Re: [Asrg] draft-irtf-asrg-criteria (was Re: request for review for a non FUSSP proposal)

[Alessandro Vesely <[e-mail]>]

Steve Atkins wrote:
> On Jun 30, 2009, at 12:04 AM, Danny Angus wrote:
>> Actually I would have liked to have included *some* definition, but 
>> because members of this group hold pretty entrenched opinions covering 
>> most possible definitions I felt that on the one hand it would be 
>> impossible and on the other hand it would be unnecessary.
>>
>> I believe that it is unnecessary for two reasons, the first being that 
>> this group cannot agree a definition, yet operates reasonably 
>> successfully, and secondly there can be an empirical test for a 
>> solution, even if there is no agreed definition of spam itself.

In that case, the criteria draft should not give a definition at all. 
Instead, it should mention that proposed techniques may give the 
definition of the phenomenon they are intended to operate against, if 
necessary. A warning against incautious use of the term "spam" would 
be in order, since it is an undefinable term. It may or may not be 
useful to enumerate some of the existing definitions.

> The problem arises when someone, anyone, claims that there is 
> One True Definition of spam. The fact that that's blatantly false 
> isn't the problem. That it causes hordes of people to come out of 
> the woodwork to argue for their One True Definition of spam, causing 
> yet another rerun of the Thread That Would Not Die is the problem.

Agreed. However, the anti-spam endeavor should not have the tones of a 
religious debate. It's not. One reason why I would have liked to 
classify objections against a tentative definition is to understand 
where does that holy war spirit originate from. Is it still strong as 
in the MARID epoch, or has that lesson been learned? I'm not sure 
whether those hordes, or the thoughts that trouble them, thwart more 
than just definitions.

(The other reason is that definitions are generally useful. They are 
not true or false, they define something. They may be good or bad, 
though. Good definitions provide constructive hints; for example, the 
U in UBE suggests that a technique might attempt to maintain a 
register of presupposed solicitations.)

> (A problem that's usually best solved by killfiling anyone participating
> in that sort of thread).

I believe you meant fundamentalists rather than mere participants. I 
hope I'm not stamping on anyone's feet if I rise these questions. I do 
so because I can't stand the perpetual failure to effectively counter 
mail abuse. IMHO, it must be related to some wicked cripples that 
undermine anti-spam work. New solutions arouse no interest, 
independently of their technical content. Some bafflingly conclude 
that spam is a natural fact of life, that cannot be even diminished, 
and any solution would only alter its delivery mechanisms. Preemptive 
rebuttal, I don't think it's sane. Marketing may be considered a 
natural fact of life, if regarded as the commercial facet of Darwinian 
evolution. However, if direct marketing is considered spam, the 
"spammers-are-stupid*" entries of Rhyolite's list, hinging on the 
assumption that spammers won't respect RFCs, may not apply.

It will be helpful for newcomers if Danny's paper will have an 
Introduction that explains the questions I've been tried to address 
above --lengthily and vaguely, as I don't know the answers myself. 
Bill mentioned a "dynamic equilibrium" in [1], depicting a chronic 
syndrome that can only be alleviated by a "really big push". Even if 
that's not a satisfactory answer, his words help understanding the 
problem.

--
[1 http://www.ietf.org/mail-archive/web/asrg/current/msg15369.html]