On 13 Oct 1998 17:17:03 GMT,
Dr. Greg Quinn (greg at franklin.burnham-inst.org) wrote:
>Richard P. Grant (see_sig at cmtech.co.delete.uk) wrote:
>: In article <6vtc0v$hfl$1 at franklin.ljcrf.edu>,
>:greg at franklin.burnham-inst.org (Dr. Greg Quinn) wrote:
>:>: : Wouldn't it be a good idea to set up some kind of automoderation
>: : for this (and other bionet) newsgroup, which could simply
>: : consist of a spam filter?
>:>: There is a way to moderate groups so that you have a human moderator who
>: passes or deletes the first message from an address, then lets the 'bot
>: know that a particular address is OK (or not), so all posts from that
>: address will be accepted (or deleted, if need be).
>:>>I think all that's needed is a filter which removes messages that contain
>phrases such as "18 year old nymphos", "naughty" and "make $$$$$$$$$$$ in
>a day". It's no big deal to sift through the spam but it is fairly easily
>avoidable.
I'm not sure if that which I describe below is what Greg had in mind, but
there are "spambots" that automatically delete articles. How they work
is much tied into how the NNTP (Usenet) protocol works.
When a spambot detects an article that matches a keyword in a "delete"
list, it sends an cancel-article for the message-id of the offending
article to a newsgroup called "control". When this cancel-article
propagates to other sites (every Usenet site MUST carry the newsgroup
"control"), they will expire or delete the article in question from their
news-spool.
This mechanism is actively used at several sites to control spam posts
to many non-bionet newsgroups. I do not know the legality of running
a spambot for a bionet newsgroup and am hence reluctant to run one
myself, or offer the resources to run one from.
I principle, I concur that a bionet-spambot would be wonderful.
Ashok
--
Ashok Aiyar, Ph.D.
McArdle Laboratory for Cancer Research
aiyar at ebv.oncology.wisc.edu
