[rdfweb-dev] seeAlso

Jim Ley jim at jibbering.com
Sun Aug 10 00:35:56 UTC 2003

"Julian Bond" <julian_bond at voidstar.com>
> Chris Croome <chris at webarchitects.co.uk> wrote:
> >One thought, when your weblog has a RSS 1.0 feed would it make sense
> >to add a seeAlso to point to the blog RDF?
> Two thoughts I've had in this area.
> I wonder if we need a new property or a convention for seeAlso that is
> only used to point at FOAF files.

FOAF certainly has no ability to change the semantics of rdfs:seeAlso,
that's owned by the w3, so FOAF people can't say anything about it.   As
I've said many times before "FOAF files" are somewhat misleading, is a
400,000 triple document that uses foaf:mbox 50 times a FOAF file?  it
certainly could be useful to a FOAF consuming app, so that would make it a
FOAF file, and if that's a foaf file, I'm sure you agree anything can be.
rdfs:seeAlso is a generic linking mechanism, nothing more, FOAF people can't
add any more semantics to that.

Creating a subProperty of it would allow you to define extra semantics,
however subProperties do raise the implementation bar at the moment, since
you need to look for rdfs:seeAlso and all sub-property's, if you teach the
sub-properties into the start-up code, this isn't too expensive, but to be
truly useful you'd want to do it all the time, to catch new subProperties
when you meet new rdf schemas.  This becomes harder.

> Then it's quite hard to say what a "FOAF" file is.

I'd say it's impossible, and for me a pretty unimportant one, as none of my
use cases have any such idea.

> But it's still annoying as a
> scutter developer to get pointed at files that aren't even RDF.

Remember you're pointed at URI's they may not have RDF now, but they might
return RDF later... (are you sending appropriate accept headers btw?)

> The other thought is for sites that generate large quantities of FOAF
> (such as Ecademy). Would it be useful to produce a scutter plan for the
> whole site? How should I signpost it's existence?

I think it would be, although it might be better to have a number of
aggregrated files, as that would keep your bandwidth down, you could then
keep scutters away from the real files, and improve your bandwidth


More information about the foaf-dev mailing list