We naturally use Enterprise Connect as a time to try and talk with as many enterprise end users as we can about their biggest concerns. This year, one of the things that came up much more than ever was the issue of desktop video—mostly as a concern, something that end users seemed worried about their networks’ ability to support.
Video in general was a big topic at the show; our video track out-drew all others during several time blocks. Does this mean that 2011 is going to be the Year of Video? No, I don’t believe it does. I think it means that enterprise decision-makers are aware that desktop video usage is only going to spread, and mostly likely in unpredictable ways.
It's the unpredictability that causes the concern. If there’s a top-down mandate to use an application, or if there’s a clearly-unstoppable bottom-up phenomenon—the iPhone being a good example—you at least know what to expect and can plan for it. If video is something that’s just a part of every new Unified Communications interface, and if cameras are so cheap that everyone has one, then you have the potential for an unmanageable torrent of traffic to hit your network.
Or not. Lots of people use video right when they get their new UC client, then the novelty wears off or they find they don't really need it, and they quit using it. If a network manager had provisioned the network for that initial wave, she’d wind up with a huge amount of unused bandwidth. Not the worst thing ever--bandwidth tends not to stay unused for very long—but still not particularly efficient.
But if desktop video is going to be in widespread use, what should it look and sound like? I don't think I've ever seen a desktop client deliver video that I didn’t find more distracting than I found helpful, the quality was so uneven.
There are some indications that quality does matter. Skype asks you to rate the quality of your call, be it video or audio, after you complete it. How effectively they translate that feedback into network improvements, I don't know. But at least asking people the question may be a way of beginning to set up a little higher expectation for quality. After all, it was Skype's superior voice quality over the open Internet that really made the software the success it became—they certainly weren't the only Internet VOIP software out there, they were just the best.
On the last day of the Enterprise Connect exhibition, I stopped by the Psytechnics booth and chatted with CEO Anthony Finbow, and Paul Barrett, the company’s new CTO (side note, in case you missed it: Psytechnics just got acquired by NetScout). Paul mentioned "MOS for video," which I didn’t even know existed (MOS or mean opinion score is the classic gold-standard measurement for voice quality). Evidently not too many other people know about MOS for video, because when I googled the term I primarily got results for Mos Def videos. But Paul explained that there are specifications and guidelines for testing and rating video on the MOS scale, and he suggested that this measurement could come into wider use.
So we come back around to the Cisco line: Video is the new voice. That's not true today, for many reasons, not least of which is that the video that’s as accessible as voice is generally not as good, primarily because the networks it runs over are not built to carry it with that level of quality. My opinion is that video usage won't really take off until it reaches some minimum standard of quality—not perfect, necessarily, but at least good enough, on a consistent basis, that it's not a distraction. Video will be the new voice, not when you notice it, but when you don't.
