W3C home > Mailing lists > Public > public-grddl-wg@w3.org > June 2007

#faithful-infoset wordsmithing

From: Jeremy Carroll <jjc@hpl.hp.com>
Date: Tue, 19 Jun 2007 23:02:38 +0100
Message-ID: <4678527E.9000107@hpl.hp.com>
To: GRDDL Working Group <public-grddl-wg@w3.org>


Summary:
- suggested change to mention xmlFunctions-34 in GRDDL spec
- suggested changes in response to Chime's @@s, also concerning the WG
decision to postpone #faithful-infoset

==========

I note that Chime has a number of @@s wondering where to mention the new
resolution. This messages makes some concrete suggestions.

I wonder too whether the decision should have resulted in some other
minor changes to the GRDDL specification text, for example, the new
resolution explicitly mentions tag issue xmlFunctions-34, could this be
worked into the relevant paragraph like this:

1. In section 6, in paragraph 4 change the sentence ending "is
unspecified" to:

i.e.
[[
is unspecified; this specification anticipates that the
resolution of TAG issue
<a href="http://www.w3.org/2001/tag/issues.html?type=1#xmlFunctions-34">
xmlFunctions-34
</a>
will provide further
clarification and guidance.
]]

2. While I like Dan's 'unspecified', I still think it may be helpful to
be a bit clearer, I haven't really understood what was wrong with text I
have suggested before such as, changing the same "is unspecified" to

[[
is as defined in other Recommendations, but otherwise unspecified
]]

(the vagueness about which Recommendations is deliberate: I think we are
more likely to have consensus around standards-driven behaviour, than
about precisely which standards!)


I've separated these two out to aid discussion. Combined, and with the
preceding text they are:
[[
Whether or not processing of XInclude, XML Validity, XML Schema
Validity, XML Signatures or XML Decryption take place
is as defined in other Recommendations, but otherwise unspecified; this
specification anticipates that the
resolution of TAG issue
<a href="http://www.w3.org/2001/tag/issues.html?type=1#xmlFunctions-34">
xmlFunctions-34
</a>
will provide further
clarification and guidance.
]]


3. In the test cases, the main suggested change is at the #xinclude
test, to explicitly mention xmlFunctions-34

a) change the link
<a href=
"http://lists.w3.org/Archives/Public/public-grddl-wg/2007Feb/att-0017/31-grddl-wg-minutes-edited.html#item02">
resolution</a>
regarding <em>faithful infosets</em>
to
http://lists.w3.org/Archives/Public/public-grddl-wg/2007Jun/att-0046/SV_MEETING_TITLE_--_6_Jun_2007.htm#item04

b) Change the sentence beginning "In particular, ... " to the
following:
[[
In particular, the output illustrates a situation where the XML
processor invokes XInclude processing at a low-level and presents the
expanded <a href="http://www.w3.org/TR/xinclude/#processing">result
infoset</a> <a href="#ref-XInclude">[XINCLUDE]</a> to the GRDDL-aware
agent.
Note that most browsers do not perform the XInclude, and
thus their behavior does not correspond with the output shown.  This
pair of tests anticipate that the resolution of TAG issue
<a
href="http://www.w3.org/2001/tag/issues.html?type=1#xmlFunctions-34">
xmlFunctions-34
</a>
will provide further guidance concerning them.
]]

Concerning the note about browsers, personally, one of the things that
has motivated me over a number of years to work on RDFa and then GRDDL
is a better alignment between the Web and the Semantic Web. From this
perspective, I think it is important, that given (X)HTML input, a GRDDL
aware agent behaves similarly to a Web Browser. The note is intended for
readers like me. I realise that some on the WG take a more XML centric
line, and may choose to not take much notice of browser behaviour.


Concerning the other @@s in the test editors draft concerning this, my
suggestions would be:
4. No change:
[[
GRDDL is silent @@reword to indicate intent of postponing
#issue-faithful-infoset?
]]

no rewording needed.

5. Suggest delete:
[[
However, at the time when this version of the GRDDL Test Cases was being
composed, XProc was still in early stages of development and thus not
very testable by GRDDL-aware agents. @@Is this a fair assessment?
]]
I am not sure, I don't think it matters - this text is not necessary. If
questionable text is not necessary, delete it. So my suggestion would be
to delete this sentence

6. Suggest delete:
On
http://www.w3.org/2001/sw/grddl-wg/td/grddl-tests#noxinclude
[[
Note that the unexpanded infoset and its corresponding XPath Data Model
(See: B XML Information Set Mapping [XPATH]) could instead have been
presented to an XProc pipeline with an explicit XInclude component. @@Is
this the kind of XProc usage the spec is advocating and would an
explicit XProc snippet be useful?
]]
Again, I don't think this text is necessary, and as you seem to have
some doubts about it, deleting it is the safer option.
You have mentioned XProc in the introductory blurb, that probably suffices.

============

BTW I believe there is HP consensus that such changes would be an
improvement!

Jeremy (speaking for HP this time!)


-- 
Hewlett-Packard Limited
registered Office: Cain Road, Bracknell, Berks RG12 1HN
Registered No: 690597 England
Received on Tuesday, 19 June 2007 22:02:59 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 14:11:50 GMT