W3C home > Mailing lists > Public > public-grddl-wg@w3.org > February 2007

Re: Best way to embed Semantic Data

From: Harry Halpin <hhalpin@ibiblio.org>
Date: Fri, 09 Feb 2007 18:44:51 -0500
Message-ID: <45CD0773.5070707@ibiblio.org>
To: Simone Onofri <simone.onofri@gmail.com>
Cc: public-grddl-wg Group <public-grddl-wg@w3.org>

    Good questions!

    Without chair's hat on, I'd say the problem with microformats is
when you want to combine lots of them they can be ill-defined, and also
sometimes you want to express semantic data (say, data about rare
birds!) that there isn't a microformat for. However, since there is so
much microformat data out there, and it is *easy* to use and has a
user-community, it makes sense to use microformats when the data you
want to express maps to microformats, and use GRDDL to get RDF out.

    However, if you want something you can deploy now that can express
any semantic metadata in RDF, the way to go is Embedded RDF. It has a
few minor limitations (No XML Schema datatypes) but works out the box
with a GRDDL client to extract RDF.

    RDFa does a lot of things Embedded RDF does, but (correct me if I'm
wrong here BenA and IanD), but likely will be a full-fledged first-class
citizen RDF syntax with Schema data-types, etc. However, it's still in
the works, so I'd keep it on the radar for future projects.


Simone Onofri wrote:
> Dear WG Members,
> Thanks for Teleconference of this week, It's my second teleconference
> in my life and a great experience.
> I'm working hard to gain all information on GRDDL available in order
> to comment and help in the developing of documentation. Well, I
> "gleaned" web, more home pages (like Dan's [1] and Ian's [2] home
> pages) and works of Danny like hDOAP [3]. So, excluding RDFa [4] (that
> approach can be similar to eRDF). On the Primer [5],  there are to
> main ways to include in a Semantic layer on a XHTML document, in order
> to extract it via GRDDL. The peculiarity and power of GRDDL is to use
> (without data replications) information that are both Human and
> Machines readable.
> I'm talking about Microformats [6] and eRDF [7]. And the Question
> is... what is the best way to include the Semantic Layer?
> So, I'm testing for Microformats on my home page and it's more simple
> to implement and most diffused (especially for blogs). I'm also
> reading about eRDF (I'll implement soon) and code remain cleaned and
> clear. In particular when applied to a huge number of dialects.
> Third way is to adopt personal transformation and personal rules for
> embedding semantics but I try to exlude this except cases in which
> existing application is simply transformed via XSLT.
> In which case is better to use eRDF than Microformats and vice versa?
> Thanks for attention, first of next teleconference I'll send my little
> "official" presentation.
> Best Regards and Happy Week End,
> Simone Onofri
> [1] http://www.w3.org/People/Connolly/
> [2] http://iandavis.com/
> [3] http://dannyayers.com/2005/05/26/hdoap-a-microformat-for-doap/
> [4] http://www.w3.org/TR/xhtml-rdfa-primer/
> [5] http://www.w3.org/TR/grddl-primer/
> [6] http://microformats.org/
> [7] http://research.talis.com/2005/erdf/wiki/Main/RdfInHtml


Harry Halpin,  University of Edinburgh 
http://www.ibiblio.org/hhalpin 6B522426
Received on Friday, 9 February 2007 23:44:57 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:39:10 UTC