Re: Valid representations, canonical representations, and what the SW needs from the Web...

From: <Patrick.Stickler@nokia.com>
> > From: Jeffrey Winter [mailto:JeffreyWinter@crd.com]
> > Also, how
> > could you ever express meta-metadata?
>
> Or did you mean something else by "meta-metadata"?

I think Jeff was pointing out that they RDF returned could itself have
metadata attached to it.  Using OPTIONS, you could do something like:

OPTIONS http://example.invalid/foo

returns

200 OK
Content-Type: text/meta-URI

http://example.invalid/meta/foo


Now you can just use the good ol' GET, PUT, and DELETE to work with the RDF
document at that URI.  Further, it would then be possible to issue:

OPTIONS http://example/invalid/meta/foo

and get back

200 OK
Content-Type: text/meta-URI

http://example.invalid/meta/meta-foo


And so on...


I should also point out that Roger Costello and David Jacobs were
recommending a similar technique using a new http header.  Details for that
can be found on distributed-registry at yahoo groups.  Personally, I prefer
the OPTIONS method.

---
Seairth Jacobs
seairth@seairth.com

Received on Wednesday, 5 February 2003 11:01:33 UTC