- From: Jim Barnett <Jim.Barnett@genesyslab.com>
- Date: Fri, 10 Aug 2012 10:10:43 -0700
- To: "Harald Alvestrand" <harald@alvestrand.no>, <public-webrtc@w3.org>
"don't know" might be useful as a trigger for negotiation, but ultimately the app has a binary decision to make - use dtmf or not. "don't know represents an uncertainty that needs to be resolved, or that should be treated as equivalent to 'false'. - Jim -----Original Message----- From: Harald Alvestrand [mailto:harald@alvestrand.no] Sent: Friday, August 10, 2012 12:58 PM To: public-webrtc@w3.org Subject: Re: CHANGE: Re: [Bug 18485] New: Change DTMF API to be on PeerConnection On 08/10/2012 05:52 PM, Martin Thomson wrote: > On 10 August 2012 00:12, Harald Alvestrand <harald@alvestrand.no> wrote: >> I assume some responders won't support 4733. So even if all browsers >> support it, canSendDTMF isn't guaranteed to be true always. > Or, as you would have it, canSendDTMF isn't guaranteed to be true even > if it was true in the past. That bothers me a little. > Renegotiation is theoretically able to add or drop support for 4733 at any time, so that is trivially true. We could have canSendDTMF return "don't know" rather than true or false when negotiation isn't done yet. I don't know if that adds value; people have quoted the use case for canSendDTMF as "hide the keypad when canSendDTMF is false"; I'm not sure what is useful here.
Received on Friday, 10 August 2012 17:12:05 UTC