W3C home > Mailing lists > Public > www-tag@w3.org > March 2002

Re: section 1, intro, for review

From: Christopher Ferris <chris.ferris@sun.com>
Date: Mon, 18 Mar 2002 10:38:43 -0500
Message-ID: <3C960A03.6010308@sun.com>
To: Chris Lilley <chris@w3.org>, www-tag@w3.org
Interesting. However, I think that without this explanation,
as to why DOM is listed as a technology for designing new
formats, others will be similarly confused.

I think that omitting XML Schema from this list just
because it is also an instance format is unwarranted. I
would suggest that it be added back even if the list
is not intended to be exhaustive.

I am also puzzled by the absence of XML Infoset in the
list.

Cheers,

Chris

Chris Lilley wrote:

> On Monday, 18 March, 2002, 13:48:20, Paul wrote:
> 
> 
>>>as well as technologies for designing new formats (XML, Namespaces,
>>>
> PC> DOM).
> 
> PC> How does DOM permit someone to "design a new format"?
> 
> One of the fascinating things about working on the design of SVG 1.O
> was the realization that there were at least two, non-intersecting
> communities in terms of design.
> 
> One group (the traditional XML/SGML group) who were primarily
> interested in the DTD as the 'real spec' and would phrase questions
> about the prose in terms of "the DTD says blah so the text is wrong
> when it says ....."; they would consider the object model to be a mere
> temporary in-memory representation of the real XML and as incidental,
> or someone elses problem.
> 
> Another group (the dynamic HTML/server-side programming group) who
> were primarily interested in the object model as the 'real spec' and
> would phrase questions about the prose in terms of "object foo
> inherits from blah so the text is wrong when it says ....."; they
> would consider the xml form as a mere temporary serialisation, used to
> ship object models across the net - incidental, or someone elses
> problem.
> 
> So given the experience of designing (the first time, for me) a markup
> language and its object model at the same time, I suggested to Dan
> that DOM be added along with XML.
> 
> Of course, it is possible to design a new format without a DOM, just
> as it is possible without namespaces or indeed without XML.
> 
> PC> And shouldn't XML Schema be in this list?
> 
> Yes. Originally, it was on that list. Trouble was, it was in both
> columns (it can be included in an instance, for example in XForms, and
> also is one of those 'meta' things like XML and Schem,as (languages
> for writing languages).
> 
> So having it on either list (or both) introduced a need for further
> explanation. Since the list is not exhaustyive but illustrative, the
> solution chosen was to remove W3XC XML Schema from the list.... there
> are probably better solutions.
> 
> 
> 
Received on Monday, 18 March 2002 10:39:39 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 26 April 2012 12:47:05 GMT