W3C home > Mailing lists > Public > public-lod@w3.org > June 2008

Re: The king is dressed in void

From: Peter Ansell <ansell.peter@gmail.com>
Date: Fri, 13 Jun 2008 08:19:07 +1000
Message-ID: <a1be7e0e0806121519y5d469aew924865990677a621@mail.gmail.com>
To: "Giovanni Tummarello" <giovanni.tummarello@deri.org>
Cc: public-lod@w3.org, "Semantic Web" <semantic-web@w3.org>

I see VOID as going past the need to have a search engine in order to
decide which sparql endpoints you need to use to effectively make
particular queries based on sample graphs provided by either SPARQL
construct queries on the dataset sparql end point or by way of an
example document specifying typical rdf statements that are contained
in a data set.

Your method does not currently enable someone to backtrack to which
endpoints they should use to get more information about someone if
they don't just want to do a text search, or rely on someone pre
indexing billions of rdf statements for them. Linked Data should be
about self-discovery. If someone ever finds a URI used on the semantic
web they can now find a way to a sparql end point with more related
information. Ie, if you found
http://dbpedia.org/resource/Bioinformatics as a URI on the web and
sindice.com had not yet indexed it for you then you could discover it
by chopping off the path and accessing the robots.txt file provided by
that domain, a method currently used by web 1.0 search engine
crawlers. Then you can discover the end_point and a typical example,
along with provenance and subject information by using the void method
(robots.txt->sitemap.xml->void rdf->either SPARQL CONSTRUCT or example
file). Of course, this mechanism will not work directly for purl.org
users.... but it is a good start as far as I can tell for datasets
which utilise their own domains for naming and hosting. Purl.org users
could be supported if you acknolwedge that the real provider is the
one redirected to by the 302 redirect from purl.org and you could
start the discovery process there.

The method of discovering this information,  don't require an extra
level of complexity as the void rdf simply adds the sparql endpoint
information, example information, and provenance information.

I don't see VOID as having more than one class and two or three
properties when it is eventually created.

Class: void:Dataset

Property: void:sparql_end_point, void:example_file, void:example_uri
(which you can use with a SPARQL construct on the
void:sparql_end_point in order to get the same as a void:example_file)

The rest of the descriptions seem to be allowed for by current
vocabularies such as foaf and dc so the actual specification will be
very highly modular and hence easy to implement and agree on IMO.

Cheers,

Peter

2008/6/12 Giovanni Tummarello <giovanni.tummarello@deri.org>:
>
> Wasnt RDF all aabout being self describing?
>
>  if i say "giovanni works in research" .. do i really need a
> vucabolary that says "this rdf contains informations that describe
> what people claim to be working on" that's a suicide. If this is the
> case (which i totally dont believe) then the king is seriously naked
> and there is no hope whatsoever that RDF is going to have any
> relevance (and there i say it)
>
> to find one such file, instead of having to invent agree and markup
> i'd say its much easier to do something like [1] or [2].
> this is not marketing. its a plea to NOT jump on more layers of stuff
> when the previous layers have really to show there value and
> adoptability still. Solve some simple use cases first then jump to the
> more complex one.
>
> Giovanni
>
> [1] http://demo.sindice.com/search?q=*+%3Chttp%3A%2F%2Fwww.w3.org%2F2006%2Fvcard%2Fns%23title%3E+%27research%27&qt=advanced
>
> or http://sindice.com/search?q=http%3A%2F%2Fxmlns.com%2Ffoaf%2F0.1%2Fknows&qv=http%3A%2F%2Frichard.cyganiak.de%2Ffoaf.rdf%23cygri&qt=ifp
>  (documents which contain statements in which someone claims to be
> knowing richard)
>
> [2] http://forum.sindice.com/showthread.php?t=10
>
>
Received on Thursday, 12 June 2008 22:19:43 UTC

This archive was generated by hypermail 2.3.1 : Sunday, 31 March 2013 14:24:16 UTC