Re: [rtcweb] Is there room for a compromise? what about no MTI?

[Dave Crocker <[e-mail]>]

On 12/22/2013 5:40 PM, Bjoern Hoehrmann wrote:
> If you order a product with "WebRTC" and get a text pager or
> ringphone that do speak the wire protocol, you might well feel
> mislead because you associate "WebRTC" with smartphone video
> conferencing or whatever.


This is an essential point.  It highlights the human factors/marketing 
reason that profiling is problematic:  a single label for a broad set of 
functions that aren't always all provided is misleading.

This is the reason that each type of capability needs its own, 
distinctive name.

This warrants repeating:  To the extent that different styles of use 
will mean that some features are not supported, a single generic name 
will be misleading.

Perhaps not surprisingly, the incremental approach to specification
lends itself nicely to this issue, whereas profiling does not.

And to the extent that folks think subsetting (profiling) is not the
issue, please re-read some of the language being used, such as:

> a statement indicating that devices that can't send or receive a
> given media type need not concern themselves with the MTI codecs of
> that type.

That's profiling.

On the other hand, distinctive labels, such as

    1.  rtcweb base

    2.  rtcweb video

    3.  rtcweb audio

    4. rtcweb audio + video

are names that distinguish one product type from another.  (I don't care 
what the actual names are, as long as each distinguishes the set of 
features from other, different sets.)

d/

-- 
Dave Crocker
Brandenburg InternetWorking
bbiw.net