Re: What's an "XML Fragment"?

[Paul Grosso <pgrosso at arbortext dot com>]

There is no particular relationship between the use of the word
"fragment" in XSLT and in the Fragment Interchange spec.  The
latter made a point of defining the term (and related terms) in 
a specific way related to fragment interchange.  The former used
the term in a more generic way without considering the earlier
formal use by the Fragment Interchange spec.  (Other specs--the
DOM, the URI spec, and at times XPointer and XLink--have also 
used the term in diverse ways.  Unfortunately, "fragment" is 
one of those words [like "entity", but perhaps even worse]
that everyone likes to use in a slightly different way, and
few think to define it carefully.)

Despite the fact that I'm on the XSL WG, I cannot answer the
detailed questions about what "fragment" should mean in the
XSLT spec, and it would therefore be best for someone more 
familiar with the actual editing of the XSLT spec to respond 
to questions about the use of this "term" (if it is, indeed, a
"term" and not just a use of the english word) in the XSLT spec.

paul

At 10:34 2000 10 09 -0400, John E. Simpson wrote:
>At 08:31 AM 10/09/2000 -0400, Elliotte Rusty Harold wrote:
>>At 4:43 PM -0400 10/8/00, John E. Simpson wrote:
>>>I have no inside information, obviously, but I just read that as a 
>>>reference to XML fragments as defined by the apparently moribund XML 
>>>Fragment Interchange WD (last updated 6/30/99). Although nothing seems to 
>>>be happening with that WD, it was still fairly current as of the time the 
>>>XSLT WD went to Recommendation.
>>>
>>>See:
>>>    http://www.w3.org/TR/WD-xml-fragment#terminology
>>
>>That sounds possible. The big issue with that interpretation is that the 
>>XML Fragment Interchange WD is not listed as a normative or non-normative 
>>reference by the XSLT spec.
>
>Yeah, I know. (Presumably, since the Fragment Interchange never went to 
>Rec, it would have to be non-normative.) After I posted that message last 
>night, I went back to look at the earlier XSLT drafts; Fragment Interchange 
>wasn't referenced from them, either. The only connection I can establish 
>between the two is that Paul Grosso of Arbortext was on both WGs; you might 
>try e-mailing him to ask if the connection is coincidental, or intentional 
>but undocumented.


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