much overlap. It would be a burden not only on the spec writers,
but also on every implementer down the line.

>From what I've heard, the main arguments against this sort of thing are
'XSL patterns are too simple' and 'XPointer is too complex'.

So, how about 'XPointer Level 1' for the XSL spec and 'XPointer Level 2'
for the heavy lifting?

Jonathan Borden wrote:
>         Or in XSL patterns:
>
>         //chapter[21]/v[12]
>
>         XSL patterns are a better XPointer (less weight more filling...)

Comments?

.micah


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


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


Current Thread