The point is not whether the chunked encoding is in or out of HTTP/1.1 In fact I would if anything prefer it to be out if anything. What I was reacting to was a suggestion that the footers be dropped from chunked and put in HTTP/1.1. This is a change I can see absolutely no technical justification for whatsoever. I also fail to see any political advantage which would mean getting the spec to proposed standard earlier. Thinking that reports of such a change might have been subject to a certain degree of confusion I thought it best to send a private message to the chair rather than exciting the whole list over what could well have been a simple misunderstanding or confusion. If people were indeed worried by any additional complexity involved in the handling of footers they need only construct a very simple FSM to strip out the additional items. What I am concerned about that in a rush to get 1.1 out the door decisions will be made which will cause us compatibility problems in 1.2. It would be something of an embarassment to have to introduce a second encoding for no other reason than the first one had been fumbled. In summary I suggest that we ensure that wherever possible the 1.1. spec requires implementations to be tolerant in accepting and call to attention the areas where extension is likely. PhillReceived on Saturday, 9 March 1996 11:21:44 UTC
This archive was generated by hypermail 2.4.0 : Thursday, 2 February 2023 18:42:57 UTC