W3C home > Mailing lists > Public > public-lod@w3.org > March 2010

Re: Should dbpedia have stuff in that is not from wikipedia - was: Re: A URI(Web ID) for the semantic web community as a foaf:Group

From: Bernard Vatant <bernard.vatant@mondeca.com>
Date: Tue, 30 Mar 2010 21:40:55 +0200
Message-ID: <9d93ef961003301240s1fca9402i357ca1f17b5f545@mail.gmail.com>
To: Hugh Glaser <hg@ecs.soton.ac.uk>
Cc: Kingsley Idehen <kidehen@openlinksw.com>, Tom Heath <tom.heath@talis.com>, "KangHao Lu (Kenny)" <kennyluck@csail.mit.edu>, "public-lod@w3.org" <public-lod@w3.org>
Hi Hugh and all

... skipping Kingsley-related stuff :)

>   .... This is an interest to me because there is a whole load of other
> stuff that
> appears under the dbpedia banner, mostly concerned with sameAs with other
> resources (some of which I disagree with).

Pat Hayes and Harry Halpin have a nice paper for LDOW 2010 about use and
abuse of owl:sameAs

> I think that most people who use dbpedia are using it on the basis that
> what
> they get from dbpedia is a reflection (for good or bad, of course) of the
> contents of wikipedia infoboxes and whatever else the dbpedia team have
> managed to glean from the site.

I would say translation or re-presentation rather than reflection.
I've expanded on this notion of translation a few months ago ...
Wikipedia content itself is the result of a long chain of re-presentations
of knowledge. dbpedia is yet another another step in the translation
re-presentation of knowledge. There is a lot of added value even it's the
"same content" (whatever that means). Interpreting fields in the infobox,
expliciting their semantics, is not a simple "reflection". There is added
value, there is re-interpretaion in terms of ontologies that have not been
invented in Wikipedia, alignments of "equivalent" fields etc. And linking to
other representations is certainly part of the process.

Adding other stuff, for whatever reason, complicates the trust and
> provenance of the source.
> Exactly what is the provenance of resolving a dbpedia URI?
> Well, it is a subset of the wikipedia information, plus possibly a chunk
> more.

Indeed, but the same for anything produced by human intelligence. It's bits
of the legacy plus a chunk more. Dwarves on giant's shoulders etc.

I think that dbpedia (all praise to its amazing achievement) should restrict
> itself to publishing exactly and only what it has gleaned from wikipedia,
> and any other stuff should be published elsewhere.

IMHO "exactly and only" can't make any sense here. There is no explicit
semantics in WP, and there is in DBpedia.


Bernard Vatant
Senior Consultant
Vocabulary & Data Engineering
Tel:       +33 (0) 971 488 459
Mail:     bernard.vatant@mondeca.com
3, cité Nollez 75018 Paris France
Web:    http://www.mondeca.com
Blog:    http://mondeca.wordpress.com
Received on Tuesday, 30 March 2010 19:41:33 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:16:04 UTC