W3C home > Mailing lists > Public > ietf-http-wg-old@w3.org > May to August 1997

Re: HTTP status codes

From: Joel N. Weber II <devnull@gnu.ai.mit.edu>
Date: Wed, 23 Jul 1997 16:55:35 -0400
Message-Id: <199707232055.QAA21541@mescaline.gnu.ai.mit.edu>
To: hgs@cs.columbia.edu
Cc: http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com, confctrl@isi.edu
   Sender: hgs@dnrc.bell-labs.com
   Date: Wed, 23 Jul 1997 16:17:22 -0400
   From: "Henning Schulzrinne (BL)" <hgs@cs.columbia.edu>
   Organization: Columbia University

   SIP and RTSP re-use a number of HTTP status codes, among other
   properties. It is likely that they may need to or want to adopt other
   HTTP status codes that emerge in the future. Thus, it is desirable that
   the SIP and RTSP-specific status codes do not conflict with HTTP codes.

   One possible solution: Declare officially that HTTP will only use status
   codes up to x49 (say) and leave others for private extensions, including
   SIP and RTSP.

There may be things other than SIP and RTSP, and it may be the case that
those other protocols will define status codes which are useful in
SIP and RTSP and HTTP, so it might be better to do something like this:

x00 through x49 is for HTTP
x50 through x59 is for SIP
x60 through x69 is for RTSP
and then we can assign blocks of ten to a few more protocols later.

another possibilty is to state that x00 through x49 are for things defined
by the HTTP spec authors, and x50 through x99 are for others, and have
one central authority to control allocation of those numbers in the
x50 through x99 range.  (It is probably desireable to keep the status
codes defined by HTTP consecutive.)

But whatever we decide to do, I don't think SIP and RTSP should be
considered `private extensions'.  Perhaps `defined by people other
than the authors of the HTTP spec' is an OK description, though.
Received on Wednesday, 23 July 1997 13:58:38 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 24 September 2003 06:32:49 EDT