[whatwg] Codecs for <audio> and <video>

Charles Pritchard chuck at jumis.com
Tue Jul 7 08:39:56 PDT 2009

Kristof Zelechovski wrote:
> Audible mouse feedback is an OS thing, not an HTML thing.
While users will certainly have applications and os-level accessibility 
web designers may have their own unique methods of presenting feedback,
and I believe, given enough easy-access, innovative interfaces will come 
from it.
> I would rather have programmatic access to the MIDI synthesizer rather than
> be able to simulate it with a beep.
There is no guarantee that a host system will have a MIDI synthesizer,
and MIDI itself can quite simply be implemented with a series of audio 
clips, a sound font.
Given a very constrained environment (think old cell phone ring tones), 
you can still
"simulate it, with a beep", to provide some form of fall back.

> How do you detect that the client mixer is too slow?
How do you detect that the client player is too slow with the current 
audio tag?
> Why can't you just get the premixed jingles from the server?
You can get pre-mixed jingles from the server, but that may mean high 
in an otherwise resource constrained environment.
> Isn't the reading voice a CSS thing?  
Yes. So?
> Isn't sound transformation hard enough to deserve a complete API?  I think
> allowing playing with binary audio data is not going to help most
> programmers who do not have the slightest idea of how to deal with it.
Sound transformation does not need a "complete" API, but there are
certainly a few common filters which would make things easier on 
Given any amount of support, you'll soon see programmers developing
their own libraries.
> Imagine a Canvas interface with PutPixel only.

I've imagined Canvas with putPixel (and setPixel) and that's enough to 
arbitrary drawing, and arbitrary image formats, and that's better than 

Most programmers will be able to use a Javascript library, just as 
they've done in the past.

If you have particulars you'd like to see in an initial API, do share.

I could think of FFT/inverse FFT as a particular function which might 
make sense in an initial API,
as they're very particular functions, they're resource intensive, and 
they're in common use.

> Chris 

More information about the whatwg mailing list