Re: Testing for WebRTC

On 23/10/2013 14:09 , Dominique Hazael-Massieux wrote:
> Le mercredi 23 octobre 2013 à 14:04 +0200, Robin Berjon a écrit :
>> That said, to what degree do you actually *need* this to run with
>> multiple browsers? Would it not be better to run with a browser on one
>> side and something else pretending to be a browser that's actually testing?
> While this might be practical for some of the testing contexts (e.g
> continuous integration), I think the WebRTC Working Group should include
> in its CR exit criteria actual interop across actual browsers, which we
> should prove with testing across different browsers.

Right, that makes sense. I just think that we should be careful to 
distinguish the two. The WPT repo would normally only get content that 
can be used if you're testing on your own, IMHO. This doesn't at all 
mean that it can empower actual interop testing events, e.g. by having 
the tests somehow configurable to also work in a multi-browser situation.

I say this because it has implications on implementation for WPT; it 
doesn't mean that we can't solve both.

Robin Berjon - - @robinberjon

Received on Wednesday, 23 October 2013 13:06:06 UTC