CDT's Berman Opposes Online Anonymity

AMERICAN CIVIL LIBERTIES UNION


                       WASHINGTON OFFICE

                                                    122 Maryland Avenue, NE
 November 1, 1985                                   Washington, DC 20002
                                                    --------------------
                                                    National Headquarters
Mr. David Chaum                                     132 West 43rd Street
Centre for Mathematics and Computer Science         New York. NY 10036
P.O. Box 4079                                       (212) 944-9800
19O9 AB Amsterdam                                   Norman Dorsen
                                                      President
Dear Mr. Chaum:                                     Ira Glasser
                                                      Executive Director
                                                    Eleanor Holmes Norton
                                                      CHAIR
                                                    National Advisory
                                                    Council





        Thank you for sending me a most interesting article.   A
society of  individuals  and  organizations  that  would  expend  the
time and resources to use a  series  of  'digital  pseudonyms' to
avoid data linkage does not in my opinion make big brother
obsolete but  acts  on  the  assumption  that  big  brother  is  ever
present.  I  view  your  system  as  a  form  of  societal  paranoia.

        As a matter of  principle,  we  are  working  to  enact  formal
legal protections for  individual  privacy  rather  than  relying  on
technical solutions.   We want to  assume  a  society  of  law  which
respects legal limits rather than  a  society  that  will  disobey  the
law, requiring  citizens  to  depend  on  technical  solutions.  e.g.
require  a  judical  warrant  for  government  interception  of  data
communications rather than  encrypt  all  messages  on  the  assumption
that regardless of the lawt  the  government  will  abuse  its  power
and invade privacy.

        As a  matter  of  practicality,  I  do  not  think  your  system
offers much hope for privacy.    First, the trend toward universal
identifiers  is  as  much a  movement  generated  by  government  or
industry's desire to keep track of all citizens as it is by
citizens seeking simplicity and convenience in all transactions.
At best,  your  system  would  benefit  the  sophisticated  and  most
would opt for  simplicity.  The  poor  and  the  undereducated  would
never use or benefit from it.

        Finally where there's a  will, there's  a  way. If  government
wants to link data bases, it  will,  by  law, require  the  disclosure
of various individual pseudonyms used by citizens or prohibit it
for data bases which the government wants to link.  Since
corporations  make  money  by  trading  commercial  lists  with one
another, they will never adopt  the  system  or  if  it  is  adopted,
will use "fine printn  contracts  to  permit  selling  various   codes
used by their customers to other firms.

        The solution remains law, policy, and consensus about limits
on government or corporate intrusion into areas of individual
autonomy.  Technique can be used to enforce that consensus or to
override it. It cannot be used as a substitute for such
consensus.



                                            Sincerely Yours,

                                             /Sig/

                                            Jerry J. Berman
                                            Chief Legislative Counsel
                                            & Direrector ACLU
                                            Privacy Technology Project


   cc: John Shattuck

At 9:48 AM -0700 7/21/97, Anonymous wrote:
>                     AMERICAN CIVIL LIBERTIES UNION
>
>
>                       WASHINGTON OFFICE
>
>                                                    122 Maryland Avenue, NE
> November 1, 1985                                   Washington, DC 20002
...

In fairness to Berman, in 1985 very few people were thinking seriously
about these issues, and Chaum's paper (presumably the one published that
year in "Communicatons of the ACM") was probably seen as far-off technology
then.

I'd be more interested to see Berman's more recent views on online anonymity.

It might well be that in 1985 he saw little hope for technological
solutions, and understandably placed more faith in legislative solutions.

Now that the technology for anonymity is widely deployed, this situation
has changed.

If Berman or any of the other "cyber rights" groups were to call for bans
on anonymity, this would be news indeed. (And I don't expect them to. The
Supreme Court rulings on anonymous leafletting and speech were pretty
clearcut.)

--Tim May

There's something wrong when I'm a felon under an increasing number of laws.
Only one response to the key grabbers is warranted: "Death to Tyrants!"
---------:---------:---------:---------:---------:---------:---------:----
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."

----- Forwarded message from Jerry Berman -----

As you probably noticed, this letter was written in 1985 and represented
the position of the ACLU at that time. Times, and organizations, and
technology have changed dramatically in the last 12 years, and so has my
view on this issue.

I strongly believe both legal *and* technical protections are essential to
guarantee privacy rights (including the right to anonymity) and have
worked, with my associates at CDT and others on the Net, to promote the use
of strong ecnryption as a way to protect privacy.  I also continue to work
for  legal restrictions on electronic surveillance.

Please post this wherever you think this might clarify my position and the
debate.

Thanks,

Jerry Berman


----- End of forwarded message from Jerry Berman -----

As far as I can tell, law, custom, and technology are the three ways we
have to protect our privacy.  Cypherpunks are well aware of the
technological options, so I won't discuss them further except to note that
they probably are not, by themselves, enough.

Law has also been discussed, including the European privacy laws and
similar laws which have been proposed for the US.  One good thing about the
European mindset on privacy is that it seems to include resistance to legal
suppression of privacy technology.

