W3C home > Mailing lists > Public > www-style@w3.org > March 2013

Re: [mediaqueries4][css-speech] Media queries for multichannel audio ?

From: Florian Rivoal <florian@rivoal.net>
Date: Fri, 29 Mar 2013 16:58:26 +0100
To: www-style@w3.org
Message-ID: <op.wuptbos0f5de51@localhost.localdomain>
Hi,

Your request has not fallen off the radar, don't worry. We just have a  
very full plate, and it sometimes takes time before things get on the top  
of the agenda.

I think at this point, the most useful thing to do would be to collect  
uses cases, so that we can best judge if the solution proposed would be  
effective at solving them. The examples you gave in your previous mail are  
useful, but not really what I would call use cases though.

"Providing the right number of audio channels in a video streamed using  
the media attribute of the <source>  element or in the Media Source case  
by JavaScript using mediaMatch()" is a use case (maybe two).

I would be interested in knowing whether people want to use it for  
something else. Does it have applications in the context of css-speech,  
for instance? If yes, what are their expectations? This is the kind of  
thing I'd like to hear about before taking any decision.

Also, just to clarify, for your use case, assuming a 5.1 capable  
environment, do you want the number of channels detected to be 5 or 6?

  - Florian

On Fri, 29 Mar 2013 16:39:16 +0100, Mark Watson <watsonm@netflix.com>  
wrote:

> Florian, all,
>
> What should be the next steps on this - should I provide more detailed  
> use
> cases ? (in email, or is there a wiki ?)
>
> ...Mark
>
> On Sat, Mar 23, 2013 at 10:06 AM, Mark Watson <watsonm@netflix.com>  
> wrote:
>
>>
>>
>> Sent from my iPhone
>>
>> On Mar 23, 2013, at 6:20 AM, "Florian Rivoal" <florian@rivoal.net>  
>> wrote:
>>
>> > On Wed, 20 Mar 2013 16:03:48 +0100, Mark Watson <watsonm@netflix.com>
>> wrote:
>> >
>> >> Florian, all,
>> >>
>> >> How do I progress this issue ? Do I need to file a bug ? (where?) Do  
>> I
>> need to join the WG ?
>> >
>> >
>> > Leaving your other two issues aside for the moment, and focusing on  
>> the
>> mutichannel part.
>> >
>> > There is no need to join the WG, raising the issue in this mailing  
>> list
>> and following up on conversations (as you are doing) is normally enough.
>> Once we reach sufficient consensus, or at least understand well enough  
>> the
>> questions that needs to be answered, we can take it to one of our weekly
>> conference calls to try and decide one way or the other.
>> >
>> > From where I stand, I see a few open questions with your proposal:
>> >
>> > * Should it be number of channels, or number of speakers?
>> > To answer this questions, we would need a variety of use cases to be
>> described, to be able to judge which one fits best.
>> >
>>
>> It's the number of output channels, because:
>> - you can't know the number of speakers - the audio output may be a
>> digital one connected to an A/V receiver or other equipment which is  
>> then
>> connected to the speakers.
>> - some systems use audio processing techniques to simulate surround  
>> sound
>> without the usual number of speakers - for example soundbars.
>>
>> Some use-cases, all involving content available in stereo and 5.1,
>> provided as alternative <source> elements in an HTML media element:
>> - a PC that supports the 5.1 codec, but is outputing the audio over a
>> stereo analogue output jack
>> - a PC that is outputing the audio over a 5.1 digital audio output
>> (S/PDIF, HDMI)
>> - a TV that is outputing the audio over internal stereo speakers, with
>> only a capability for basic downmixing
>> - a TV that is outputing the audio over a 5.1 capable digital audio  
>> output
>> - a TV with internal signal processing that is capable of simulating
>> surround sound from a 5.1 input through less than 6 speakers
>>
>> > * If number of speakers, can that be reliably detected?
>> > I am not an expert on the topic, but it seems to be that the OS (let
>> alone the browser) is not necessarily aware of how many physical  
>> speakers
>> there are.
>> >
>> > * It seems to me that your main use case for this media query would  
>> not
>> actually be applying different styles based on the situation, but feed  
>> that
>> back into javascript, and use it there. If that's the case, I wonder if
>> media queries is the best tool to solve the problem.
>>
>> It's a resource selection problem, which is done today on the media
>> element using media queries (the media attribute of the <source>  
>> element)
>> and could be done in the Media Source case by JavaScript using the
>> mediaMatch() function (ensuring that the script approach matches the
>> declarative approach).
>>
>> ...Mark
>> >
>> >
>> > - Florian
>> >
>>
>>
Received on Friday, 29 March 2013 15:58:50 UTC

This archive was generated by hypermail 2.3.1 : Friday, 29 March 2013 15:58:50 UTC