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

Re: Common Gateway Interface

From: Hallam-Baker <hallam@ai.mit.edu>
Date: Fri, 30 May 1997 13:38:50 -0400 (EDT)
Message-Id: <199705301738.NAA14651@muesli.ai.mit.edu>
To: Dylan Barrell <dbarrell@hotmail.com>
Cc: hallam@ai.mit.edu, dbarrell@hotmail.com, http-wg@cuckoo.hpl.hp.com
X-Mailing-List: <http-wg@cuckoo.hpl.hp.com> archive/latest/3379
> I didn't say that we should necessarily standardise the CGI implementation but 
> we should specify which header fields MAY be passed, which MUST be passed and
> which MAY NOT be passed to CGI, NSAPI, ISAPI or JOHNDOESAPI. I also beleive 
> every RFC concerning HTTP or an extension to it which adds to the headers 
> should deal with this issue.

I don't think we should consider the issue at all. Not every API passes
headers on. High level APIs like the Common Lisp server mean that the
user does not need to even know about the headers.

There is absolutely no reason to require any header to be passed at all.
Contrawise there is no reason not to allow a calling module access to 
the raw data stream which many if not most do already.

> Failure to do this will force each Internet, Intranet and Extranet developer to 
> write his own HTTP daemon in order to be assured of having access to all the 
> information she needs. I would prefer them to be able to use standard HTTP 

??? I don't think thats the case at all and if it was it would be a case
for a different working group, possibly even a different standards body.

Since this is a principle competitive issue between the server companies
you have about as much chance of getting them to agree to a standard as
you have of solving the middle east problem.

Received on Friday, 30 May 1997 10:42:48 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 14:40:20 UTC