Re: [anti-spam-wg@localhost] Solution to Spam

[Nate Waddoups <]

On Thu, 26 Jun 2003, Mark McCarron wrote:

> I see you too ignored the part about mailing lists NOT being blocked by the 
> system.  Another professional who reads what he want to and ignores the 
> rest.

Maybe he read this part:

"2.  A 10-15 second delay between emails should be imposed."

...and did the math.  A 10 second delay is 6 messages per minute,
360/hour, and 8640/day.  A mid-sized mailing list with 200 people and 50
messages/day will generate 10,000 outbound messages per day.  Readers will
fall further and further behind... until they stop using the system
entirely.

Or maybe he read this part:

"Sorry, but we really don't care about mailing lists." 

Can you see where people get the idea that your system blocks lists?

"'Mailing lists' as I said are unimportant however, this does not mean
that they will not (or cannot) be sent. The only difference the end user
will notice will be the graphic download and the delay, other than that,
it will not effect any end user."

The delay, if it could be enforced, would stifle mailing lists, as O
described above.  But can the delay even be enforced?

Suppose the mailing list server (or spammer) circumvents the delay by
distributing outbound mail to a large list of mail servers.  If each
server has a 10-second delay, then spammers can send mail at 1-second
intervals, in full compliance with the rules, by using round-robin
delivery to 10 mail servers.  Or .1 second intervals with a list of 100
mail servers.  How will you stop that from happening?

Or maybe he read this:

"As for 'mailing lists', well, the majority of users don't really care
about them. That is what it comes down to in the end, majority rules."
Special authorisation codes can be given to legitimate companies to enable
certain bulk email features [...]"

It appears to me that you are acknowledging that the system you've
proposed, if it works as originally described, DOES block mailing lists.

You saw the problem yourself and acknowledged that a new feature (special
authorization codes) would need to be implemented to work around it.
That's fine, but in that light I don't see how you can protest that people
who think your system blocks legitimate bulk email are misunderstanding
your proposal.

Care to describe those special codes?  What makes them different from
"regular codes," and how do you stop spammers from forging them?

-- 

Nate Waddoups
Redmond WA USA
http://www.natew.com