W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > October 2002

Syntax WD - thumbs down

From: Jeremy Carroll <jjc@hplb.hpl.hp.com>
Date: Thu, 31 Oct 2002 10:43:40 +0100
To: <w3c-rdfcore-wg@w3.org>
Message-ID: <BHEGLCKMOHGLGNOKPGHDAEIPCAAA.jjc@hpl.hp.com>

This is the summary report of my review of the editor's draft of the Syntax
doc (revision 1.350).

The WG should not publish this document.
The editor has refused to address my main point.

Hence I propose that the WG:

- actions the editor to delete section 2 "An XML syntax for RDF" (approx 10
pages)
- publish the WD subject to this and other editorial changes at editor's
discretion


===

I would hope that any opposition to this proposal would come from people who
have read section 2. I would hope that any opposition took account of the
large number of very substantial editorial changes proposed by myself and
Brian (I reattach my review copy, the previous version missed two
</span>'s).

In particular if this proposal is rejected I would expect that that means
that a majority of the WG believe that, in the time available, section 2 can
be made into something that will help the reader of the document understand
the substantive content (in sections 5 - 9).
Moreover, the degree of this help must justify the length of the section.

===

I have previously indicated a willingness to compromise on having the
section clearly labelled as informative.

If there is a majority in the WG who, having reviewed the section, believe
this compromise is workable, then I suggest that:
- the section should be moved to being a non-normative appendix
- that the appendix be on a separate html page in a compound document

Moreover, I think that to make it a contribution to the community's
understanding of RDF/XML rather than an obstacle there is still a
substantial amount of editorial work to do - in terms of making the
descriptions clear, correct and concise; and in relating these descriptions
to the substantive section (particularly the productions in section 7). The
quantity of this editorial work, and the shortness of time in which to do
it, suggests that the document needs an additional editor, who I would
expect to emerge from the majority who felt that these examples were worth
the effort. I would hope that such an editor would make efforts to use
declarative descriptions of the examples, rather than the procedural
descriptions that Dave has used.

<<aside:
The deletion of the italian
<rdf:Description rdf:about="http://www.w3.org/TR/rdf-syntax-grammar">
<dc:description xml:lang="it">Il Pagio di Web Fuba</dc:description>

should be required in any case.
It is inappropriately (and untruthfully) boastful (approx: "The coolest web
page")
>>

===

I feel that the best way forward is:

- acceptance of my proposed resolution above
- if the editor believes it is necessary to give examples of use of each
production, that such examples should be added to the test cases.
- in section 7, after each production, a link to the relavant example is
given.

If WG thinks Dave's section 2 could be a contribution to the community, he
should be encouraged to work on the many editorial comments he has already
received, and publish it through some refereed channel - like - if he were
quick - he could submit it for consideration for WWW2003; more realistically
maybe ISWC2003

===

I regret that I will be unable to attend the telecon tomorrow, I trust the
chair will represent my position.

===

There are other points, the important ones Dave has already accepted.

A technical issue that we have not discussed is XML validation, which can
change the infoset over which the grammar operates. I suggest we should have
telecon time on this topic after publication - I could prepare some test
cases.

Jeremy



Received on Thursday, 31 October 2002 04:43:49 EST

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:52:35 EDT