Re: Draft-gregorio-atompub-multipart on the standards track?

["'Aristotle Pagaltzis'" <pagaltzis@xxxxxx>]

* Brian Smith <brian@xxxxxxxxxxxxxx> [2008-06-19 14:55]:
> Aristotle Pagaltzis wrote:
> > I just don't see how using `multipart/related` is in any way
> > Atom-specific.
> 
> You just said previously that the mechanism only needs to work
> for atom:content/@src. The feature requires one part to be an
> Atom entry. The only defined way to advertise support for the
> feature is to document it in an AtomPub service document. The
> behavior of the server is defined in terms of AtomPub
> operations. AtomPub is a protocol for the manipulation of Atom
> documents.

How does any of your statements not apply likewise for XOP?

For single-shot creation of Media and Link Entry, you obviously
need an Atom Entry, whether with XOP or bare `multipart/related`.
That the server supports this feature needs to be avdertised in
the service document, whether the mechanism is XOP or bare
`multipart/related`. That the server will interpret particular
combinations of media + Entry as a single-shot Media + Link Entry
creation in Atompub terms is pretty much the premise of this
spec, so applies regardless of whether XOP or `multipart/related`
is used.

> I think it is fair to say the proposed mechanism is
> "Atom-specific," and I don't think anybody here associates any
> negative connotation with that term since all the work done
> here is Atom-specific.

But the way you said it suggested that using XOP would be less
Atom-specific. So let me restate: I didn’t and still don’t see
how using `multipart/related` is Atom-specific in any way that
using XOP would avoid.

Regards,
-- 
Aristotle Pagaltzis // <http://plasmasturm.org/>