Re: [apps-discuss] Fwd: New Version Notification for draft-shelby-exi-registration-01.txt

[Ned Freed <[e-mail]>]

> On Fri, Apr 13, 2012 at 08:24:39PM +0200, Julian Reschke wrote:
> >>>>
> >>>> All true, but as far as I can tell, you didn't mention
> >>>> Content-Encoding in the message I replied to.
> >>>
> >>> Ah, I see the disconnect. In my original message, I said:
> >>>
> >>> "Why not call it application/senml+xml and encode it with EXI?
> >>> Doesn't that tell you everything you need?"
> >>>
> >>> I should have been more explicit -- what I meant the above to convey
> >>> was
> >>>
> >>> "Why not call it application/senml+xml and encode it with EXI?
> >>> That is, use Content-Type: application/senml+xml and
> >>> Content-Encoding: exi .
> >>> Doesn't that tell you everything you need?"
> >>>
> >>> So, reset. With that clarification, what do you think?
> >>
> >> Yes, absolutely, that works with HTTP.
> >>
> >> AFAIR, Core (<http://tools.ietf.org/wg/core/>) doesn't have
> >> Content-Encoding, though.
> >> ...
> >
> > ...with the constraint Ned mentioned in a parallel message (no
> > out-of-band information needed).

> I suppose that the registration for application/senml+xml could
> mandate use of a particular schema when encoded with EXI and exclude
> any additional schema (i.e. no schemaId allowed).

That's certainly doable from a registration standpoint - we even have
an Encoding Considerations section that could be used for this.

However, keep in mind that this approach could be a bit of a problem if
a new version of senml comes out that includes a schema change.

				Ned