W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > August 2001

Re: RDF Issues from MusicBrainz

From: Graham Klyne <Graham.Klyne@baltimore.com>
Date: Mon, 13 Aug 2001 13:46:53 +0100
Message-Id: <5.1.0.14.2.20010813132904.03a4dec0@joy.songbird.com>
To: Brian McBride <bwm@hplb.hpl.hp.com>
Cc: RDF Core <w3c-rdfcore-wg@w3.org>
At 12:42 PM 8/5/01 +0100, Brian McBride wrote:
>It does occur to me that it might be a good idea to tidy up the list from
>the flip charts and any other contributions such as this folks would like
>to make into a consolidated list.  To do that would require two actions:
>
>   o folks who have knowledge of direct user experience with RDF, to
>     summarize the top issues users had to the list - as Aaron has
>     just done for MusicBrainz

I'm not sure if this counts as "user" experience, but these are the 
RDF-related issues I would have raised in presenting our experimental workw
 ith RDF, which we hope will lead to applications in trust modelling...


Issues in use of RDF
Encoding of facts and rules


* Facts
   subj pred arg2 .
   subj rdf:type monad .
   subj pred (arg2 arg3 ...) .

It's easier to start with facts expressed using n-place predicates 
--s-expressions--;  we've thought about coding these in RDF's triple-basedn
 otation.  Some current thoughts are at:
   http://public.research.mimesweeper.com/RDF/RDFFactsAndRules.html


* Rules
   Need a “handle” to the fact -- a URI:

   Reification can do this
    [ a rdf:Statement ;
       rdf:subject subj ;
       rdf:predicate pred ;
       rdf:object ( arg2 arg3 ... ) ]

   Similar, but not reification:
    [ a Fact ;  functor pred ;
       args ( subj arg2 arg3 ... ) ]

 From discussions held since these notes were made, I think we may do 
better to avoid reification for this purpose, if only to avoid confusion 
over the intent.


* Variables
   Resource “stands for” a variable:
   [ a Variable ; rdfs:label “?x” ]

I think this parallels the "prince" node resolution we discussed:  the 
"variable name" is used for certain kinds of presentation, but has no 
inherent significance (noting that rdfs:label is described as being for 
presentation purposes rather than identification).


Further work
• Capturing provenance: contexts?
• e.g.
   StatedBy( http:/foo.bar.com/doc, { s1(a,b,...)  s2(c,d,...)  ... } )


I don't think any of these issues bear directly on what we are 
contemplating at the present, but suggest future issues for RDF-based 
applications and generic tools.

#g
--


At 12:42 PM 8/5/01 +0100, Brian McBride wrote:
>Hi Aaron,
>
>Thanks for this.  I'm sorry you couldn't stay for the full time.  The short
>presentations did get squeezed into the last 45 mins or so of the meeting.
>
>It does occur to me that it might be a good idea to tidy up the list from
>the flip charts and any other contributions such as this folks would like
>to make into a consolidated list.  To do that would require two actions:
>
>   o folks who have knowledge of direct user experience with RDF, to
>     summarize the top issues users had to the list - as Aaron has
>     just done for MusicBrainz
>
>   o someone to volunteer to make the summary
>
>Brian
>
>
>Aaron Swartz wrote:
> >
> > I did not get to make a presentation on MusicBrainz at the F2F,
> > however I did feel it was important that I share with you the
> > top issues that Rob Kaye, the head of the MusicBrainz project,
> > had while working with RDF:
> >
> >   - Make RDF specs easier to understand
> >
> >     The spec focuses too much on the serialization. The BNF directly
> >     in the spec is useful, but really misleading. The focus of the spec
> >     ought to be more on the graph that is created, rather than the
> >     serialization of that graph.
> >
> >   - Provide more guidelines for creating vocabularies
> >         o What should be specified as part of a vocabulary?
> >         o Provide a sample vocabulary!
> >         o Should/must a vocabulary have an RDF Schema?
> >
> >   - Provide a syntax that isn't as verbose
> >         o It should also be more intuitive. I think Sergey Melnik's
> >           approach seems reasonable:
> >           (http://www-db.stanford.edu/~melnik/rdf/syntax.html)
> >
> >     - Do you have any tips on how to write a vocabulary spec
> >       for inclusion into IETF spec? Drew Streib @ freedb wants to use
> >       MM for CD lookups and he wants to push this through the IETF.
> >       Has this been done before for an RDF vocab?
> >
> > --
> > [ "Aaron Swartz" ; <mailto:me@aaronsw.com> ; <http://www.aaronsw.com/> ]

------------------------------------------------------------
Graham Klyne                    Baltimore Technologies
Strategic Research              Content Security Group
<Graham.Klyne@Baltimore.com>    <http://www.mimesweeper.com>
                                 <http://www.baltimore.com>
------------------------------------------------------------



-----------------------------------------------------------------------------------------------------------------
The information contained in this message is confidential and is intendedf
 or the addressee(s) only.  If you have received this message in error ort
 here are any problems please notify the originator immediately.  The 
unauthorized use, disclosure, copying or alteration of this message is 
strictly forbidden. Baltimore Technologies plc will not be liable for direct, 
special, indirect or consequential damages arising from alteration of thec
 ontents of this message by a third party or as a result of any virus beingp
 assed on.

This footnote confirms that this email message has been swept by 
Baltimore MIMEsweeper for Content Security threats, including
computer viruses.
Received on Monday, 13 August 2001 10:14:37 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:38:44 EDT