Unnamed Administration sources reported that Lionel said:
[opt-in bulk email]
>Has anyone ever actually come across such a contract in real life
>or are they just urban myths?
Urban myth.
If you make damn sure that you clearly mark your bulk mail with the
website/organisation at which your user subscibed, & you record the
*way* they subscribed[0], you should be fine. It's also vitally
important that you respond promptly to email that arrives at your
domain's 'abuse@' address.
[0] Eg: IP address & time stamp from when they hit the 'subscribe me'
button on a web form, copy of the signed paper form they sent in, etc.
Likely insufficient.
Save your hide by getting verification on every entry; i.e:
1) Get request.
2) Send email to alleged requester.
3) Do nothing unless/until you get back a confirming "yes, I do want"
reply.
This is what spammers disparage as "double out-in"...
>Save your hide by getting verification on every entry; i.e:
>1) Get request.
>2) Send email to alleged requester.
>3) Do nothing unless/until you get back a confirming "yes, I do want"
> reply.
Yes, very good point. I should have included that too.
That's exactly what we are doing. Which is good
As it's still likely to end up with the most popular domains @hotmail.com, @yahoo.com, @aol.com having several thousand recipients
though I'm still interested in whether anyone has more experience
of ensuring that mail doesn't get blackholed.
I'm thinking along the lines of whether and how it's necessary to
rate limit sending to those domains, whether they don't like single
messages having more than a certain number of RCPT TO lines, whether
there are contracts that one can sign to get access to some sort of
super special non-public MX for them, etc...
and log and save everything. if there's a web form, then log the ip address
that the request came from. provide enough infrastructure that when you get
a complaint, you can rapidly provide the records.
and the "urban legend" thing is incorrect. AOL has in some cases had mailing
list providers sign agreements governing their behavior. that's the only
one i know of, but there could be others.
As it's still likely to end up with the most popular domains @hotmail.com, @yahoo.com, @aol.com having several thousand recipients
though I'm still interested in whether anyone has more experience
of ensuring that mail doesn't get blackholed.
Spam has reached such epic porportions that it is virtually
guranteed that if you send mail out on a regular basis, you
will eventually be blackholed somewhere. But if you follow
the advice here (as it sounds like you are), most sane folks
will still accept your mail.
I'm thinking along the lines of whether and how it's necessary to
rate limit sending to those domains, whether they don't like single
messages having more than a certain number of RCPT TO lines, whether
there are contracts that one can sign to get access to some sort of
super special non-public MX for them, etc...
or whether it's just all pot luck
It varies a lot, depending on the provider. However, it'd
probably help to remember that a load of mail which might
DoS a small provider will almost certainly set off alarms at
large providers...and that may get you blocked.
At my last job, we successfully flew under the radar by sending individual messages to each recipient. We were sending info to around four hundred thousand registered users of our site and some tens of thousands were at yahoo, hotmail, aol &c.
Our only problems were on our side ... we ran out of filehandles a couple times. If anyone wants to take a look at the quick and dirty perl script I wrote, you're welcome to it.
