Intent to unship: LocalMediaStream

[Andreas Pehrson]

I would like to unship the MediaStream subclass LocalMediaStream and its
stop method.

Firefox is the only major browser to still ship LocalMediaStream. It is the
type we use for the MediaStream that is returned from getUserMedia.

The most breakage I expect is for the removal of the stop method, but for
this we shipped a console deprecation warning in Firefox 44 (bug 1103188).

The alternative to stopping the MediaStream from getUserMedia is to stop
all the tracks that were originally present in it through the stop method
on MediaStreamTrack. We shipped this method in Firefox 34 (bug 1057955).

This work is tracked in bug 1258143.
https://bugzilla.mozilla.org/show_bug.cgi?id=1258143

[Mike Taylor]

On 10/11/18 9:16 AM, Andreas Pehrson wrote:
> I would like to unship the MediaStream subclass LocalMediaStream and its
> stop method.
>
> Firefox is the only major browser to still ship LocalMediaStream. It is the
> type we use for the MediaStream that is returned from getUserMedia.
>
> The most breakage I expect is for the removal of the stop method, but for
> this we shipped a console deprecation warning in Firefox 44 (bug 1103188).

Do we have telemetry on usage of stop? If not, can we collect it?

In my experience, developers aren't really paying attention to
deprecation warnings in the console (...especially the ones that don't
test in Firefox).

--
Mike Taylor
Web Compat, Mozilla

[Andreas Pehrson]

We don't have telemetry for this. On the other hand we're the last ones
still implementing this, so that takes care of the people that don't test
in Firefox that you mention.

FWIW Chrome removed this in 47. That's three years ago. And Chrome tends to
be what most services develop and test against when it comes to webrtc
(with its still actively changing spec).
https://bugs.chromium.org/p/chromium/issues/detail?id=338500

In light of this, do you still think we need to add telemetry?

[Mike Taylor]

On 10/11/18 11:13 AM, Andreas Pehrson wrote:
> We don't have telemetry for this. On the other hand we're the last ones
> still implementing this, so that takes care of the people that don't
> test in Firefox that you mention.
>
> FWIW Chrome removed this in 47. That's three years ago. And Chrome tends
> to be what most services develop and test against when it comes to
> webrtc (with its still actively changing spec).
> https://bugs.chromium.org/p/chromium/issues/detail?id=338500
>
> In light of this, do you still think we need to add telemetry?

I think if Chrome was able to successfully remove, there's a lot less risk.