Experiments on Mailing Lists

["Timothy C. May" <tcmay@got.net>]

At 7:49 AM -0800 1/4/97, Peter Hendrickson wrote:

>The moderator keeps the cash.  The moderator is pulled in several
>directions.  He or she wants to keep the money but also wants people
...blah blah blah....

All of these schemes--some of them pretty clever--for posting tokens,
reputation-based killfiles, buying and selling reputation futures, etc.,
are almost certainly far too complicated to deploy on a list like ours.

I have some direct evidence. The "Extropians" mailing list, beset with
similar perceived S/N problems several years ago, tried some of these
approaches. I've mentioned this before, and strongly urge those
contemplating such schemes to find some archives of this time period (I
don't know if they exist) and see what some of these real world experiments
did.

(Granted, there was no PGP encryption being used, but this is not really
central at this point to most of the proposed schemes.)

Here are some of the main Extropians list experiments I recall from the
1992-4 period, and some comments:

1. List-server-maintained filters. Subscribers could request that mail from
"Joe Doe" not be sent to them. Subscribers could request only certain
topics, or only certain subscribers, etc. Specific messages could be
requested (e.g., of someone being filtered).

2. A reputation market, with share prices for the repuations of specific
individuals, specific claims about the future, or even general topics. For
example, the shares of "crypto anarchy" might have started trading at 15
thornes, and later reached 110 thornes. (Thorne was the guy who eventually
got the code running, mostly.)

3. Private justice was tried, using the notion of "polycentric law." Those
with beefs with others could "file suit" with some entity. "Tim's
Protection League" could serve up certain kinds of justice, including
forcing his own clients to not post, for example. (Readers will note the
influence of Neal Stephenson's "Snow Crash" on some of the ideas, along
with the ideas of Bruce Benson, David Friedman, and others.)

What were the results?

First, much list bandwidth was consumed discussing the design of these
systems, the limitations of simple scalar measurements, and what real
ratings systems and reputation-based killfiles ought to have.

(Sasha Chislenko, now affiliated with Firefly, the distant grandchild of
such schemes, was involved.)

Second, there was an actual increase, in my opinion, in off-topic posts, as
the weaknesses of these systems were probed, and workarounds found.

Third, the aforementioned weaknesses were just too obvious. Here are some
examples:

* The reputation market was incredibly easy to manipulate. I used my
initially distributed "thornes" to drive up the share prices in my own
"tcmay" share prices. I also spent $15 of "real world" money to buy the
thornes of a guy who was not interested in using them...this $15 translated
into a _lot_ of thornes, which I could then use to buy lots of shares, and
see the share prices increase.

(I left the Extropians list in January of 1994, and at the time had the
largest portfolio...no big deal, but it shows how such markets are not
exactly very grounded in underlying reality!)

* The killfiles at the host consumed a fair amount of CPU cycles...partly
to pay for the increased services, a subscription fee for the Extropians
list was initiated....this apparently drove the membership in the list down
to much lower levels. (Which some may think is a good thing, but I
understand from friends who remained on the list after 1/94 that posting
rates dropped to such a low level that the list became much less
interesting to read...I heard a few days ago that volume is back up to
about 40 messages a day, and that the subscription fee has been dropped.)

* The private adjudication of disputes was the most absurd of all. There
was little or no incentive for "rational" judgements, and anyone could form
their own "protection" justice system. I created my own, with myelf as the
only member. This was not forbidden by the list meta-rules on such things,
so I became de facto my own justice system, not bound by any other systems.

Now, don't get me wrong....these "experiments" were interesting in their
own right. And eventually such experiments may actually work out. Certainly
many of us believe in the importance of individuals and groups setting up
similar institutions on their own intiative. The problems, however, are
apparent.

(It would be nice to see a much longer article on the results of these
experiments. It could even get published, in my opinion, as the issues are
interesting. Emergent organizational principles, polycentric law, etc.
Perhaps someone still on the Extropians list could suggest that this would
make a nice article.)

A lack of richness of interactions is part of the problem. The lack of
accountability is another. The loopholes in the meta-rules compounds other
problems. And a _lot_ of list bandwidth got consumed. The flame wars also
got even more vitriolic.

So, rather than do similar tinkering with the Cypherpunks list--not that
either Hugh Daniel or John Gilmore have given any hint they are willing to
do such tinkering--I suggest those who want to try token-based posting, or
information markets in reputation capital, or herd-consensus killfiles,
etc., set up a separate mailing list and implement whatever they wish.

While such schemes may turn out to be imortant, I'm here on this list to
discuss various ideas, not to be guinea pigs in somebody's pet idea (not to
mix my pet metaphors, or my pet peeves).

Cheers,

--Tim May


Just say "No" to "Big Brother Inside"
We got computers, we're tapping phone lines, I know that that ain't allowed.
---------:---------:---------:---------:---------:---------:---------:----
Timothy C. May              | Crypto Anarchy: encryption, digital money,
tcmay@got.net  408-728-0152 | anonymous networks, digital pseudonyms, zero
W.A.S.T.E.: Corralitos, CA  | knowledge, reputations, information markets,
Higher Power: 2^1398269     | black markets, collapse of governments.
"National borders aren't even speed bumps on the information superhighway."