[
Lists Home |
Date Index |
Thread Index
]
8/5/2002 12:45:00 PM, "Didier PH Martin" <martind@netfolder.com> wrote:
>
>
>My point is to go beyond these two poles and try to make both the
>programs and the XML documents readable. I won't say that as Mike called
>it that the DOM is a space suit but more as a 19 century diving suit> It
>was good at the time taking into account the maturity of XML and the
>knowledge people got about it. Now, its time to move to space suits and
>go beyond the DOM.
I have no trouble at all with that idea. I think I just "celebrated" my
5 year anniversary on the DOM working group, and supposedly the half-life
of computer industry standards is about 5 years, so it is definitely
time to start thinking about XML-API NG. The original DOM use case of
cross-browser scripting is about dead (I hope Mozilla breathes life into it,
but Microsoft doesn't have any interest anymore, and Dubya's antitrust
folks don't exactly inspire fear of anything worse than a wrist slap if
they don't play nicely with the rest of the industry). High-end scriptable
XML authoring tools haven't exactly taken off, and with Word supposedly
going to support any-schema XML editing Real Soon Now, I can't imagine
anyone getting into that business. So, the basic idea of DOM as
an abstract interface into a product's proprietary data structures
may be falling by the wayside. What do others think?
It would be *nice* the .NET and Java XML APIs were at least conceptually
similar, and didn't introduce gratuitous differences just for marketing
purposes (ha ha ha ha ...what a joke ... ). Or maybe it's time
to let lots of flowers bloom -- .NET, the JCP efforts, dom4j, various
vendors' proprietary APIs, etc. etc. -- and come back to an effort to
standardize when end users start to scream again, as they did about
the Dynamic HTML APIs in 1997.
So, starting with the most basic assumptions, is the idea of a
vendor-neutral read/write XML API still viable? I *think* most
besides our benevolent protectors in Redmond would agree,
as long as we're talking about a foundation layer rather than
One True Standard. How about language neutrality? Could a DOM-like
API with a language binding model that allowed NodeLists, etc. to
be mapped onto whatever language-specific interface makes the
most sense find some traction? Finally, who would use such a
beast? Is the XML InfoSet just too low-level to bother trying to
expose to ordinary users? Do most people want to use a data binding tool to
hide the XML as a serialization format, or would a clean API (e.g.,
that was integrated with XPath and XSLT, hid bizarre syntax such as
CDATA sections, and had a sensible model of how text was related
to elements) find some customers?
|
