ietf-asrg
[Top] [All Lists]

Re: [Asrg] ASRG progress? META-DISCUSSION(META-DISCUSSION(POSTAGE))

2008-12-02 19:43:05

On December 2, 2008 at 09:33 dhc(_at_)dcrocker(_dot_)net (Dave CROCKER) wrote:
Folks,

g'day.

As nearly as I can tell, all of this latest flurry of discussion on the ASRG 
has 
been exploring topics that have been explored repeatedly elsewhere, 
consuming 
(and wasting) much energy, with no useful results.

Perhaps I've missed the productive output.


Probably because after a few go-rounds for a few days interspersed
with a lot of "should we be talking about this?" someone jumps up
again barely one business day later and re-opens the "should we be
talking about this?" meta-discussion.

I had another, closely related, question which may've gotten buried in
the noise about the group's mission in general:

       How do we know when we're done?

I'll break out a sub-category:

       How do we know when we've made progress?

So I thought I'd ask:

      Since this is supposed to be a "research" group, what has been studied 
that is new?

Well, clearly a lot of people here had firmly held beliefs about
e-postage like "it pre-requisites micro-payments and a solution to
problems thereof" which have now been at least somewhat disabused.

That is, e-postage probably doesn't necessarily depend on
micro-payments, only certain probably degenerate e-postage designs
rely on micro-payments, so that long and firmly held "disproof" may be
specious.

That's some sort of progress. Not universal agreement, I don't think
we have that, but at least some acknowledgment that it might be the
real issue (pre-requisite, or not?)

The conversation has now moved on to whether the double-spending
problem has to be "solved" for e-postage to be workable, but that's
only in the past, um, ... I think it was first raised seriously here
(tho referenced previously, let's not argue that!) over the weekend.

It all might take more than a week or two, that's all it's been in
this round, particularly a week or two belabored incessantly with
meta-discussions of "should we even be talking about this" (not to
mention a long holiday weekend) every other day, to work through this
big, big issue.

If it were wholly up to me I'd find a way to ban wide-open
meta-discussions from this group (and many others.)

So my next question is: What have we learned from meta-discussions?

It's not that they have no use, but having one every 48 hours or so on
the same topic isn't quite right either.

They tend to be more a measure of the personal interest of the
individual opening the meta-discussion in the topic than an actual
substantive procedural discussion.

At any rate, there's also a huge semantic problem in that the usual
tone of a meta-discussion is to toss what might really be a much
richer set of issues (e.g., are economic incentives useful? how?)
under a catch phrase (oh we've discussed E-POSTAGE before it doesn't
work) in an attempt to stop the whole thing.

  Semi-Sarcasm: Oh, we've discussed stopping spam before, obviously we
  can't solve it or even slow its growth, can't we just move on to
  something more productive (whatever that is)?

      What new avenues have been suggested that have (or look likely to) 
develop 
group consensus as promising?


That mathematically closed algorithms for spam detection, prevention,
avoidance, and removal may not be a necessary pre-condition for an
approach to be viable? So what's the second choice?

That is, how leaky can an approach be and still be worthwhile? How
airtight must it be? How do we even measure any of that?

      What has been learned?

That the never-ending and uncontrolled background hum of
meta-discussion about whether the current topic will produce anything
can't possibly be productive?

Can't we at least have a designated meta-discusser with some rules
(like they can only raise the issue every so often, not every few
hours) rather than a near daily "why are we talking about this"
discussion raised by whomever gets the whim or has lost interest in a
thread?

Didn't we just go through this very meta-meta-discussion and the
moderator suggested that if it was clearly marked POSTAGE in the
subject so those not interested could easily move past it?

I think we reached that consensus on Thursday evening, barely one
business day ago. Yet here we are again:


  Date: 27 Nov 2008 17:26:43 -0000
  From: John Levine <johnl(_at_)taugh(_dot_)com>
  Subject: Re: [Asrg] e-postage -- NEW POLICY

  >What I sincerely do not understand is why it is so important for some
  >people to not just fast forward through the messages which they
  >personally believe will not be fruitful but instead to stop and try
  >their level best to discourage the conversation entirely.

  Well, OK.

  New rule: all future messages about e-postage, attention bonds, and
  other pay-to-spam schemes MUST have the word "postage" in the subject
  line.  Then we can put it in our killfiles.

  I mean it, by the way.



---

So another question raised here is: Who is in charge?


I'd now suggest that all meta-discussion be marked similarly so we can
all "killfile" it but somehow, since it is an attempt to influence
discussion in other threads, that won't be satisfactory.



         Meta-discussion is the devil's playground - me


As to something more concrete I'm contemplating trying a taxonomy of
e-postage ideas so the good and the bad and the ugly can be weeded out
more quickly and concisely: ``Oh, micropayments, that's 4.2.1.a, we
don't do 4.2.* anymore, and it's not a necessary pre-condition for
e-postage which means disproof of micro-payments is not a disproof of
e-postage, please read the taxonomy and, if you must, start a separate
4.2.1.a thread, but please don't present it as a deal-breaker because
it's not, it's only a deal-breaker for a very small set of possible
designs of e-postage.''

-- 
        -Barry Shein

The World              | bzs(_at_)TheWorld(_dot_)com           | 
http://www.TheWorld.com
Purveyors to the Trade | Voice: 800-THE-WRLD        | Login: Nationwide
Software Tool & Die    | Public Access Internet     | SINCE 1989     *oo*
_______________________________________________
Asrg mailing list
Asrg(_at_)irtf(_dot_)org
https://www.irtf.org/mailman/listinfo/asrg

<Prev in Thread] Current Thread [Next in Thread>