From: Charles Lindsey (chl@clerew.man.ac.uk)
Date: Fri Apr 16 2004 - 17:13:18 CDT
In <407DE495.9020403@erols.com> Bruce Lilly <blilly@erols.com> writes:
>Charles Lindsey wrote:
>> In <406CC294.5090302@erols.com> Bruce Lilly <blilly@erols.com> writes:
>>> Neither the term nor reasons
>>>why the process might legitimately take place are defined.
>>
>>
>> It is going to happen, whether it is legitimate or not.
>If it is not legitimate, then we should state so clearly. If there are
>known instances where it happens regardless, we should be especially
>clear that those cases are forbidden.
Why? There is no reason to forbid it, so long as it is done in accordance
with whatever rules we specify. Our job is to provide rules that will
enable the Usenet community to function. Putting up arbitrary barriers
just to show that "we make the rules and we are in charge" will do nobody
any good (and such rules would be ignored anyway).
>If the article in question has been successfully relayed, and is refused
>relaying by other sites, the server has finished it's job -- there is no
>cause for "re-injection". If the article has never been successfully
>relayed, there is no harm in presenting it again as a proto-article
>for injection -- the alternative is that the article simply does not
>propagate beyond the server in question, in which case the user may notice
>that lack of propagation and manually resubmit his article, which will
>have the same effect.
Yes, but in the case where this sort of thing is likely to be important
(examples I am aware of include control messages and important announce
groups) the actual relaying/posting is likely to be done by some automated
script whose failure may not be immediately apparent to the poster. It is
quite normal for such articles to be inserted into the network at two
sites, and if one of them is expecting relaying and the other is expecting
posting, then reinjection may well be the inevitable consequence.
>That is utter nonsense; if the article (N.B. not proto-article) cannot
>be relayed (why on Earth is A trying to relay to B, when it is an established
>fact that B does not accept relayed articles from A?) then that is the end
>of the transaction between A and B regarding that article. There is again
>no justification for "re-injection".
Because if you had been following the discussion we had a month ago, you
would have seen that life on the real Usenet is not so clean and tidy as
you so fondly imagine, and there are indeed examples of sites which are
prepared to do exactly that (for example by permitting the use of the NNTP
IHAVE command by anyone). All sorts of strange practices go on out there
in the real world, and some of them provide a real service to the
community (and others are, of course, just plain broken).
Anyway, I have now posted the actual modifications to the draft needed
for the Injection-Date header, so let us continue the discussion in terms
of what I have actually written.
-- Charles H. Lindsey ---------At Home, doing my own thing------------------------ Tel: +44 161 436 6131 Fax: +44 161 436 6133 Web: http://www.cs.man.ac.uk/~chl Email: chl@clerew.man.ac.uk 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