Seems like just the other week there was the strong possibility of vendor-neutral support for video as a standard element in HTML 5. Now it's all up in smoke no thanks to disagreement on what codec to implement as the base standard, and disagreement over a free-and-open spec vs. a for-pay spec.

The original idea was simple. Implement a tag in HTML 5 that allows browsers to play back video streams natively, with no plugins or third-party add-ons like Flash. People liked the idea, but they couldn't agree on whether to support H.264 (widely-used, patent-encumbered, not free) or Ogg Theora (patent-free, but not widely used). End result: the folks compiling the spec have decided to forget about defining a codec standard, which means the tag is effectively useless as any kind of standards-defining measure.

When I originally wrote about this issue, my take was this: H.264 may be patent-encumbered, but it's also been used that much more broadly. People are reluctant to embrace an open standard that simply hasn't had as wide-scale a shakedown, especially for something as demanding as video. Theora needed to show that it could be time-tested as a solution, some public implementation people could point to as an example.

That left three choices: a) accept an open standard that was barely being used in the first place (Ogg Theora), b) accept a standard that was in wide use and had a better chance of being supported, even if it was patent-encumbered (H.264), or c) some third path, perhaps involving freeing up an existing codec from patent protection. None of these things seemed attractive enough, and so the end result was d): let the whole thing fall by the wayside, and let video remain the province of the proprietary software world.

The most frustrating thing about this whole mess is how the folks who were against adding Ogg Theora to the mix claimed (in part) that this was because there was no guarantee that the format couldn't be subverted by a patent claim in the future. But that doesn't mean H.264 is immune from being subverted in the same fashion -- although the MPEG LA, the folks who devise the royalty structure for H.264 and related patents, haven't exactly helped make any of this easier either. Their market is bigger than the web alone, so they are, I guess, reluctant to play favorites and make it easier for Mozilla and Opera and the other non-MPEG folks to adopt MPEG without breaking a sweat.

What I'd like to see happen is two things. One, a broader use of Theora as a backend codec, so that content can be parallel-encoded as Theora and served up in that format if needed. This gives Theora a chance to provie itself and, if need be, improve itself in the field.

The other is a firmer set of cost structures for licensing H.264 for serving up web video -- say, total indemnity for both encoding and playing back individual clips under five minutes in length, in perpetuity. That means the little guys who serve up movie trailers and other goodies get a free pass, but the folks who want to serve up industrial-strength streaming video (Joost, NetFlix, etc.) pay for that privilege. There's something like this right now in the H.264 licensing, but it's got a sunset provision, so people are rightfully leery of touching it lest it come back to bite them and empty their wallets.

What I suspect will happen is this. The

And if you think this is a horrible, messy, slow-as-January-molasses way to do it, you're right. If we look back on this one day and regard it as one of the many nails in the software-patent coffin, I won't be surprised.

InformationWeek Analytics has published an independent analysis of the next-generation Web applications. Download the report here (registration required).

Follow me and the rest of InformationWeek on Twitter.