Re: [codec] WG Review: Internet Wideband Audio Codec (codec)

["Roni Even" <[e-mail]>]

Hi,
In line
Roni Even

> -----Original Message-----
> From: [e-mail] [mailto:[e-mail]]
> Sent: Thursday, December 24, 2009 11:47 AM
> To: Roni Even
> Cc: 'Tschofenig, Hannes (NSN - FI/Espoo)'; [e-mail];
> [e-mail]; [e-mail]
> Subject: Re: WG Review: Internet Wideband Audio Codec (codec)
> 
>     Date:        Thu, 24 Dec 2009 08:50:30 +0200
>     From:        "Roni Even" <[e-mail]>
>     Message-ID:  <[e-mail]>
> 
>   | I am not sure but are you suggesting that the IETF will define the
>   | requirements, metric and quality assessment requirements and all
> proposed
>   | codecs should provide the results and then the WG will choose the
> best codec
>   | bases without discussing the codec itself. This is what I would
> call a
>   | selection process (at least in ITU terms).
> 
> The WG can decide how it wants to go about the process, I'd just prefer
> that
> the charter not (effectively) rule out selection of something that
> already
> exists with an assumption that something entirely new will be created.
> 
>   | The problem is that the IETF process allows anyone to contribute to
> existing
>   | work hopefully leading to a better the end result.
> 
> Of course, but also be aware that there's no one definition of
> "better".
> Something that can be defined quickly and used immediately might be
> much
> better than something it takes 5 years to create and more to implement,
> even if the invented one saves a little bandwidth or has better loss
> recovery characteristics.

This is the IETF process for better or worse, I asked similar questions and
the response is that the IETF decide what is better is based on rough
consensus. 
BTW: my personal view is that your suggestion is in line with the process at
the ITU when doing codec selection, but there are people who prefer doing it
at the IETF using the IETF procedures for other reasons. 


> 
>   | What about the change control, does it stay with the original
> contributor or
>   | can the IETF modify the codec based on input from other parties,
> which means
>   | that the codec may change by the IETF anyhow.
> 
> The IETF will have change control over its protocol, of course, which
> may
> cause it to diverge from that upon which it was originally based.  And
> yes,
> everything changes with time.
> 
> kre