RE: MSE testing status update

We have done track implementations, but they aren’t currently passing the tests.  We intend to resolve these, either by updating the tests (assuming that’s where the problem is) or updating our implementation.

From: Matt Wolenetz [mailto:wolenetz@google.com]
Sent: Tuesday, August 11, 2015 8:40 AM
To: Michael[tm] Smith <mike@w3.org>; Paul Cotton <Paul.Cotton@microsoft.com>
Cc: Jerry Smith (IEP) <jdsmith@microsoft.com>; <public-html-media@w3.org> <public-html-media@w3.org>
Subject: Re: MSE testing status update

Regarding the SourceBuffer.{audio,video,text}Tracks implementation corresponding to the spec, in Chrome we're tracking that with https://crbug.com/249428. While I agree that in the current state, we could not transition out of CR, I am expecting that Chrome at least will get these implemented.

Jerry, what do you think the prognosis is for Edge support for this portion of the MSE API? I hope this feature is not something we need to mark at risk in the spec.

Note, the other MSE-interface related Chrome bug mentioned during today's MSE teleconference was mentioned earlier in this thread: Chrome is failing a significant number of MSE w3c web-platform-tests interface.html tests. This is tracked in Chrome with https://crbug.com/504170.)

On Wed, Jul 8, 2015 at 5:30 AM Michael[tm] Smith <mike@w3.org<mailto:mike@w3.org>> wrote:
Paul Cotton <Paul.Cotton@microsoft.com<mailto:Paul.Cotton@microsoft.com>>, 2015-07-06 19:41 +0000:
> Archived-At: <http://www.w3.org/mid/BN3PR0301MB118535D834D4DD5A67AF3CB2EA930@BN3PR0301MB1185.namprd03.prod.outlook.com>
>
> > Many of the *Track interface pieces are yet to be implemented in Chrome,
>
> Are these currently interfaces implemented in any other MSE implementations?

I believe they are not supported in an UAs at all yet.

Certainly we don’t have two UAs that support them.

Given that, it’s not clear to me why they weren’t listed as “at risk” in
the CR publication. We’re definitely not going to be transition the spec
out of CR if we can’t document that we have at least two implementations of
those features (and we actually don’t even have tests for those features
themselves yet—all we have at this point is just the simple test that
checks whether the interface exposes members for them as expected).

One option of course is to drop them from the spec (and move them to a v2)
and take just the sets of features to Rec for which we can demonstrate that
we have at least two implementations.

  —Mike

--
Michael[tm] Smith https://people.w3.org/mike