The last is custom.  This approach appears in cypherpunk discussions as an
emphasis on contractual relations between people and the organizations
receiving data.  It is this area I would like to discuss.


What makes a good privacy contract?  What should you expect when you buy
something?  What is the standard contract?  What exceptions must be clearly
noted?  How does society decide these cultural issues?

I strongly believe in standard contracts because they greatly reduce
transaction costs.  Where the parties need different terms, they can
negotiate exceptions.  However, we do not have standard privacy contracts
which match our new communication and database technologies.  We don't even
have a good public debate on the issues involved.

Unfortunately, it seems that the only way to raise these issues to the
radar scope of the news media/public is to have some congresscritter
propose a law.  Even more unfortunately, said law usually preempts people's
right to contract with each other, rather than just setting a framework for
the contract negotiations.  There must be a better way.


-------------------------------------------------------------------------
Bill Frantz       | The Internet was designed  | Periwinkle -- Consulting
(408)356-8506     | to protect the free world  | 16345 Englewood Ave.
frantz@netcom.com | from hostile governments.  | Los Gatos, CA 95032, USA

On Tue, 22 Jul 1997, Bill Frantz wrote:

> As far as I can tell, law, custom, and technology are the three ways we
> have to protect our privacy.  Cypherpunks are well aware of the
> technological options, so I won't discuss them further except to note that
> they probably are not, by themselves, enough.
... 
> The last is custom.  This approach appears in cypherpunk discussions as an
> emphasis on contractual relations between people and the organizations
> receiving data.  It is this area I would like to discuss.
...
> What makes a good privacy contract?  What should you expect when you buy
> something?  What is the standard contract?  What exceptions must be clearly
> noted?  How does society decide these cultural issues?

A contract is something within law and enforcable by courts, so you aren't
really talking "custom".  And even with private contracts, not everything
is enforcable.  Other times something like a contract (IANAL) is created
such as when I buy a product unless it says "as-is" it has some implicit
warranties under the universal commercial code.

Contracts are good things, but they are cumbersome because they have to be
negotiated on the spot [like the man at computer city who wrote on the
back of his check for their "club" membership - acceptance of this ... 
says you won't send me junkmail;  They deposited the check and he sued and
won].

On a web page, I usually cannot send an E-NDA in the form they want me to
fill out (so I often leave spaces blank or with obviously bad numbers and
some comment to contact me).  This is usually less efficient than phoning.

This leaves Custom as such, a social instead of a legal contract.  The net
has been good at policing its own.  Web pages asking for email addresses
aren't supposed to forward them to spambots.  They now often state this
explicitly.  If someone says something wrong in a discussion group, they
get flamed, and if they do it enough, they are placed in everyone's kill
file.

The problem is that culture takes a while to learn, and with everyone with
different customs comes to the net, they may not see anything wrong with
doing something they are used to (for a physical world example: some local
immigrants from an area where barter was the custom tried haggling on
price-tagged items until they figured out what was going on). 

Custom is determined by evolution, not by specification.  And that cannot
be accelerated, and will be a problem until everyone has been on the net
long enough to establish a common set of manners.  How much information
does someone retain?  Enough so that a web page presents your desired
configuration without having to retype it?  Should they pass this on to a
sister site?  A different company? 

The worst thing that can happen is to codify custom.  I can see that as
the reason to avoid "Voluntary Ratings", because as soon as a misrated
(who determines?) page comes up, it becomes cause for a legal action, and
then everyone will want it to be manditory and enforced.

So a measure of politeness (indicating what the content might be so it can
be filtered) becomes a political battle, almost as if I don't go out and
say something offensive in public every day then I am not defending my
first ammendment rights - which is only slightly less bizzarre than some
of the judicial decisions.

--- reply to tzeruch - at - ceddec - dot - com ---

At 12:22 AM -0700 7/23/97, Bill Frantz wrote:

>I'm not sure I agree that the evolution of custom can not be accelerated.
>If we actively discuss the proper limits to use of private information, are
>we not accelerating the development of consensus about what are reasonable
>policies.  It is precisely the lack of such discussion, and the lack of
>trust that accompanies it that leaves us in the situation we find ourselves
>in; between the well defined Netiquette of the ARPANET days, and whatever
>our commercial net will evolve.
>
>There are a lot of choices.  For example:
>
>(1) Don't remember anything about me as an individual.
>(2) Don't share any individual data.
>(3) Feel free to share, but leave my name off.
>(4) Sell the data, but give me a piece of the action.
>
>I can imagine that any of these will be acceptable to some people.

But these "agreements" have two sides.

None of the four choices you list are of any interest to me, for example,
and whether you find them acceptable is of no interest to me.

(I am not trying to be rude to Bill, just making the point forcefully that
I don't particularly care that these four choices are acceptable to "some
people.")

I of course remember _lots_ of things about people, I share those memories
on occasion (without requesting permission), I mention names, and I
certainly don't recall every giving one of the subjects of my memories a
cut of the action.

