[whatwg] H.264-in-<video> vs plugin APIs

[Mike Shaver]

On Sat, Jun 13, 2009 at 10:08 AM, Chris DiBona<cdibona at gmail.com> wrote:
> No, but it is what I worry about. How agressive will mpeg.la be in
> their interpretation of the direction that theora is going? I don't
> think that is a reason to stop the current development direction (or
> the funding of it) but I thought that Dirac, with the BBC connection,
> might make a better opponent politically than Theora.

I have reason to hope that Mozilla would be a good opponent
politically as well; that was certainly one piece that we were glad to
bring to the table.  Not that I have anything against Dirac, and would
love to see support for it as well, but I think it's farther from
being web-practical due to bandwidth minimums than Theora is.

> It is client compatibility first, and global/edge bandwidth restricted
> as well. I'd prefer to ship with the reference libraries and have told
> the team as much.

I would certainly like to understand the reasons for Chrome shipping
with H.264 support, but this thread has confused me a little.

As I understand it, you have to have it because, per your analogy with
plugins, there is a lot of legacy content with H.264 that will be made
available via the <video> tag so you're forced to provide support or
risk market irrelevance...

...but that legacy content is virtually *all* from Google properties...

...but Google can't provide Theora video because of...

a) client compatibility limits (circular with the above, though
Firefox will provide ~25% of the web with Theora support, vastly
larger than I think even the most optimistic projections of
Chrome+Safari with H.264 when Chrome ships <video>, so maybe we can
out-egg that chicken)

b) bandwidth concerns (but even if Theora took _double_ the bandwidth,
and _all_ the content was converted overnight, that's still only a 25%
increase in bandwidth, plus a few percent for Chrome when it ships
<video> as well)

So if we can remove the bandwidth spectre -- and I really believe we
can do that, if it hasn't already been done with the state of the art
encoders -- it sounds like it unwinds to client compatibility, which
happily Mozilla and Google have some significant influence over!

Mike