Re: [Asrg] criteria for spam V2

> From: Dave Crocker <dhc@dcrocker.net>

> ...
> VS> Trying to define spam as that which we our comptuers can detect is a
> VS> serious mistake.  As you and others have often said, we cannot hope
> VS> to eliminate all spam.  An equivalent statement is that we cannot hope
> VS> for our computers to detect all spam.
>
> Frankly, I am hoping that we simply get away from the term.  No, I do
> not think we can agree on a universal, all-encompassing definition for
> spam and, no, I do not think we can prevent or detect all spam.
>
> That's why I am in the group that believes we should settle on a
> definition of something that we CAN operationalizem, for which improved
> control will mean improved Internet usability.

It would also be a major mistake to define "unsolicited bulk email"
as that which our computers can detect.

As in any technical design process, the first step is to figure out
what we're doing.  Nit picking about whether the word "bulk" in the
requirements document should be 2 or 3 is destructive.  We must address
the fuzziness of "bulk" only when we worry about reducing the requirements
to something for the computers.  Judging from other contributions
about "largely automated mechanims" and so forth, we are not ready
for that.  Besides, your bit about the student does not seem to be
about resolving the fuzziness of "bulk" but finding contractions
between the fuzzy human notion and what computers will do and so
throwing out the notion entirely.

Perhaps I've misunderstood.  Are you instead saying that 2 is the wrong
value of "bulk" for computers?  If so, what number do you prefer?

For what it's worth, most of the thousands of operators of DCC clients
seem to prefer either values between 20 and 100 or the special infinite
or overflow count of targets that means "incredibly bulky because the
message hit a spam trap."  (One of the critical parameters for a DCC
client is the local threshold for "bulk.")

Vernon Schryver    vjs@rhyolite.com
_______________________________________________
Asrg mailing list
Asrg@ietf.org
https://www1.ietf.org/mailman/listinfo/asrg