Excerpts from Kevin Smith's message of Thu Apr 29 17:17:20 +0100 2010: > On Thu, Apr 29, 2010 at 5:09 PM, bear <[email protected]> wrote: > > IMO clients and servers should now "do the right thing" with Atom > > payloads and we can get rid of most of it. > > Ah, but my client does "do the right thing" with Atom payloads - it > doesn't support them, so it doesn't advertise that it supports them. > The right thing for the identi.ca bot to do, when faced with a client > that isn't advertising that it supports atom payloads (or xhtml-im > payloads) is almost certainly for it to not send them. Fortunately, > this isn't a complicated protocol fix - we've got XEP-0115, which is > deployed and working for this kind of thing. >
Yes, I think this would work. Also another possible way to tackle it closer to the client would be SIFT (see http://mail.jabber.org/pipermail/standards/2010-March/023222.html ). Matthew > /K
