April 4, 1994


Barbed Wire Fences in Cyberspace: The Threat Posed by Calls for
Ownership of Transactional Information

By David R. Johnson

People leave tracks as they move through electronic networks. Whenever
we do business, electronically or otherwise, the other party knows
something about our preferences and activities. But when we use the new
electronic vending machines and digital cash, the merchants with whom we
deal can keep a detailed record that reveals much more, in aggregate,
about our activities.  Store and forward messaging systems can continue
to store after they forward. As we all conduct more and more of our
lives on the networks, we will necessarily create a much more detailed
digital record of those lives.

Concerned about the threat to privacy created by such electronic
dossiers, some have called for new laws granting each of us "ownership"
of all the transactional information generated as we move around the
network. Third parties would have to bargain with us to obtain the right
to use "personal" information. This may produce a sort of cattle drive
vs. sheep herder battle on the electronic frontier. Either "information
wants to be free" or we can all put barbed wire around the tracks we
leave -- but we can't have both a free information range and a system of
information ownership. The First Amendment implications of any such
privacy regime are staggering.

This real contract is, in a sense, internal to every user of the
networks.  You want your own information to be private, at least to some
degree. But you want to be able freely to find out about everything and
everyone else.  You want to get customized marketing services and
eliminate junk mail.  But it's annoying to think that the marketing
companies, not you, profit from knowing what you want.  You want to be
free to investigate and speak your mind about any subject.  But it can
be devastating when someone else feels free to broadcast a critical
comment about you across the world. April 4, 1994 Barbed Wire Fences in
Cyberspace

Ironically, despite recent calls to create a proprietary interest in
personal information, it may not matter who "owns" this data. 
Economists tell us that the market -- especially a super-efficient
low-transaction-cost online market -- will operate quickly to create
trades that allocate information rights in an optimal way no matter what
the starting point. If consumers "own" their personal data, critics of
the privacy initiatives suggest, they will trade away this transactional
privacy for convenience every time. Even if we all get to sell our
privacy in this way, the price will be low and the economic benefits
will flow primarily to the rich (whose preferences are worth more to
marketers). So we are faced with the prospect of a range war over
ownership of data that may, in the end, settle nothing of importance and
cost us all something to create the markets in which we negotiate
transactions in these rights.

Let's go back to privacy basics.  What we all fear most is the ability
of the government or a personal enemy to put together a detailed
dossier, compiled from easy-to-obtain electronic sources, that becomes
the basis for a damaging publication or prosecution.  And we want to
know the ground rules under which we transact any given interchange --
are we in a quiet corner, whispering, or in a public square trading in
the open?  The way to avoid both confusion and undue invasion of
property, in light of these basics, is not to create a new property
interest. It is to allow (and require) system operators to establish and
enforce clear rules -- and then to allow system users to decide where to
conduct their business taking into account the degree of privacy they
want with respect to particular types of transactions. The operationally
efficient way to provide optimal privacy and openness is to let users
decide which sysop's rules are right for them -- and to select the
systems through which they want to conduct their business.

Even if we select a legal regime based on choice among communication
systems -- as opposed to a default "ownership" principle -- there will
be a need for new law regarding detailed electronic transactional data. 
We will need to be sure that users can find out about the rules of any
given system with which they interact and enforce the rules against both
sysop and other parties they meet online.  We will need direct
prohibitions or warrant conditions applicable to potential abuses of
governmental powers. And we will need to consider whether to ban
outright the publication (or threat of publication) of certain types of
detailed dossiers for tortious reasons -- such as extortion or coercion.

As the electronic frontier matures, there will be continual conflicts
between those who want free access to information and those who want to
participate in the networks without automatically giving up all rights
to control the flow of the resulting transactional data.  Before we
treat these conflicts as appropriate matters for a new type of
"property" law, however, we should remember that almost everyone will be
on both sides of any new "fence."  If we approach this problem by
allowing a variety of new spaces to develop -- allowing users to reward
and punish optimal sets of rules by sending their business to the stores
of sysops who get the rules right, we'll have saved a lot of unnecessary
transaction costs and we'll all end up owning a more civilized patch of
cyberspace that is appropriately (by our own lights) accommodating to
both privacy and freedom of speech.