ietf-asrg
[Top] [All Lists]

Re: [Asrg] 0. General - Define Spam

2004-01-05 09:30:28
"Jon Kyme" <jrk(_at_)merseymail(_dot_)com> wrote:
This might seem like a "good thing" - but of course this means that there
is no way of comparing proposals meaningfully. e.g. Scheme (a) will
probably be (very close to) 100% effective at stopping the set of email
messages that the creator states it is targeted at. Unless the creator is a
complete idiot. Scheme (b) is likewise 100% effective for the (overlapping
but not identical) set it is designed for. How does this help anyone?

  Because the two different schemes have *clearly* defined what part
of the spam problem they are trying to solve.  They can therefore be
compared on the level of "which part of the problem they solve", and
not just on "how they solve the problem".

  If we had one "perfect" definition of spam, then we would have two
ways of comparing anti-spam schemes:

  1) solely on their technical merits

     This is difficult, as the methods will be doing wildly different
     things.

  2) On the part of the spam problem they address, AND on #1

     For this, we will still need a separate definition of the part of
     the problem the scheme addresses, independent of any "perfect"
     definition of spam.

*You* have rechartered the group to work on the poorly defined "spam"
problem. Previously, the group worked on issues raised by communication
characterised by (lack of) consent.

  I must admit that my preference would be to have the charter discuss
consent.  We can then come up with a "perfect" definition of spam
which fits within the consent model:

  SPAM: For a particular system, any message locally determined to be spam.

  Nice, circular, and nearly useless.  But it does focus the solution
where it should belong: *I* determine what is spam on my system, and
any system which is set up to permit *other* people to make that
decision on my behalf should be explicitly outside of the group.

Furthermore, the group is now chartered to focus on "approaches that can be
defined, deployed and used in the near term".

  Given the blatant opposition to getting anything done, I'm inclined
to believe that the SMTP-based email system will collapse before
people are willing to do anything.

  I've seen a ~10x increase in spam to my systems in the past two
months.  It's *ridiculous* that there's so much opposition to doing
anything.  Of *course* any anti-spam system will be short-term, and
won't solve the problem.  But it should give us the breathing room to
work on a more complete solution.

  i.e. I'd rather not have my doctor say "You've got terminal cancer,
but we're not going to treat it.  It's easier for us to wait until
you're dead, before we operate.  We think that by then, we'll have a
better treatment available."

  That's blatant incompetence and negligence.  But on the anti-spam
front, it's accepted practice.

  Alan DeKok.

_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg