[fetchmail]Need old versions of fetchmail, raising objection against Patent DE 10243243 B4

[Matthias Andree]

Jan Kechel <jan@kechel.de> writes:

> i am going to raise an objection against T-Mobile's Software-Patent  DE=20
> 10243243 B4.
> Yes, it's a german patent-application, but it try to translate it to=20
> english:
>
> german original:
> "Verfahren zur empf=E4ngerseitigen automatischen Behandlung von=20
> unerw=FCnschter elektronischer Post in Kommunikationsnetzen"
>
> my english translation:
> "Process for receiver-sided automatically handling of unwanted=20
> electronic mail in communication-networks"
>
> So, i found at=20
> http://www.catb.org/~esr/fetchmail/fetchmail-features.html that:
>
> Since version 3:
> Following SMTP 571 response to a From line, fetchmail no longer=20
> downloads the bodies of spam messages.
>
> Since version 2:
> When forwarding mail via SMTP, fetchmail respects the 571 "spam filter"=20
> response and discards any mail that triggers it.
>
> To rise an objection, i need to give evidence that any publicly=20
> available program, for example fetchmail, already implemented such a=20
> feature before 17. October 2002.
>
> PS: you can get the original patent-application from:
> https://publikationen.dpma.de/lst_pat_xpt.do?query=3DPN+%3D+DE10243243B4
> (it probably works only at your second try because of the=20
> session-management of that site)

Jan Kechel <jan@kechel.de> writes:

> i found out that spam-filtering is not fetchmail's work ..
> I'm going to use bogofilter and Paul Graham's article "A Plan For Spam"=20
> <http://www.paulgraham.com/spam.html> for my objection.
>
> Thanks anyway and sorry for the interruption.

A central item of said patent is claim #1, "eine Analyse auf serienweise
vorhandene inkrementierte Benutzerkennungen" (analysis for incremented
user identifiers present in quantities) - neither Paul's method nor
fetchmail nor bogofilter implement this analysis.

You should still raise the objection now (deadline Apr 26) and perhaps
send in the reasons later, if that's possible - I personally think that
patents on software methods are the wrong way.

Perhaps a poll on Bugtraq and pertinent fora (SPAM-L) can find prior art
to base your objection on.

Let me know off-list how the story ends.

--=20
Matthias Andree