Doubts regarding XSL and DOM

[keshlam@xxxxxxxxxx]

>DOM is an API used by a prog. lang/application, XSL is neither <...>

We're _almost_ in synch. XSL is an application. As such, the question of
how it models XML internally isn't directly relevant; you're talking to XSL
rather than to the model.

>I feel it would be proper to have "DOM-like read-only access" methods as
>they would leverage the XSL pattern syntax by giving them a DOM like feel.

Think of the DOM as a useful subroutine library. That's all it is.

The "feel" of the pattern syntax doesn't particularly have to be DOM-like.
Where the two are discussing similar things, there are definitely
advantages to using similar words -- but that mostly means that both XSL
and DOM should track the XML terminology, when it applies.

The syntax is already very different (XSL patterns are based on the
XML-Query proposal, right?), and for good reason: a query language can
express things that the underlying representation may not directly support,
and is oriented toward some combination of terseness and clarity of
expression rather than being anchored in a programming-language-like
syntax.

Of course _if_ the DOM adds richer querying mechanisms next time around,
and _if_ the XML-Query syntax happens to be directly representable in terms
of those additions, then you might in fact wind up simply passing your
query string to the DOM... in which case the XSL syntax would arguably
become part of the DOM syntax, rather than vice versa. XSL is certainly
going to be a strong influence on any DOM query mechanisms, but until and
unless the DOM WG announces a direction we won't know exactly how strong.
______________________________________
Joe Kesselman  / IBM Research
Unless stated otherwise, all opinions are solely those of the author.



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