I'm just going to chime in because I think Mark wants voices to assess consensus at least. Like back at the SF interim, I don't care too much about this. I think re-framing is not a big deal. But if vendors want to avoid reframing, I don't mind enabling them to do that. That said, I don't see that it's necessary to move the application data/headers length check to the HTTP layer, rather than the framing layer. The information is all there in the framing layer. I think the main reason to move it to the HTTP layer is if we want a HTTP/2 framing layer intermediary not to have to do this length check. I understand that Roberto thinks this is desirable. It was not clear to me if Jeff did too. I don't feel too strongly about this either, but mostly just wanted to call it out. Again, I basically don't really care :) As long as we still have the checks somewhere and we still have that 14 bit data length check, then I'm fine with this. Cheers. On Mon, Apr 21, 2014 at 2:07 PM, Jeff Pinner <jpinner@twitter.com> wrote: > On Mon, Apr 21, 2014 at 2:06 PM, Martin Thomson <martin.thomson@gmail.com>wrote: > >> On 21 April 2014 13:20, Jeff Pinner <jpinner@twitter.com> wrote: >> [lots] >> >> > Best quoting ever :) >
Received on Monday, 21 April 2014 21:59:16 UTC