W3C home > Mailing lists > Public > ietf-http-wg@w3.org > October to December 2020

Re: How to express no matching results in HTTP SEARH method?

From: Poul-Henning Kamp <phk@phk.freebsd.dk>
Date: Wed, 04 Nov 2020 13:24:57 +0000
To: Julian Reschke <julian.reschke@gmx.de>
cc: ietf-http-wg@w3.org
Message-ID: <63989.1604496297@critter.freebsd.dk>
--------
Julian Reschke writes:
> Am 04.11.2020 um 03:40 schrieb Sawood Alam:

> > cases. Also, 204 is a success response, which means a client might
> > repeat the request, while a 4xx response would suggest that there is no
> > point in repeating the request without making changes to it.
>
> If the request target exists, and accepts and processes the query, the
> response ought to be a 2xx, no?

Any librarian will tell you that "Found" is much simpler than
"Not found", which has a lot of possible semantics:

	Not found, though it should have been.

	Not found, and will never be found.

	Not found, but will certainly be found sometime in the future [($time)].

	Not found, and cannot possibly be found until ($time)

	Not found here, but try asking ($uri, [$query_params])

	Not even looking, you have been asking too many queries recently

	Not telling, (temporary/permanent/authorization needed)

	Found, but you cant have it (temporary/permanent/authorization needed)

	...

It might be a good idea to hash out a couple of general mechanisms
in the spec, to provide mechanisms for traffic/load-engineering,
at the very least something like "Dont-repeat-this-SEARCH: <seconds>" ?

-- 
Poul-Henning Kamp       | UNIX since Zilog Zeus 3.20
phk@FreeBSD.ORG         | TCP/IP since RFC 956
FreeBSD committer       | BSD since 4.3-tahoe    
Never attribute to malice what can adequately be explained by incompetence.
Received on Wednesday, 4 November 2020 13:25:14 UTC

This archive was generated by hypermail 2.4.0 : Wednesday, 4 November 2020 13:25:15 UTC