On Tue, Sep 13, 2016 at 1:46 PM, juan <juan.g71@gmail.com> wrote:
On Tue, 13 Sep 2016 17:47:42 +0000
Sean Lynch <seanl@literati.org> wrote:


> I used to have a filter that prevented Gmail from marking cpunks mail
> as spam, but I have intentionally avoided fixing that since the list
> address changed in hopes of training it not to suck so bad. So far I
> don't seem to have had much luck.


        So apparently the system ignores the fact that messages come
        from a mailing list, and treats them as if they came directly
        from the original sender's address. And then wrongly marks some
        as spam. So it's doing it wrongly, twice.

I'm not sure how manually blocking based on mailing list headers makes any sense. And there are plenty of mailing lists that get lots of spam. This used to be how cpunks worked when it was distributed, as has been discussed here recently. The end user was expected to run a spam filter against the list, subscribe to a filtered mirror, or just put up with it.
 
        Adding a rule to stop it filtering messages with trait
        'comes-from-cypher-whatever' works, but it seems to be slightly
        misaligned with the modern philosophy of "automatic system
        built for retards, I mean, for Important People Too Busy To
        Learn Anything".

Well, the "retards" probably aren't going to care much about a few mailing list messages being sent to their spam folder erroneously. I can't say I've actually cared about any of the messages that got sent to spam, I'm just pedantic about such things. Today, though, the only "false positive" was someone I would have blocked anyway. So is it really a false positive?