Re: rdf-text telecon agenda

On Tue, May 26, 2009 at 6:35 PM, Sandro Hawke <sandro@w3.org> wrote:
>
> Alan, Eric, and I just talked through the status of this effort, and
> came up with an agenda based on trying to make sure we're all in
> agreement on the things we think we're in agreement on, and then going
> on to share ideas about the remaining issue(s).
>
>   Changed IRC channel: #owl
>   Teleconference code: OWL1  (6951)
>   10am Eastern  Time
>
> This agenda has lots of PROPOSED resolutions.  This group has no formal
> decision-making authority, so these resolutions should be understood as
> reflecting consensus among the people who attend.  Brief +1 / -1
> response via e-mail could be helpful, too.
>
> 1.  set of language tags
>
>    BCP-47 vs RFC-3066
>    Clarify that we're interpreting RDF Concepts as
>        linking to 3066 or it successor.
>    double-check with addison?
>
>    PROPOSED: We understand that when RDF Concepts referred to RFC
>    3066 it really meanted "RFC 3066 or its successor" (which is
>    currently BCP-47).  We'll add a note to this effect to this spec.
+1
>
> 2.  change of name of datatype
>
>    PROPOSED: The datatype previously known as rdf:text should be
>    called rdf:PlainLiteral
>
maybe rdf:plainLiteral? just follow the example like xsd:unsignedLong.
But I don't have a strong opinion for P/p, and is happy with whatever
the majority decide.

> 3.  change of title
>
>    PROPOSED: The title will no longer mention i18n.  It will be
>    something more like: A Datatype for RDF Plain Literals
>
+1

> 4.  how much of i18n stuff to remove?
>
>    PROPOSED: Pending approval from Michael Sperberg-McQueen, we'll
>    remove the 3rd intro paragraph (from LC version).  It talks about
>    xml:lang, etc.
>
+1
>    (That paragraph was expanded in response to Michael
>    Sperberg-McQueen's comment.  Assuming Michael is okay, we'll just
>    drop that paragraph.)
>
> 5.  action: we need a new Abstract
>
+1

> 6.  plain literals without language tags
>
>    PROPOSED: rdf:PlainLiterals will map 1-1 to RDF Plain Literals, so
>    Plain Literals with and without language are both handled by
>    rdf:PlainLiteral.
>
0 (need to think more)

> 7.  backward-compatibility goal
>
>    This spec is not asking anyone to change their RDF implementation.
>    We're not adding market pressure to add the d-entailment.  RDF
>    folks can freely ignore this spec, without harm.
>
>    PROPOSED: The spec will be clear that while this spec formally
>    specifies an XML Schema datatype, we do not promote or suggest or
>    pressure RDF or SPARQL software or data to be modified to
>    support/use this datatype.
>
+1 (if this will reduce the main tension between this doc and RDF/SPARQL people)

> 8.  interoperability goal
>
>    PROPOSED: We'll say something about how rdf:PlainLiteral typed
>    literals are not supposed to to leak out and break the
>    backward-compatibility goal.
>
+1 (I believe it follows from 7)

> 9.  How to meet the interoperability goal...?
>
>  .. brainstorming, sharing ideas, etc ...
>
>     * Pat's approach using RDF'
>
>    Status of Table 3?
>
>    What do we say specifically about SPARQL?
>
>      - it shouldn't be be in the queried graph (but this this
>        isn't about SPARQL)
>      - it shouldn't be in the BGP
>      - it shouldn't be in a filter
>               STR("foo@"^^rdf:PlainLiteral), LANG( ), DATATYPE( )
>      - it shouldn't be in CONSTRUCT
>
>
>
>
>
>



-- 
Jie Bao
http://www.cs.rpi.edu/~baojie

Received on Wednesday, 27 May 2009 05:42:13 UTC