Scoring based on lines of quoted material (>)

Scoring based on lines of quoted material (>)

[David Maslen]

Has somebody done, or could they suggest a way to do, a method of
scoring I could use in a mailing list to score down mail that has more
lines starting with > than otherwise.

Generally I find it pretty annoying when somebody quotes pretty much
the entire thread so far, and adds a "me to" comment of 1 line plus a
10 lines of signature on the end of that.

I think I'd like to score up messages where content is greater than
quotation, and penalise heavily a message where content is less than
say 10% of the lines in a message.

I think this will catch the "me to" thread quoters, and the flame bait,
reply to each point of your message type posters.

Re: Scoring based on lines of quoted material (>)

[Kai Großjohann]

David Maslen <david.maslen@iname.com> writes:

> Has somebody done, or could they suggest a way to do, a method of
> scoring I could use in a mailing list to score down mail that has more
> lines starting with > than otherwise.

More and more people seem to be interested in this.  Maybe you could
pool your resources with Frederick and Thilo?

(Frederick and Thilo, I hope you don't mind being mentioned and cc'd
here.)

kai
-- 
~/.signature: No such file or directory

Re: Scoring based on lines of quoted material (>)

[Lars Magne Ingebrigtsen]

David Maslen <david.maslen@iname.com> writes:

> Has somebody done, or could they suggest a way to do, a method of
> scoring I could use in a mailing list to score down mail that has more
> lines starting with > than otherwise.

I think the way to implement this would be to add a `f' "match" type
to the body score thing.  Then the "match" could be a user-defined (or 
pre-defined) function that would return the score of the article.
Then we could have a function that returned a score based on the
ratio between quoted text and new text.

Anybody want to implement this?

-- 
(domestic pets only, the antidote for overdose, milk.)
   larsi@gnus.org * Lars Magne Ingebrigtsen