Agree. I think all the existing methods can be polyfilled/retained for some period of time, since they all thunk trivially to the new methods we are proposing. We've survived worse flag days already (e.g. PeerConnection -> RTCPeerConnection) without major incidents. On Thu, Nov 14, 2013 at 10:18 AM, Martin Thomson <martin.thomson@gmail.com>wrote: > On 14 November 2013 07:55, Adam Roach <adam@nostrum.com> wrote: > > Are we planning a story here around backwards-compatibility? Or are you > > envisioning some kind of flag-day on which all the existing WebRTC apps > > simply stop working? > > That's a good point. As originally proposed, the concept of a flag > day was not necessary. The idea is to add the proposed items to > existing interfaces, without removing any existing points. As Justin > observes, polyfill is possible for a good number of existing > methods/attributes, so retaining them in browser interfaces is > trivial. > >
Received on Thursday, 14 November 2013 20:32:01 UTC