[rdfweb-dev] foaf.rdf as a requirement

[Danny Ayers]

> >There have been suggestions of using a standard naming convention of
> >"foaf.rdf" for the file -- so then we can use Google as another
> way to find
> >FOAF documents.
> >
> >But thats not a requirement.
> >
>
> Actually this is an excellent idea !   But I think it *should* be a
> requirement.  I'm aware of the trend when writing standards to make only
> minimum restrictions,  but there are some times where the benefits of
> resticting far outweigh the loss of freedom.  This is such a case.
> Check out the list of foaf.rdf files provided by [1] below.  To be on
> this list is so very easy,  you give up so very little ... if we all did
> it, then the foaf space would be far easier to find.   Why not specifiy
> this as a requirement ?
>
> [1] http://www.google.com/search?q=foaf.rdf

I wouldn't lose any sleep if this was a requirement, but I'd lean towards it
not being. Certainly worth considering. My first response is that it seems
to give the filename an extra significance which might be misleading. Also
I'm not sure of the true value of Google's list, at least three of the hits
on the first page aren't foaf files.

It does raise a good question though, how to handle "formally FOAF'd as" :
Let's say my primary FOAF file is currently http://wayne.org/foaf.rdf but
for some reason it moves to  http://waynetta.org/foaf.rdf. Would it be
better to use http 301 Moved, or something explicit within the RDF? In the
new file and/or the old?

Cheers,
Danny.