W3C home > Mailing lists > Public > public-rif-wg@w3.org > April 2008

Re: CURIE proposal ...

From: Axel Polleres <axel.polleres@deri.org>
Date: Tue, 22 Apr 2008 16:30:25 +0100
Message-ID: <480E0491.5070902@deri.org>
To: Michael Kifer <kifer@cs.sunysb.edu>
CC: "Public-Rif-Wg (E-mail)" <public-rif-wg@w3.org>

Michael Kifer wrote:
> Axel wrote:
>> Michael Kifer wrote:
>>> Jos wrote:
>>>> Therefore, I would suggest to adopt the following suggestions by Axel:
>>>>
>>>>> 1) We use UNQUOTED prefix:ncname to denote CURIEs which expand to QUOTED 
>>>>> IRIs
>>>>  > 4) For symbol space IRIs (i.e. IRIs after the ^^) we only allow eithr
>>>>  > the unquoted prefix:ncname writing or the angle bracketted name.
>>>>
>>>> i.e.,
>>>> mailto:chris = "http://....#chris"^^rif:iri = 
>>>> "http://....#chris"^^http://.....#iri
>>>>
>>>> CURIES should *not* be allowed as the first part of a constant name, 
>>>> i.e., my:curie^^rif:iri should not be allowed.
>>> I really dislike this particular proposal. It appears to me as an ugly
>>> hack. 
>>  >
>>> In some contexts a macro would expand into just a concatenation and
>>> in other contexts into a concatenation plus ^^rif:iri. Yuck!
>>  >
>>> I am more sympathetic to shortcuts where macro-expansion is well-defined.
>>> For example, in unquoted contexts and inside <...> (which can be used as
>>> a shortcut for urls). Then you can have
>>> <mailto:cris> = "http://....#chris"^^rif:iri = "http://....#chris"^^http://.....#iri
>> this (CURIEs within angle brackets) is against the conventions used in
>> Turtle (it resembles XML though, but there a prefix:ncname pair is a 
>> QName and not a CURIE).
> 
> Exactly. They use QNames and we need curies. There is too much confusion in
> people's minds, and we should not further that. It was bad enough to use
> the same symbol : for both qnames and curies. (I was arguing to use a
> different syntax, as in SWSL and WSMO.)
> 
>> As for a generic macro definition mechanism XML already offers entity 
>> references.
> 
> You are confusing things again. We are talking about the ***PRESENTATION
> SYNTAX***. Not about XML! In XML we decided to not use Curies. In fact,
> I just realized that all our XML examples are wrong in that respect: we
> need to change them to use entities.

yes.

>> To me it appears that there is a mess already: XML is different from 
>> Turtle is different from RDF/XML... All I intended to propose was 
>> adopting ONE of them (Turtle) without compromises, instead of inventing 
>> yet another one.
> 
> No, you are proposing to import that mess into RIF.

I am really not. I agree that mixing up things is a mess. I am entirely 
happy without a generic macro mechanism and reusing the Turtle way of 
CURIEs. I just said, if you want another, more generic macro mechanism, 
XML entities are already there and we might consider reusing them.

Axel


-- 
Dr. Axel Polleres, Digital Enterprise Research Institute (DERI)
email: axel.polleres@deri.org  url: http://www.polleres.net/

rdfs:Resource owl:differentFrom xsd:anyURI .
Received on Tuesday, 22 April 2008 15:31:28 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 2 June 2009 18:33:48 GMT