Hi Hadrien,

On Wed, May 18, 2011 at 03:25:57PM +0200, Hadrien Gardeur wrote:
> >
> > On reflection, I don't think the link template idea above is a good
> > solution,
> > because the client has no way of finding out what inline content types are
> > available.
> >
> > I wonder if a better solution would be to invent three new pieces of
> > machinery:
> >
> > * a new @inline-type extension attribute for use on atom:link elements
> >
> > * a new Accept-Inline HTTP header
> >
> > * a new <accept-inline> element for use within app:collection elements in
> > Atom
> > service documents
> >
> 
> These are three different problems though...
> Erik seems to be strictly interested in solving the first one, and in this
> case, would a single attribute with a secondary mimetype be enough ?

Yes, absolutely. I was just thinking that, if you had all three pieces of 
machinery, then you would have the tools to solve a range or problems all 
related to encapsulated media types, including Erik's. This might be a 
reasonable scope for an internet draft, for example.

> I don't think that we need a new rel value, or that we should limit this new
> attribute to a specific rel either. While you're strictly talking about
> alternate versions of the same content, this attribute would also be useful
> for other use cases (for example I might publish a music review as an atom
> entry, and provide a link@rel="related" with type="application/zip" and
> inline-type="audio/mpeg").

Good point.

Cheers,

Alistair

-- 
Alistair Miles
Head of Epidemiological Informatics
Centre for Genomics and Global Health <http://cggh.org>
The Wellcome Trust Centre for Human Genetics
Roosevelt Drive
Oxford
OX3 7BN
United Kingdom
Web: http://purl.org/net/aliman
Email: [email protected]
Tel: +44 (0)1865 287669

Reply via email to