[
Lists Home |
Date Index |
Thread Index
]
On Mon, 3 Feb 2003 09:05:58 -0500, Elliotte Rusty Harold
<elharo@metalab.unc.edu> wrote:
> I don't believe that. I specifically don't believe there are alternative
> syntaxes for the Infoset (of if there are, they're not XML)
I think this is probably the most important meta-issue that the "XML" world
has to face. If a fork occurs, this will probably be the rock that it
forks around, because the W3C has already build two classes of
Recommendations -- those built on the Infoset (DOM [sortof],
XPath/XSLT/XQuery, XSDL, and SOAP) and those built on the syntax. But this
is a permathread, and we've all stated our positions numerous times.
I don't think the "alternative syntaxes aren't XML" position is
demonstrated in the real world, FWIW, whatever its philosophical merits.
The classic example is that you can use DOM in a browser (if you're
careful) irrespective of whether the input document is XHTML, valid HTML,
or tag soup that the browser "fixed up" into an Infoset. Other examples
are the various SAX parsers for non-XML syntaxes that generate documents
that can be transformed with XSLT (see www.wikiml.org for one example, if
memory serves). XQuery is designed to (and empirically can, in a demo that
Dare was showing at XML 2002) query an infoset that is concretely realized
in an SQL database as if it were "XML". If some tool turns a syntax into a
form that DOM and XSLT can work with, it passes the XML "smell test" as far
as I'm concerned. We could argue about whether these are best practices or
good things, but they exist.
>
> I also think its important for there to be a plethora of languages. One
> language does not fit all. Competition is good. I do not believe in the
> centralized "There can be only one" philosophy.
I certainly agree with this! To fork, or to accomodate, should be made on
pragmatic rather than philosophical or marketing grounds. If the data-
markup-language and the text-markup-language don't really share all that
much, they should fork. But let's not do this lightly... remember that
long list of things one gets "for free" with XML because it is universally
supported. If half (ahem, more like 90%) of the vendors and customers fork
off to something new, hard-core XML people will long for the good old days
when the data-oriented people were whining about DTDs and PIs rather than
ignoring XML completely. :-)
|
