- From: Willy Tarreau <w@1wt.eu>
- Date: Fri, 18 Jan 2013 20:55:34 +0100
- To: Phillip Hallam-Baker <hallam@gmail.com>
- Cc: James M Snell <jasnell@gmail.com>, "ietf-http-wg@w3.org" <ietf-http-wg@w3.org>
On Fri, Jan 18, 2013 at 02:40:41PM -0500, Phillip Hallam-Baker wrote: > Why do HTTP request messages have dates in them anyhow? > > If they do not cause a state machine to behave differently then lets get > rid of them. > > Dates that are content metadata are useful but putting the time in a > synchronous protocol has always seemed rather silly to me. The only use I > can see is to find out if the requester can config their machine to have > the right TOD on the clock. If that is all we are worried about then DNS > type 32 bit dates would be fine. DateTime in the DNS actually wraps every > UNIX epoch. We could also have a date relative to the moment it's emitted. That way, we get rid of the synchronization issues and risk of overflows while keeping a short format (32 bit). Just an idea. Willy
Received on Friday, 18 January 2013 19:56:02 UTC