The easy way, I suppose, would just to embed SparkWeb as a "side-by-side" pod, with some type of real or faked single sign on scheme, but it sure would be nice to have support for a native XMPP/Jabber-based IM client in AFCS...or XMPP/RTMP interoperability.
Can anyone at Adobe speak to concrete plans on this front? We're looking at Jive SBS right now for some of these reasons, plus support for VOIP integration via Asterisk, Red5, and others.
To that end, any plans for SIP/VOIP integration in AFCS (beyond the Audio support currently available)?
To my mind, I don't know if I'd characterize what you're describing as "interop". Maybe the term is closer to "rendez-vous"? As you say, if you're trying to do "push to collaborate", you typically need some medium through which to expose presence of the recipient, and push a URL at the intended recipient - in this case, likely the URL of the client swf which will take you into a specific AFCS room, or perhaps (if you're already in a swf/AIR app) the URL of the room itself (roomURL, to plug into a ConnectSession).
We've fairly studiously stayed neutral on the topic of rendez-vous, simply because we haven't seen exactly how this might play out. We started AFCS at a time where we were hoping XMPP would take over all these presence / rendez-vous problems, but luckily we decided to punt until we knew more - to date, I'd say Twitter or FaceBook Connect might be even more compelling option for this.
Anyhow, all of this is to say that we're going to focus on opening up the server so that you can publish and subscribe over HTTP, in a set of RPCs which will be very close to XMPP XEP-60. Hopefully from there any variety of true interop cases should be possible.
As you well know, there is already a proliferation of occasionally convergent, but generally divergent, collaboration technologies, applications, platforms, and providers out there today. In an "always connected" world, the typical actor will find his/herself needing to jump between/across their multiple "communities" using, quite often, different modalities and devices (voice, video, text, e-mail, blog entry, activity stream via phone, netbook, handset, etc.). To "find someone" and to initiate real-time collaboration with them is thus quite difficult in the abstract. Providing interop, starting with areas such as presence and authentication, list management, connection initiation, and routing, would be a really good place to start. Many vendors are exploring this within their own technology stack or in "like stacks", but there still doesn't seem to be a clean way to achieve it in a hetergeneous environment today.
Starting with something relatively easy and somewhat standardized, such as XMPP-based services (specifically chat and presence) is a logical place to begin.
I simply want a "push to collaborate" button, so to speak, that can find the people/person/(devices?) I want to interact with and initiate the relevant mode of collaboration across whatever boundaries may need to be crossed. Not too much to ask, is it? 😉
But back to the initial question, is this something that could be implemented via some type of server plug-in in AFCS at some point?
At some point, we will be looking at all of this, but it's not in short-to-medium-term plans yet. When you say "interoperability", it would help a lot to hear the use-case you've got in mind. We're definitely interested in hearing what you'd need!