From: szabo@netcom.com (Nick Szabo)
Date: Mon, 25 Oct 93 04:24:49 PDT
To: jamie@netcom.com (Jamie Dinkelacker)
Subject: Re: A favor re Detweiler (ha ha ha)
In-Reply-To: <9310250655.AA25136@netcom.netcom.com>
Message-ID: <9310251120.AA21774@netcom5.netcom.com>
MIME-Version: 1.0
Content-Type: text/plain



Jamie Dinkelacker:
>[wonders why Detweiler hasn't phoned himself or Tim May]
> The only reason Detweiler has not executed this test is that he avoids truth.

Harsh, but insightful.  Detweiler seems to have a world-view 
in which people are and should be marching morons who make up their 
minds via peer pressure, perceived majority, consensus, etc. rather 
than by listening or reading for content, thinking, or taking action for
themselves.  A phenomenon which disrupts this cherished "democratic" 
state of affairs is "utterly perverted" and "evil".  S(he) who promotes 
such a change, either specifically or as a side effect of trying to 
return a modicum of privacy to the panoptic, permanently recorded net, 
is a "traitor" who should be "thrown in jail", no less.  Detweiler's 
emotional vision of making every poster "accountable" to mob rule is the 
most convincing argument I've heard yet for privacy on the net, the stronger 
the better, the sooner the better.  I wonder how long it will take 
Detweiler's rhetoric of "perversion" to spread and color other cypherpunks
activities.  If maintaining 'nyms indicates a multiple personality
disorder, then perhaps posting anonymously suggests an inferiority
complex, having something to hide is the sign of a sociopath, etc.
If it's a new and different cultural phenomenon with nuances
that take time to learn, it might well be easier to defame it
as a psychological abnormality, and brand its practioners as 
"criminals", than to learn about it, adapt to it, and/or convert it 
locally to better suit one's own tastes. 

BTW, "Medusa" probably won't last very long after doubling up
with her other 'nym to flame people, having her 'nyms praise
each other, etc.  This isn't theoretical stuff -- a wide variety
of these kinds of things have been tried on BBS's for years. 
There's a big track record out there we can look at.  Making a mistake 
that unmasks your 'nym is very easy: between leaving these pointers 
lying around, the extreme difficulty of maintaining a distinct style 
or personality for an extensive amount of quality posted material, and the 
"treachery" of aquaintences (who can use anon remailers if need be
to umask the 'nym).

Thus, I find Detweiler's speculation of massive political
manipulation by net "pseudo-spoofers" (isn't that redundant)?
to be paranoid and fantastic.  I wonder if Detweiler is as worried 
about the small "clique" of newswrire writers,  
quoted almost verbatim by most print media, and without much more 
than stylistic changes by TV and radio news, a few dozen people
communicating anonymously and via mouthpieces like Dan Rather
to hundreds of millions of people in the English-speaking world?   
This "manipulation" of perceived consensus and majority opinion is 
something a few dozen net "pseudospoofers" on smart 
drugs posting under a thousand 'nyms could never hope to match.

The closest the net comes to political spoofing is
probably voting for Usenet newsgroups -- completely
insecure, there may have been some close votes swayed by
spoofers, but the world doesn't revolve around the results.
Mostly folks don't care enough about newsgroup voting to bother 
to spoof it.

We have two options for electronic voting -- we can make it more
secure, and we can make it less important by changing majority-driven
insitutions with market-driven institutions.   I'm all in favor
of making e-voting more secure, but I don't think it's worth
sacrificing our privacy or civil rights to do so.  I don't
find majority rule to be the only, "SACRED"
way to get important things done in cyberspace.  Right now voting
plays practically no role in cyberspace, and that's also what
the forseeable future looks like.

I've seen 'nym unmasking happen extensively both on BBS's and Usenet.
For example, there was a "Holocaust Revisionist" spoofer
who posted under a series of nyms to Usenenet, the old silly 
argument about only 1 million Jews were killed not 6 million,
the Nazis really intended to facilitate emigration of Jews to
Palestine and instead the Allied bombing trapped them en route
in the camps, ad nauseum.  Professing such opinions can make 
you "accountable" really quick, in many lasting ways.  In this case, 
the spoofer was booted off of techbook.com, the BanishedCPU BBS 
newsfeed was cut off when he signed up there, etc.  

None of this spoofer's 'nyms lasted long; the style was too similar
(consistent differences via purposeful mispelling, different usage 
of capital letters, etc., overdone regional jargon, etc. are easily 
recognized for what they are).  Perhaps it's selection effect, but I've 
never seen a 'nym actively maintained for more than a year without 
unmasking or changing over to another name to avoid that 'nym.  In
anon posts of significant length I've recognized many style 
pointers right off the bat.  Eg, many of us independtly made the 
L.Detweiler/S.Boxx link, I've readily recognized the origin of 
several of Tim May's anon posts, etc.  If we can recognize
links from just one or two posts, how much more difficult to
maintain the ongoing politically persuasive rhetoric suggested by 
Detweiler.  Not saying it's not possible, just that it would take an 
extreme amount of careful effort.  This isn't evidence of "criminality"
or "perversion", it's evidence that the "spoofer" cares enough
to go to those lengths, much as Detweiler cares enough to go
to great lengths to fill up my mailbox with name-calling and
elaborate scenarios.

Anybody know what the world record is for 'nym maintence?

Really important votes, market interactions, etc. should be
made quite a bit more secure than Usenet voting, but at least in 
the case of markets, does not require physical identity to be tied 
to 'nym.  (Cf. for example Chaum's work and per-organziation pseudonyms
and credentials that can be transferred between these 'nyms, 
the ongoing discussion of reputation-building, etc.)

Nick Szabo				szabo@netcom.com