Re: [rtcweb] Finishing up the Video Codec document, MTI (again, still, sorry)

[Harald Alvestrand <[e-mail]>]

On 12/04/2014 03:10 AM, Gaelle Martin-Cocher wrote:
> There was a single hum for the three categories (browser, non-browser, compatible endpoints), right?
>
> That makes a lot of sense for non-browser entities that are in fact WebRTC-compatible enpoints (apps, or gateways or else) to push the burden on the browser vendors as those entities will need to interact with browsers.  Hence "hum"...
>
> I think it would make a lot of sense to have different "hums" or questions for the two different categories.
> That will bring clarity on what the "non-browser" yet WebRTC endpoint is, versus what the WebRTC-compatible endpoint is (let aside gateways).
> It is not clear to me that we would have had the same results for each category if there was two (or three) different questions.

It is not clear that the questions are independent.

In fact, it is pretty clear to me that some participants who were 
willing to go with this package deal were quite unwilling to commit to 
all the pieces of the package if they were presented one at a time, and 
they had to answer the question not knowing how all the other pieces 
came out.

I think this is a textbook case of a "package deal".