Re: [ogpx] HTTP(S) and XMPP as transports of OGP application layer messages

[Mojito Sorbet <[e-mail]>]

Infinity Linden wrote:
> For
> instance, does it REQUIRE request/response semantics? (i.e. - if i'm
> writing a compliant server layering OGP over XMPP, and i receive an
> OGP "request" in an XML stanza, am i REQUIRED to send a response?)

Well, if you are "requesting" something, that should "require" a 
response.  For example, the <iq/> stanza in XMPP does this. for both 
"get" and "set" operations.  The need for a response does not block the 
channel from carrying other stanzas in the meantime, which I think is an 
advantage over HTTP.

But if you are just reporting an event, as in a <presense/> or 
<message/> stanza, I do not see the need for any reply at all.   The 
whole reason for using something over TCP is so that we do not have to 
worry "did it get there?"

I am not suggesting that <presence/>, <iq/>, or <message/> stanzas as-is 
are appropriate for OGP.  I only mention them as examples of how an XMPP 
stream can carry both types.

XMPP today does some form of negotiation at session startup, mostly for 
determining login credentials and encryption level, so there is some 
precedent for that.