Re: RE-VERSION

On Sun, 10 Aug 1997, Scott Lawrence wrote:

> 
>   The requestor has a version, and sends it (which may, at the
>   requestors option for whatever reason, be less than the highest
>   version it could use); the responder has a version and sends it, but
>   the response itself must always be valid according to the rules of
>   version sent by the requestor.  What could be simpler?
> 

Nothing could be simpler or more obvious! That is why a number of
people have been arguing this is what the spec SHOULD say.  
Unfortunately, it says nothing of the kind and Scott like many
people before him has been confused by it.

There is already an entire RFC devoted to trying to explain the
meaning of the version header precisely because the meaning isn't
the "obvious" one.  Let me quote:


   "An HTTP client SHOULD send a request version equal to the highest
   version for which the client is at least conditionally compliant, and
   whose major version is no higher than the highest version supported
   by the server, if this is known. ...

   An HTTP client MAY send a lower request version, if it is known that
   the server incorrectly implements the HTTP specification, but only
   after the client has determined that the server is actually buggy."

Thus, sadly, Scott's assertion 

  "The requestor has a version, and sends it (which may, at the
  requestors option for whatever reason, be less than the highest
  version it could use)"

is not true.  The requestor can only send a lower version number
if it knows the server is buggy.


John Franks 	Dept of Math. Northwestern University
		john@math.nwu.edu

Received on Sunday, 10 August 1997 07:37:21 UTC