- From: Tatsuhiro Tsujikawa <tatsuhiro.t@gmail.com>
- Date: Sat, 22 Feb 2014 00:48:00 +0900
- To: Cory Benfield <cory@lukasa.co.uk>
- Cc: "ietf-http-wg@w3.org" <ietf-http-wg@w3.org>
- Message-ID: <CAPyZ6=K=3UBm-L8bNYk=8rKgtUMdLcDQLGhwHRt8KKrP8sFK9g@mail.gmail.com>
Hi, On Fri, Feb 21, 2014 at 11:45 PM, Cory Benfield <cory@lukasa.co.uk> wrote: > On 21 February 2014 14:36, Tatsuhiro Tsujikawa <tatsuhiro.t@gmail.com> > wrote: > > We have 17 implementations so far. Are there anyone who use debug field > in > > GOAWAY in a meaningful way? > > Hyper doesn't use it, and the most it'd ever do with it is throw it in > exceptions or print it in logs: I doubt it will ever take action on > it. > > However, on a purely selfish level I don't object to having it there. > It doesn't add any significant complexity to my code or to my users' > code, and there's no spec requirement that I pay any attention to it > at all. From a pure implementation standpoint I treat that field as > opaque data, and so I'm happy to take it or leave it. > > Of course, if someone has a philosophical reason to remove it or keep > it in I'll happily defer to their wisdom. > > Cory > My argument is that if debug field is just there and not cared by the receiver and is just discarded, then there is no point to keep it in the protocol specification. If you cannot expect anything in that field from the other implementations because it is opaque and completely undefined. On the other hand, if a person or group is developing a pair of client and server, then their own logging is much more useful in my opinion. Best regards, Tatsuhiro Tsujikawa
Received on Friday, 21 February 2014 15:48:48 UTC