Re: [codec] #5: Mention DTMF in requirements

[Peter Saint-Andre <[e-mail]>]

How is the never-ending debate among DTMF signalling *methods* in-scope
for the Codec WG? I think that Henning brought this up in Anaheim only
to make sure that we test some DTMF tones. The signalling method is out
of scope for the codec itself.

On 4/2/10 7:53 AM, stephen botzko wrote:
> Are you two suggesting that in-band DTMF is a MUST?  Or alternatively a
> SHOULD?
> 
> Stephen Botzko
> 
> On Fri, Apr 2, 2010 at 9:48 AM, James Rafferty
> <[e-mail] <mailto:[e-mail]>> wrote:
> 
>     I'd agree with Steve that are still many deployments which do not
>     use RFC 2833 or RFC 4733. In our gateways, we've had to support
>     interworking variations of tone support such as INFO and in-band, in
>     addition to the RFC 2833 / RFC 4733.
> 
>     James
>     -----Original Message-----
>     From: [e-mail] <mailto:[e-mail]>
>     [mailto:[e-mail] <mailto:[e-mail]>] On
>     Behalf Of Steve Underwood
>     Sent: Friday, April 02, 2010 2:27 AM
>     To: [e-mail] <mailto:[e-mail]>
>     Subject: Re: [codec] #5: Mention DTMF in requirements
> 
>     On 03/29/2010 02:22 AM, Marc Petit-Huguenin wrote:
>     > On 03/28/2010 11:00 AM, stephen botzko wrote:
>     >
>     >> I would agree with this if I saw reasonable evidence that a
>     >> preponderance of gateways and sending systems provide the
>     signaling in
>     >> these RFCs.
>     >>
>     >> Since I am not sure that this is the case, I am unconvinced that
>     we can
>     >> totally remove the requirement.
>     >>
>     >> I'd also say that an encoder that detects the DTMF tones and
>     outputs the
>     >> RFC 4733/34 events would fully meet the requirement.
>     >>
>     > As former CTO of a VoIP provider, I never saw a PSTN provider not
>     supporting at
>     > least RFC 2833 (even if one of them did not declare it in its SDP)
>     >
>     > Perhaps the question can be asked at the next SIPit event.
>     >
>     Its true that RFC2833 is widely deployed. Its even true that many
>     systems have updated to RFC4733. Sadly, its also true that there are
>     still many quirky implementations widely deployed, and a lot of people
>     still need to interwork with audio DTMF.
> 
>     Steve