tea.mathoverflow.net - Discussion Feed (Spamassasin marks answer email as spam)

joro comments on "Spamassasin marks answer email as spam" (20080)

joro — Fri, 07 Sep 2012 06:23:15 -0700

@Willie I have no pain and don't complain. Just point out that a major anti spam software flags MO's email - this might affect other users too. This is completely OK for me, I am sensitive to spam false positives and waste more than average time deleting spam.

WillieWong comments on "Spamassasin marks answer email as spam" (20079)

WillieWong — Fri, 07 Sep 2012 06:11:11 -0700

@joro: while I feel your pain, instead of fighting the times, I've found it much much simpler to just tell my MUA to ask nicely for a nicely parsed output from lynx or something like that. (Mutt with autoview text/html works pretty well).

But you are of course right: having a text/plain part that just tells you to see the HTML part defeats the entire point of multipart MIME; it'd be better even to just have no text/plain parts compared to having one that is useless.

For that matter, I wonder if the e-mails in SE2 are any better?

joro comments on "Spamassasin marks answer email as spam" (20077)

joro — Fri, 07 Sep 2012 05:13:41 -0700

I don't mean to criticize MO, but the answer email doesn't follow email best practices in my conservative opinion. I suppose best practices are the text/plain part of the message to be the html stripped from formatting. On all the mailing lists I am, html only email (or text/plain "see the html") is extremely rare and always criticized. Even friends who send me html only email annoy me. Not to mention my MUA doesn't understand html :-)

Will Jagy comments on "Spamassasin marks answer email as spam" (20068)

Will Jagy — Thu, 06 Sep 2012 11:19:40 -0700

Stupid but relevant: a friend of mine at UT San Antonio told me that his department email marked almost anything as spam if the subject line contained punctuation. Probably the main exception would be Re: So I began deleting punctuation from subject lines when i send, to anyone, and things got a little better. But you can't win. Our own Suvrit was planning to visit Beresford Parlett at Berkeley. i told Suvrit to also call (he was visiting Stanford already) but he didn't. So when I called Parlett, it was all a surprise to him. It turned out that there had been changes to Thunderbird about two weeks before, and he had a number of important messages in spam.

joro comments on "Spamassasin marks answer email as spam" (20065)

joro — Thu, 06 Sep 2012 01:46:21 -0700

In my case the software is run on the ISP mail server so there is no feedback to the filter.

I suppose the main bayesian problem is the text/plain part of the email is too short and including the markdown source as text/plain will fix the issue.

WillieWong comments on "Spamassasin marks answer email as spam" (20055)

WillieWong — Wed, 05 Sep 2012 08:51:08 -0700

Isn´t the Bayesian filter (which generate the majority of your Spam score) trainable by the user?

I am not exactly sure how MO can change the software (assuming we can even do that) to evade all Bayesian SPAM filters, since that heavily depends on how a user/sys-admin trains the filter....

joro comments on "Spamassasin marks answer email as spam" (20049)

joro — Wed, 05 Sep 2012 04:25:11 -0700

Spamassasin marks answer email as spam

FYI: Got email notification which was marked as spam by the widely used anti spam software spamassasin.

Here is the log:

  From: MathOverflow 
  Subject: *****SPAM***** 1 Question Has 1 Answer - MathOverflow
  X-Spam-Checker-Version: SpamAssassin 3.2.1 (2007-05-02)
  X-Spam-Status: Yes, score=9.3 required=6.2 tests=BAYES_99=6.1,HTML_MESSAGE=1.8,
    MIME_QP_LONG_LINE=1.396 autolearn=no version=3.2.1
   X-Spam-Report:
    *  6.1 BAYES_99 BODY: Bayesian spam probability is 99 to 100%
    *      [score: 0.9963]
    *  1.8 HTML_MESSAGE BODY: HTML included in message
    *  1.4 MIME_QP_LONG_LINE RAW: Quoted-printable line longer than 76 chars

I can live with this and don't complain, but more aggressive spam settings might bury the email deeper.