[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Nmh-workers] Stanford disliking my emails -- update + question

From: Bob Carragher
Subject: Re: [Nmh-workers] Stanford disliking my emails -- update + question
Date: Thu, 23 Apr 2015 18:03:53 -0700

On Thu, 23 Apr 2015 11:00:42 +0100 Ralph Corderoy <address@hidden> sez:

> Hi Ken,
> > Thirdly ... I think it's ridiculous that Stanford's anti-spam
> > rules trawl through Received headers (which are defined as
> > being free-form) and look for suspicious hostnames when
> > you've already sent that email through your email provider
> > and it has a valid DKIM header; gmail has already certified
> > that the email came from an authenticated user, why does
> > Stanford care what your local hostname is?
> I suspect "Home IP address (in San Jose) sending via LHLD to
> gmail" from it is just one factor in its ham/spam decision.
> Bob says with "Received:  from ayukawa" some are ham, but some
> are still spam.  Perhaps that means it made little difference.

It did make some difference.  Previously, posts to Stanford
mailing lists sent using NMH sometimes were silently discarded.
Mostly, they were spam-tagged.  Now, they get through, with
reduced or no tagging.  (Oddly, though, when I send from GMail's
web interface, some of them are *still* discarded.  So, you're
correct, Ralph:  this wasn't the *entire* cause of the problem.)

> Seems odd so much of Bob's mail to Stanford is considered spam.
> Wonder what he's writing about.  :-)  Can Stanford recipients
> seems Bob's emails in their spam folder and train Proofpoint
> that they're ham?

I'm a dancer, not a spammer!  B-)  (The messages are about dance
events or dance-related topics.)

The Proofpoint setup is such that, assuming the message wasn't
discarded, recipients see tagging in the Subject: line, like:

     Subject: [SPAM:###] I am not spam!

The number of hashtags ("#") indicates the level of suspicion, up
to 5 hashtags -- though it's not a linear relationship.  One #
indicates a suspicion score of 0.5 (max: 1.0); further #s
indicate an additional 0.1 on the score.

Stanford provides a web interface for @stanford.edu recipients,
although there's a way to register your own (and, independently,
have your email forwarded on).  If students don't change any
default settings, then messages with 5 #s are placed in their
spam folder, while those with 4 #s or fewer reach their inbox.  I
wouldn't be at all surprised if students never change them.

Deceptively (in my opinion), their "help" page on spam implies
that they can indeed train the system.  But I confirmed with
Stanford IT that (at present) only the IT people can forward
messages to Proofpoint for training.  Which means that the only
way for me to reduce the spamminess of my messages is to modify
and experiment with them myself.  B-[  The lorem ipsum generator
website might help:



reply via email to

[Prev in Thread] Current Thread [Next in Thread]