By Carl Ford
Here is an interesting question. If you wanted to get High Definition Voice out in the marketplace, where would you start?
My readers are probably too young to remember Candice Bergen and the Sprint Pin drop commercials, or the AT&T retaliation of Whitney Houston pumping up the base if not the volume.
Wireless calls have not been a better experience with voice. In fact, I would say it's made any sound troubles acceptable.
I remember the first time I heard the Global IP Sound ILBC codec (not a wideband codec but better voice quality) and thought to myself, "This is great! Who is going to deploy it?" The answer turned out to be Skype, at least until they rolled their own wideband solution.
In recent years the number of wideband codecs available has increased, and companies like Microsoft and Polycom have been advocates. But how does one start deploying?
Skype's self-contained strategy was a great starting point, and we get to hear the issues of being off-net and condensing the voice to fit the PSTN. I'm not sure I would advocate every network creating their own Skype-like service. While Peer-to-Peer friends would celebrate, more factional Internet apps seem to make Metcalfe's law moot.
Others would rather be advocates of Web 2.0 solutions that ignore the PSTN. While this is attractive, it loses something when facing the realities of reaching people anytime, anyplace and anywhere. Logically, wouldn't that suggest wireless is the place to deploy wideband codecs?
AudioCodes advocates building IP Islands and working with all wideband players in their High Definition strategy. It sees that Microsoft's RTA is going to be in the Enterprise, with G.722 in DECT phones sold by cable operators and AMR-WB the logical deployment by wireless operators. It also supports the Skype codecs.
All well and good, but the synergy comes in gathering the islands together, and here, the subject of federation comes in and the question is, who is going to do it? I can make a case for a bunch of federating strategies, but traditionally the vision of federation breaks down and the companies become extended walled gardens.
In order for a federation to really work, it needs to be something less than a service and more of an addressing scheme. A registry for like to find like and perhaps point to a transcoding solution along the way. I would love to say that I know of such an entity, but for today, solutions are on the drawing board, but not implemented, to my knowledge.
Another more likely example works like peer-to-peer among the islands, but I have not seen this strategy in the field. However, the quality is good enough that people may choose it for themselves and the network upgrades over time, just because of consumer demand.
Brick by brick by the castle is built.