From: Charles Lindsey (chl@clw.cs.man.ac.uk)
Date: Thu Oct 07 1999 - 04:39:31 CDT
In <87k8p0gwgd.fsf@erlenstar.demon.co.uk> Andrew Gierth <andrew@erlenstar.demon.co.uk> writes:
>>>>>> "Charles" == Charles Lindsey <chl@clw.cs.man.ac.uk> writes:
>A relayer has to keep a history file of _some_ sort, even if it is not
>serving articles; otherwise it will waste an awful lot of resource in
>refeeding articles.
> Charles> checks a Date for staleness, what cut-off does it use?
> Charles> Presumably an arbitrary number such as 14 days. Yes, that
> Charles> could be written in, but we would need to agree a number.
>No such number could be agreed, because it depends on the depth of
>one's history file, and one's own preferences; e.g. I have my cutoff
>set to about 7 days, but I keep several more days of history than that
>(because without it I was rejecting too many old articles rather than
>refusing them).
OK, I take your point about relayers keeping history. So I now have the
following wording:
In order to prevent the reinjection of expired articles into the news
stream, relaying and serving agents MUST refuse articles whose Date header
predates the earliest articles of which they normally keep record. Relaying
agents MUST NOT modify the Date header in transit.
That is the strongest statement we could make. Does anyone want to water
it down so as to indicate that such agents should "mostly" do so, or
bychanging the MUST into a SHOULD? Personally, I think it is OK as it is.
-- Charles H. Lindsey ---------At Home, doing my own thing------------------------ Email: chl@clw.cs.man.ac.uk Web: http://www.cs.man.ac.uk/~chl Voice/Fax: +44 161 437 4506 Snail: 5 Clerewood Ave, CHEADLE, SK8 3JU, U.K. PGP: 2C15F1A9 Fingerprint: 73 6D C2 51 93 A0 01 E7 65 E8 64 7E 14 A4 AB A5