W3C home > Mailing lists > Public > semantic-web@w3.org > September 2009

Re: tweet2rdf vocabulary convergence

From: Joshua Shinavier <parcour@gmail.com>
Date: Tue, 29 Sep 2009 00:46:41 -0400
Message-ID: <151f299b0909282146w3651702fp1270196f7f4c4692@mail.gmail.com>
To: bnowack@semsol.com
Cc: Semantic Web <semantic-web@w3.org>
On Mon, Sep 28, 2009 at 5:35 AM, Benjamin Nowack <bnowack@semsol.com> wrote:
>
> Hi,
>
> Morton Swimmer suggested that there might be broader interest to talk
> a bit about RDF extracted from tweets, so here we go:
[...]

Hi everyone,

Great discussion, and very timely for me, too (I'm building a service,
called TwitLogic, for real-time search on structured data harvested
via nano-,picoformats and the Twitter streaming API -- paper and demo
coming soon).  I'd like to add a note on representing data *embedded
in* tweets, as opposed to data strictly *about* tweets.

As I see it, there are a few distinct kinds of structured data to be
gotten from Twitter:
1) Twitter-generated data about users
2) Twitter-generated data about tweets
3) user-embedded data about tweets
4) user-embedded data about everything else

Most of this thread has been about bits of vocabulary for expressing
#1 and (especially) #2: authoritative information about Twitter users
and individual microblog posts provided by Twitter through its API,
RSS feeds, etc.  #3 is also about individual microblog posts, but
requires some trust in the user who embeds the data and the parser
which interprets it.  #4 is data about *anything* -- the location of
an airport, whether or not you like Obama, the fact that @romeo loves
@juliet, etc. -- expressed in general-purpose formats like Twitter
Data and MicroTurtle.

So... I'd like to suggest that this vocab cloud should also include a
property for the relationship between user-generated information and
the tweets it is derived from: something like the Web of Belief
ontology's "hasSourceDocument" (which relates a graph reference with a
Semantic Web document) but with a range of sioc:Post, foaf:Document or
such, so an application like mine can link up user-generated
descriptions with tweets described by some of these other services.
Having looked through a few provenance-related ontologies (WOB Core,
PML Trust, CWM Reason), I have not yet seen anything that entirely
fits.  Thoughts?

Best regards,

Josh


Joshua Shinavier
Tetherless World Constellation PhD student
Troy, NY (USA)
+1 509 570 6990
Received on Tuesday, 29 September 2009 04:47:26 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 1 March 2016 07:42:15 UTC