In a free society, it is not possible or acceptable to control what others
remember or gossip about. Or even sell commercially.

"Custom" only applies to those who adopt the custom--the "law" is for
everyone else. The question is: do we have a law demanding that people not
remember certain things, or not gossip about what they've observed? I think
even a totalitarian society will have well-known problems enforcing such
laws.

I'd've thought this obvious, but Bill's post makes me wonder.

--Tim May

There's something wrong when I'm a felon under an increasing number of laws.
Only one response to the key grabbers is warranted: "Death to Tyrants!"
---------:---------:---------:---------:---------:---------:---------:----
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."

At 8:52 AM -0700 7/22/97, Unprivileged user wrote:
>Custom is determined by evolution, not by specification.  And that cannot
>be accelerated, and will be a problem until everyone has been on the net
>long enough to establish a common set of manners.  How much information
>does someone retain?  Enough so that a web page presents your desired
>configuration without having to retype it?  Should they pass this on to a
>sister site?  A different company?

I'm not sure I agree that the evolution of custom can not be accelerated.
If we actively discuss the proper limits to use of private information, are
we not accelerating the development of consensus about what are reasonable
policies.  It is precisely the lack of such discussion, and the lack of
trust that accompanies it that leaves us in the situation we find ourselves
in; between the well defined Netiquette of the ARPANET days, and whatever
our commercial net will evolve.

There are a lot of choices.  For example:

(1) Don't remember anything about me as an individual.
(2) Don't share any individual data.
(3) Feel free to share, but leave my name off.
(4) Sell the data, but give me a piece of the action.

I can imagine that any of these will be acceptable to some people.


-------------------------------------------------------------------------
Bill Frantz       | The Internet was designed  | Periwinkle -- Consulting
(408)356-8506     | to protect the free world  | 16345 Englewood Ave.
frantz@netcom.com | from hostile governments.  | Los Gatos, CA 95032, USA

At 12:53 AM -0700 7/23/97, Tim May wrote:
>(I am not trying to be rude to Bill, just making the point forcefully that
>I don't particularly care that these four choices are acceptable to "some
>people.")

It is this kind of attitude I wish to encourage thru custom, certainly not law.

>
>I of course remember _lots_ of things about people, I share those memories
>on occasion (without requesting permission), I mention names, and I
>certainly don't recall every giving one of the subjects of my memories a
>cut of the action.
>
>In a free society, it is not possible or acceptable to control what others
>remember or gossip about. Or even sell commercially.
>
>"Custom" only applies to those who adopt the custom--the "law" is for
>everyone else. The question is: do we have a law demanding that people not
>remember certain things, or not gossip about what they've observed? I think
>even a totalitarian society will have well-known problems enforcing such
>laws.

When I was in Malasia, I saw signs in many stores which said in effect, In
accordance with Islamic law, we offer fair prices.  My experience with
prices in stores with these signs is that they were lower than in some of
the other stores.

A similar approach could evolve for network sites.  A sign saying that we
follow the US Chamber of Commerce's code on fair information practices
could easily evolve.  My questions are, what should that code contain?
And, how many different codes do we need?


-------------------------------------------------------------------------
Bill Frantz       | The Internet was designed  | Periwinkle -- Consulting
(408)356-8506     | to protect the free world  | 16345 Englewood Ave.
frantz@netcom.com | from hostile governments.  | Los Gatos, CA 95032, USA

Tim May wrote:
> In fairness to Berman, in 1985 very few people were thinking seriously
> about these issues, and Chaum's paper (presumably the one published that
> year in "Communicatons of the ACM") was probably seen as far-off
> technology then.
> 
> I'd be more interested to see Berman's more recent views on online
> anonymity.
> 
> It might well be that in 1985 he saw little hope for technological
> solutions, and understandably placed more faith in legislative solutions.
> 
> Now that the technology for anonymity is widely deployed, this situation
> has changed.

I think we've all been thru this before, with the Esther Dyson flap and
so on.  That was then, this is now.

Online anonymity was not much of a real issue ten years ago, and I doubt
most readers had formed much of an opinion on it.  On private BBSes it
was prevelant, and on academic and corporate networks it was virtually
nonexistant.  As these networks began to merge, not surprisingly people
came down on both sides of the issue, but their opinions were backed more
by convention than by the facts of the situation.

I think that as time has progessed, people have come to see the reasons
behind it.  Five years ago, there was considerable debate over whether
anon.penet.fi should exist, today almost everyone takes for granted the
right to post anonymously.  For all the animosity he has caused, Spamford
Wallace has shown us why you don't want people to know your email address.
Some of you may remember the comments made on the list by a certain law
professor, who said he wouldn't post to usenet because he didn't want to
get junk mail.  In hindsight we can laugh at that remark in light of the
obvious solution.

> Now that the technology for anonymity is widely deployed, this situation
> has changed.

Is it so, or rather is it that the technology has been deployed in response
to a changing situation?