the time for XSLT to rally around XHTML is now.... was RE: [xsl] xsl/xslt coding standard

["James Fuller" <james.fuller@xxxxxxxxxx>]

just a general comment about XML.....it does have a component to our current
doc thread

I think that a lot of current issues, especially those discussed on other
xml lists ( xml-dev )come from having very little in the way of constraints.
Take the simple idea of having a standard documenting....format.

the simple steps to confusion

a) The goal is to be able to comment within stylesheets your code, and
possibly other bits of useful meta data like author, version, etc

b) a review of 'whats out there' reveals a many headed hydra of markup
languages that all overlap with the simple requirement of just documenting
our lowly xsl templates

c)since we know we will be interoperating with many different xml
vocabularies, we would like to have the same idiom within other
markups.....so being good little bunnies we immeditaly expand our
scope...which I think is an anti-pattern within XML, this immediate scope
expansion because we can ( its easy in XML ), for our case an XML document
markup that survives not only in XSLT but allows for any document markup
within itself is a bit over the top ( compared to initial requirement )

d) so now we are stuck in a situation where through scope expansion ( btw
which was my fault...I think Jeni simply wanted a nice xsl:doc element )
either creates the 'thread of 1000 parts' or where no one can agree...and a
recent review of the new XSLT 2.0 ( btw nice job to the editor and those
involved, its a good example of a 'living document' reflecting all the
issues that XSLT is experiencing ) shows the complexity that can arise when
interacting with other key xml technologies.

In the past web world, we had HTML and we tried to shoehorn it everywhere
( the browser wars which ensued added much pain to satisfying compatibility
between browsers ) to fit all our requirements; much to the chagrin | humour
of SGML'ers worldwide. With XML now we can deconstruct existing successful
HTML idioms and repopulate the world with all sorts of markup, creating a
frightening amount of overlapping specifications and redundancies; not to
mention introducing a mixed processing model of using markup in situations
that historically programming existed !

lets look at XLink as an example of something that was a 'simple no brainer
fundamental' success factor of the web' changing into a experiment that has
failed miserably...I fault no one in the XLink effort, I think that people
need to learn from one place to another, and the distance between an <a
href=""> to that proposed by XLink was too large. I think when you decouple
something you fundamentally remove the reason why it was successful...in our
example <a href=""> linking is successful because in HTML when I refer to a
link...it refers to either an HTML document, URI or resource ( image); all
URI's no doubt ( no flames pls ).

XLink's resolution or scope is far to large with little to no coupling other
then to xml itself ! The idea, within new XHTML, that everything can have an
href will be fundamentally more compelling then any XLink usage.....which
goes some way to saying coupling can be a good thing.

back to our documentation example, why come up with something new for doc
XSLT....XHTML is robust enough esp with modularization to be used for our
documentation purposes, whilst being able to contain other markups and
extended if need be.

I am surprised by the lack of XSLT usage/adoption/articles and XHTML...maybe
this is because most people have constrained XHTML as just a reformulation
of HTML...which of course it is not.

cheers, jim fuller





cheers, jim fuller


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