W3C home > Mailing lists > Public > www-xml-linking-comments@w3.org > July to September 2002

RE: XPointer architecture (Re: [xml-dev] XPointer and XML Schema)

From: Eric van der Vlist <vdv@dyomedea.com>
Date: 09 Aug 2002 18:35:00 +0200
To: Ronald Daniel <rdaniel@interwoven.com>
Cc: xml-dev@lists.xml.org, www-xml-linking-comments@w3.org
Message-Id: <1028910901.7741.305.camel@ibook>

Hi Ron,

Actually, in my message [1], the point I wanted to make was twofold and
I think that XPointer should either:

1) Be less permissive and include a schema() part 
2) or be more permissive and let people choose whatever algorithm they
like to determine how ids should be identified instead of imposing to
use a DTD or a W3C XML Schema.

Your answer excludes the first alternative and I can aggree on this but
think that in this case, the usage of a DTD or a W3C XML Schema should
be non normative examples of technologies which may be used.

Otherwise, I don't see the logic of imposing these 2 technologies if the
schema to use kept to the choice of each application.

Sorry for the delay in my answer but I have been far from my emails for
a while.

Eric 

[1]http://lists.w3.org/Archives/Public/www-xml-linking-comments/2002JulSep/0029.html

On Tue, 2002-07-30 at 20:12, Ronald Daniel wrote:
> Hi Eric,
> 
> The XML Linking Working Group discussed your message [1]
> in our conf. call of 2002-07-17. (My apology for the
> delay in reply, I've been tardy in writing this up.)
> 
> As Henry Thompson stated in his personal response, the
> group does not think a schema() pointer part is required,
> although it is a very interesting idea.
> 
> So why is it not required?
> 
> With the xpointer() scheme, (and other schemes in the
> XPointer Framework), there are many ways that a reader
> can construct a link to a portion of a document. They
> do not have to rely only on the presence of ID attributes.
> (For example - if we wanted to identify the element bearing
> a 'foo' attribute with a value of 'bar', we could override
> the declaration of 'foo' to make it an ID. But we could do
> other things, such as use basic XPath predicates, as in:
> 
>   http://www.example.org/some.xml#xpointer(//*[@foo="bar"]) 
> 
> While there are many options for a document reader to identify
> parts of a XML document, there is essentially only one way
> for a document creator to indicate an element as a likely
> target of a link. They must provide an attribute which has
> somehow been declared as an ID. Because there are so few ways
> for the document creator to express that intent, and because
> the document readers do not actually *require* a schema override
> in order to address any part of a document they might wish, it
> seems best not to rush to define such an override. Given that
> the linking WG's charter expires at the end of the year, we
> believe it is best not to try to define such a thing at this
> time.
> 
> Please let us know if you find this answer acceptable, or
> if you think we really do need to define such a thing within
> the lifetime of this group.
> 
> Best regards,
> 
> Ron Daniel Jr.
> (acting chair of the XML Linking WG)
> Standards Architect
> Interwoven, Inc.
> 803 11'th Ave.
> Sunnyvale, CA, USA  94089
> Tel: 408 530 5922
> Cell: 925 368 8371 
> Email: rdaniel@interwoven.com
> 
> Visit www.interwoven.com
> The Leader in Enterprise Content Management
> 
> 
-- 
See you in San Diego.
                               http://conferences.oreillynet.com/os2002/
------------------------------------------------------------------------
Eric van der Vlist       http://xmlfr.org            http://dyomedea.com
(W3C) XML Schema ISBN:0-596-00252-1 http://oreilly.com/catalog/xmlschema
------------------------------------------------------------------------
Received on Friday, 9 August 2002 12:35:34 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 October 2009 08:39:44 GMT