From: Charles Lindsey (chl@clw.cs.man.ac.uk)
Date: Mon Oct 15 2001 - 04:20:20 CDT
In <200110120311.f9C3B9s26498@panix2.panix.com> Seth Breidbart <sethb@panix.com> writes:
>With a cancel that's verified by Cancel-Lock, and which arrives prior
>to the article, the site hasn't yet elected to honour it (after all,
>it might fail the test). Hence there's no problem with the current
>wording; however, it isn't clear enough. It should say something
>about a site not being able to decide about a cancel when it arrives,
>and how to handle that case.
Eh? If a cancel arrives witha Cancel-Key, then you know to wait until the
article arrives so you can check that it fits the Cancel-Lock in the
article.
But the problem is when an article with a Cancel-Lock is cancelled by an
old-style cancel, with no Key or anything. Then you have to decide whether
to honour it in advance of the article (which you would want to do in the
case of a spam cancel) or whether to wait for the article to arrive.
That is an impossible choice, unless one can assume that all spam cancels
will be signed. That is not so at present, but do we agree that is a point
we want to aim for?
Yes, the wording needs to be changed to reflect that dilemma.
-- 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@clw.cs.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