W3C home > Mailing lists > Public > www-archive@w3.org > September 2002

Re: [RSS-DEV] Why not separate the RSS part form the RDF part?

From: Seth Russell <seth@robustai.net>
Date: Sat, 7 Sep 2002 14:23:48 -0700
Message-ID: <00a501c256b4$dbb57700$657ba8c0@c1457248a.sttls1.wa.home.com>
To: "Dan Brickley" <danbri@w3.org>, "rss-dev" <rss-dev@yahoogroups.com>
Cc: <www-archive+rss@w3.org>

From: "Dan Brickley" <danbri@w3.org>

> > Why not separate a super simple (yet extensible) RSS ... say like
Aaron's
> > RSS 3.0 proposal, or whatever .. from the full power of RDF ?
>
> We did. It's called RSS 1.0! RSS 1.0 limits the full RDF syntax, and
> allows for RDF extension data to be included both inline or by reference.

Personally I have no problem with RSS 1.0.  My proposal was for those who
want to push for simplification.  Having a simple feed that points to full
blown totally unrestrained RDF files is one solution that allows us to cover
all design goals.  Come to think of it,  If Dave Winer would allow optional
semref element in his feeds, that would solve the problem.

> This fragment:
>    <rdfs:seeAlso>
>          <rdf:Description
rdf:about="http://rdfweb.org/people/danbri/rdfweb/danbri-foaf.rdf">
>                <dc:format>application/rdf+xml</dc:format>
>                </rdf:Description>
>     </rdfs:seeAlso>

I have no problem with this.  All the information is all on one node and we
know exactly what parser to use should we choose to dereference the link.

> I would just write:
> <rdfs:seeAlso
rdf:resource="http://rdfweb.org/people/danbri/rdfweb/danbri-foaf.rdf"/>
> since rdfs:seeAlso is a pretty strong hint that the thing pointed to is
> RDF.

This is a problem for automated agents that do not want to *guess* at hints,
nore can they attempt a dereference before they render their results.

It was your own sccuterplan that broght this to my attention.  I did as best
as I could to guess which referecnes were to RDF and which were not.  See
the screen shot below.  Note the ratio of references to RDF documents (red
~read~> buttons) to non RDFdocuments (blue ~web~> buttons).

<http://robustai.net/sailor/screenshots/scuterplan.html>

Look, suppose the web were designed so that when a browser rendered a
hyperlink it had to *guess* whether to build a button for the human user to
click on or not.  Do you think that such wishy washy hyperlinks would have
been as widly deployed as our blovid <a href> ?

> rdfs:seeAlso is deployed and deployable already.
> BTW http://www-106.ibm.com/developerworks/xml/library/x-foaf2.html and
> http://www-106.ibm.com/developerworks/xml/library/x-foaf.html gives a nice
> overview of how we're using this technique in FOAF to link RDF documents
> together.

I've done a lot of browsing around your RDF files. Congratulations, seems
you have autored more RDF than anyone else that I know of.    But had you
depolyed a tightly defined semref instead of rdfs:seeAlso, it would have
saved me and my sailor from going off on many a wild goose chase.

Seth Russell
http://robustai.net/sailor/
Received on Saturday, 7 September 2002 17:24:31 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 7 November 2012 14:17:22 GMT