Re: Public feedback on HTML5 video

On Thu, Dec 31, 2009 at 10:39 PM, Robert O'Callahan
<robert@ocallahan.org> wrote:
> On Thu, Dec 31, 2009 at 6:17 AM, Tab Atkins Jr. <jackalmage@gmail.com>
> wrote:
>>
>> So, browser devs, do you think that, if autobuffer=off was supported
>> and present, you would treat it any differently than if autobuffer was
>> missing entirely?
>
>
> As Chris D said --- no.
>
> Browser vendors already have ample incentives to conserve bandwidth. Apart
> from anything else, it's useful to be able to load a page with 50 <video>
> elements in it and not grind to a halt. Since "autobuffer" exists to signal
> that the video is likely to be played, one must treat absence of autobuffer
> as a signal that the video is unlikely to be played, and it's obvious that
> the correct response to the latter signal is to conserve resources by not
> buffering. Arguably this could be made a little clearer in the spec.

I think that's the minimum that needs to be done. I'd be happy with
that, but it has to be spelled out clearer in the spec.


> IMHO this entire thread is an overreaction to a Webkit bug. They'll fix the
> bug, meanwhile, everyone just relax :-).

And Chrome. Also, right now, with they way the spec is written, it's
not a bug, because it's spec conformant. This is the main issue I
have.

Cheers,
Silvia.