Re: Minutes of W3C Audio WG Teleconference 2013-08-28

> * AOB 
> 
> Jer asked about testing. It’s the next big work item after we solve the big architectural 
> issues. 

Yes, it’s going to be a big bit of work. I occasionally help and prod people to contribute tests to our repo, and we are making slow progress, but it’s really tough to find people to commit the time to help out.  

https://github.com/w3c/web-platform-tests/tree/master/webaudio 

It’s not as simple as just moving the tests from the gecko or webkit over either, as they need porting to use the W3C test harness and checking against the current spec to make sure they reflect the wording of the spec and not the reality of the particular implementation. I wrote some notes on how to do that: 

http://blog.chrislowis.co.uk/2014/04/30/testing-web-audio.html 

I feel like requiring a (failing) test with every change that gets merged into the spec would be overkill and slow things down, but obviously it does mean that the spec is moving at a faster rate than the test suite!  

Cheers,  

Chris  



On 29 August 2014 at 13:24:02, Olivier Thereaux (olivier.thereaux@bbc.co.uk) wrote:
> Dear all,
>  
> The minutes of our latest teleconference, on 28th August 2014, are online at:
> http://www.w3.org/2014/08/28-audio-minutes.html
>  
>  
>  
> * progress on AudioWorker/ScriptProcessorNode
>  
> Need to assess whether the question by ROC was an objection, but the proposal is making  
> progress. We are now thinking of providing a monkey patch to help transition.
>  
> There was also a discussion about changing AudioProcessingEvent to use float32arrays,  
> change name, mark old one as deprecated.
>  
>  
> * communication/education on AudioWorker and changes to SPN
>  
> ChrisW will write a blog post, and Olivier will talk to the W3C staff/comms team to get  
> support (maybe a W3C news item?). It sounds like we will publish the new interface and  
> publicise its use (also: console warnings) for a while before phasing out the old SPN.  
>  
>  
> * heartbeat pub
>  
> Agreement to freeze a draft of the web audio API spec into W3C’s /TR space as soon as we have  
> a solid draft with the AudioWorker proposal. We will then work towards a /TR draft for  
> web midi, assessing whether that should be a last call draft.
>  
>  
> * suspend()/resume()/release() on AudioContext
>  
> No controversy, but we want to check with TAG that the release() method does not set a problematic  
> precedent.
>  
>  
> * Increasing buffer for power consumption purposes
>  
> Seems we have a rough agreement in principle, API proposals yet to be discussed in depth  
>  
>  
> * AOB
>  
> Jer asked about testing. It’s the next big work item after we solve the big architectural  
> issues.
>  
>  
> * Next meeting
>  
> Tentative: 11th September.
>  
> --
> Olivier Thereaux - BBC Internet Research and Future Services
>  
>  
>  
>  
>  
>

Received on Friday, 29 August 2014 14:21:11 UTC