Re: The abysmal state of interoperability for the m-section of SDP

Hi Nils,

Yes, I apologise, we haven't moved away from the track IDs yet, so while
Firefox is spec compliant, we would have preferred a bridging
implementation with the old function available for longer. We tend to move
slowly to the latest implementations for two reasons:

* Not all browsers have implemented it yet, therefore we would move from a
functioning system to a non functioning system.

* Old browser versions are around much more than you might expect - e.g. we
support a hospital that still has Chrome 65 as their standard version.

Transitioning from one version of doing things to another while being in
production is hard when you have a diverse interoperability problem where
some of the browsers you rely on move faster than others. I guess what this
means is that it's now time for us to move. 😀

This report wasn't so much a criticism of Firefox as it was an experience
report from the trenches to show that reality moves slower than you think
and supporting backwards compatibility is worth more than you think.

I apologise I misspoke on Firefox who are indeed following the standard of
today - it's us who haven't caught up with that yet.

Kind regards,
Silvia.


On Wed., 4 Sep. 2019, 1:35 pm Nils Ohlmeier, <nohlmeier@mozilla.com> wrote:

> Hello Silvia,
>
> > On 24Aug, 2019, at 22:28, Silvia Pfeiffer <silviapfeiffer1@gmail.com>
> wrote:
> >
> > In the past, we had no interoperability between Firefox and Chrome
> > because Chrome supported Plan B (which incidentally worked really
> > well) and Firefox did Unified Plan. We waited patiently for Chrome to
> > support Unified Plan and rejoiced when it came. But we were out of
> > luck. Moving to Unified Plan with Chrome didn't make sense for us,
> > because Firefox decided to make secondary streams not use the ID that
> > is passed in, but use something else altogether. So we held back,
> > because we still couldn't make it work interoperably. (Thanks Firefox
> > for not following the standard.)
>
> Since you are claiming that Firefox is not following the standards here it
> would be very helpful if you could point us at the spec which we are not
> following.
>
> It is my understanding that track IDs are no longer guaranteed by the spec
> to be match any SDP m-sections identifiers.
> According to the Unified Plan MIDs are suppose to be used for that. It’s
> hardly Firefox fault that majority of the industry hasn’t switched to that
> (yet).
>
> Best regards
>   Nils Ohlmeier
>