Re: [webrtc-stats] Should there be an API version flag in the stats?

I think this needs wider discussion, so adding my 2 cents.

Currently the UAs return widely different structures and metrics. So, we
run daily tests on WebRTC API to check if anything changed, and what the
differences are (new stats, data type changes, API additions, etc). I know
a lot of people run these kind of checks. So adding a version number would
indeed simplify that process.

However, what is the expectation when  the version number changes? that the
getStats() API returns a particular type of structure, the data types
returned by the metrics are conformant (double instead of int)? and/or all
MTI metrics are implemented?

Instead of UAs changing their versions arbitrarily, perhaps using the dated
version of the spec that was implemented by the UA might help.

On Thu, May 26, 2016 at 6:03 PM, Harald Alvestrand via GitHub <
sysbot+gh@w3.org> wrote:

> alvestrand has just created a new issue for
> https://github.com/w3c/webrtc-stats:
>
> == Should there be an API version flag in the stats? ==
> When an UA's interface changes, it would be nice to have a safe way of
>  detecting that a change has happened. Should the implementation be
> encouraged to expose some kind of "version number"?
>
> Please view or discuss this issue at
> https://github.com/w3c/webrtc-stats/issues/49 using your GitHub
> account
>
>


-- 
Founder, CEO, callstats.io
http://www.callstats.io
Analytics and Optimizations for WebRTC.

We are hiring: www.callstats.io/jobs/

Received on Tuesday, 7 June 2016 10:19:36 UTC