I have played the guinea pig to the mobile carriers' video models since VCast launched years ago. From VCast's VOD model to the early iterations of MobiTV streaming, SprintTV to Qualcomm's MediaFLO, I have tried living with all of these models. None of them ever came close to being part of my media rituals.

The key problems were apparent from the start. Generally, I don't ever think in terms of platforms, so I never crack open a phone or even start a Web browser and think "oh, I would like a video experience now." These services quite simply are forgettable. They are packaged simply as buckets of video. Even if I recall that they are there, I have to drill and drill to find what I want. There is no alerting mechanism to remind me of the one piece of content I really do want to see, and so little effort has been put into surfacing the content that is relevant to me. Sorry, but in my mind the manner in which many of these companies crafted and sold mobile video (as a TV grid in your pocket) is precisely what doesn't map well against the mobile experience.

And then there are the practical and technical considerations. The start-up and navigation times are still daunting enough to make me think twice before launching any mobile video platform. I am old enough to remember the old black and white set in Grandma's bedroom taking minutes to warm up. The buffering and the signal hunting of most "next generation" mobile video actually succeeds in capturing the feel of 1957 over-air TV.

I think some of the successes with video in the app stores make some of the old on-deck models look frowsy. Apps like the CBS March Madness, MLB at Bat, PGA Championship, TMZ, etc. capture my interest for video because their specific topics capture my interest and they all deliver more than video. They deliver experiences and content part of which involves video. They are targeted, easily accessible, rich, diverse and at least five to 10 clicks closer to me than any video on a deck.

Thankfully, this lesson is not being lost on some of the originators of the mobile streaming media platforms. "In the March Madness app, the average session time is almost three times the average of the core MobiTV experience," says the guy who should know, Ray DeRenzo, CMO, MobiTV, which made that app with CBS and also runs video experience on millions of feature phones. "March Madness" certainly featured the live telecasts and highlights of the games, but it also offered text, contextual data, and other material that kept viewers longer or drew them in even when a game wasn't on. That "core" product to which he refers is the classic MobiTV streaming media infrastructure the company has been selling for years under its own brand and as white label solutions for Sprint TV, AT&T and Verizon. Across the platforms, MobiTV claims 7 million subscribers. The app experience is about to help the traditional deck-based mobile TV experience change.

"The app stores have been a good testbed for us to innovate and introduce new functionality," says DeRenzo. With platforms that stretch across carriers and technologies, it is pretty much impossible to iterate changes and test new ideas in packaging and pricing. On the app platform, MobiTV recently launched with NBC a Notre Dame Central app that plays live games for its rabid fans. A new Ultimate Fighting Championship app is the first I know to offer pay-per-view in an app. In all of these cases, the apps are highly targeted to a specific vertical, the video is wrapped in rich contextually relevant content, and the pricing/packaging is flexible.

In bringing some of these lessons back into the on-deck mobile video model, DeRenzo says to look for a refresh of the core product later this year or early in 2010 that is more personalized and less like the classic cable TV grid. "We'll let users set their favorites or watch content in live or video on demand modes or download and store it on their device."

Content discovery may also change so that users become more like programmers who consult bookmarks more than a grid. Users could have their own channel autopopulate with content from a range of media providers. This more-verticalized approach lets users choose music, sports, news, financial categories and have video content pour in from multiple sources. The company already sells a product like this in the financial news niche, Mobi4Biz, but it is only available on select devices. Here you can create a watch list of companies to track, and the metatagging in the video creates a VOD playlist of personalized content. If MobiTV can find a way to link such a system into a polite alerting system or bring that personal channel to the surface of the deck, then we might have something closer to a mobile video consumption model.

I am still guardedly optimistic that mobile video is going to be huge, but I think it is still unclear what shape the giant will take. Mobile platforms don't just extend media; they change it. Mobility forces consumers and content providers both to rethink what kinds of media consumption patterns are possible and desired. How these lessons are learned -- and then, how these revised expectations from media are applied backward to Web video and living room TV -- is going to be the media story of the next decade. We're all back in first grade, but the class has no teacher. Time to improvise.

This commentary is insightful. I recommend it to others.