Re: [syndication] Re: Thoughts, questions, and issues.

[Ian Graham <ian.graham@utoronto.ca>]

On Tue, 22 Aug 2000, Edd Dumbill wrote:

> On Mon, Aug 21, 2000 at 04:47:40PM -0700, Gary Teter wrote:
> > I just don't understand, practically speaking, why RDF needs to be a part of
> > the syndication process. 
> 
> Key point. Syndication is just one thing you can do with RSS. There are
> many more interesting things. Syndication as a task requires a different
> set of stuff than describing a web site, this seems to me to be what Ian
> Graham is saying.

Yes, I would agree with that ;-)
 
> So, RSS is "site summary". If you choose to syndicate that metadata,
> cool. There are many other things one could do with it.
> 
> In my opinion RSS and syndication should be separated. Syndication is
> about routing, rights, scheduling etc. That applies to any content,
> whether metadata, images, MP3s or whatever.

Yes -- that's what I'm making a stab at in the syndication model I
posted at http://www.java.utoronto.ca/news/syndication-3.txt. Still
very rough, but I think a useful first stab (and much shorter than ICE
;-) )

> I don't think discussions conflating the two ideas of metadata and
> syndication will ever reach a happy resolution.
>
> -- Edd
> 
> PS. As for RDF, it's one way of skinning the cat, and quite a funky one
> at that.  Why not let's try it and see how it goes?
> 

To me (and I think to others) the unspoken question is: when is it better
to create your own application-specific vocabulary and syntax, and when is
is better to use RDF to re-express such a vocabulary and syntax in the
'generic' RDF language?  The answer is -- who knows?  There's only one way
to find out, and thats by trying it out and seeing which works best.

Best --

Ian