On 2014-08-20 15:50, Kevin A. McGrail wrote:
On 8/20/2014 6:40 PM, Dave Warren wrote:
On 2014-08-20 14:54, Kevin A. McGrail wrote:
I follow Chris Santere's definition: Spam is about Consent not Content.
I follow a capitalist view that doing business with a company is
implied consent (I usually follow a rule similar to some legal
precedence where you have to have done RECENT business with them,
say 1-2 years at max).
So with that simple set of rules set, here's my interpretation:
#1, not spam. Clear implied consent. Someone hired them and gave
them the email address.
#2, not spam ASSUMING it's a real paystub. Again, Someone hired
them and gave them the email address. If it's a fake paystub, it's
just a phish which is a type of spam.
If we're talking strict consent, none of these have the consent of
the owner of the mailbox -- The recipient doesn't do business with
these senders, it's misdirected rather than malicious.
Do they have the consent of the owner of the domain?
How is this a misdirected email?
Imagine you start receiving paystubs to your email account this Friday,
and receive another every two weeks. They're addressed to "Kennith
McGrail" and mention they're related to your work for a company name of
"Pigs, Cows, Cows and Cows, Inc".
Now that's not your name, or a company you work for, but it's close
enough that the legitimate recipient is possibly [email protected]
or another similar typo. You've previously received stuff from other
humans looking to contact him too, after he gave out your address
instead of his own.
Attempts to notify the company sending the paystubs, Pigs, Cows, Cows
and Cows and the real recipient have failed (or you simply don't want to
bother taking the time -- It's not your job to do so, but in this case,
the user has made the attempt)
To you, you haven't given consent, so it's spam (under a strict "Consent
is all that matters!" you haven't given consent -- I'd consider it a
candidate for blocking by sender locally)
But to the other Kennith McGrail (as well as every other staff member
using that payroll company), this mail is definitely wanted, and so I'm
not sure if it's a good idea for the entirety of the SpamAssassin
project's userbase to learn to block it or not, which is the bar I use
for my ruleqa corpus.
If the sender had a way to make it stop, I wouldn't be calling it spam
or suggesting it's anything but the result of an honest typo. But that
isn't the case here; here it has reached the point of being abusive. But
abusive does not necessarily mean it's spam.
(And again, I'm not really worried about this specific case, it's one
sender with a consistent sending address, so it's trivial to block --
But I'm looking for guidance for this type of situation when sorting in
general. For cases #3 and #4, based on your feedback they're already in
my spam corpus queued for tonight's run)
--
Dave Warren
http://www.hireahit.com/
http://ca.linkedin.com/in/davejwarren
