Change Should Be Welcomed
Well, I am more than glad that Larry
thinks that I have joined issues on the matter, and have been able to penetrate
this world notwithstanding my ivory-tower-libertarian credentials. Indeed, I
think that his answer is so useful precisely because it does not rely on any
private norms of the cyberspace community in dealing with the broader issues.
Substantively, the question he poses at the outset of his letter is one that we
should all answer: Why is this space different from any other space? Why and
how will markets reduce freedom there?
In answering that question, I will start again from the other side, to make
the case that the complications and issues that he sees in cyberspace are
reflections of similar issues that have developed in other markets, where the
issue is the same: How do we overlay private businesses of one sort or another
on some common grid or infrastructure? It is clear that architecture matters in
both kinds of space and in pretty much the same kind of way. As a first
approximation, we need to have public highways because it is simply too costly
in ordinary space to build roads that stand one beside the other each limited
to a certain group. But the converse hardly follows from that observation:
There is no need for all roads to be public. A large ranch can have lots of
private highways for internal communication. Common gated communities have
roads that exclude the public by allowing limited entry by all members of this
commons. And if someone could acquire the land to build a private road beside a
public one, we should not begrudge him because he removes some traffic from the
common infrastructure.
Now what happens if the cost of private roads really drops, so that everyone
can afford the right level of individualization? The answer is that we get more
private roads. This is what has happened in the telephone industry: The party
line (by which several families shared a single telephone line) went the way of
the dodo when the cost of technology allowed for cheaper phone lines to be
developed. And so it is on the Net. The reduction in cost could easily allow
the creation of private lines side by side with that open space; and if we have
these two regimes, public and private, operating side by side, we can recognize
it not only as a profound change in the Net, but a welcome change that is
predictable as a matter of general economic theory.
Larry is less sanguine about this because he thinks that the architecture of
these private structures does more than make commerce more efficient. It also
makes control cheaper. But here I can accept the descriptive conclusion but
doubt whether he has pointed to a matter of concern. It is as not as though
when AT&T starts its broadband system that it can compel me to join it.
There still remains the older public road for those who do not want to go along
the private road. The presence of AT&T allows for more sorting in the
marketplace, which means that those people who care more about freedom can move
in one direction, with the legions of hackers and researchers; while those who
care more about security can migrate to a network that gives them protection
against anything from advertisement to smut. It is hard to see any systemwide
threat to liberty when new technology increases the ability of all individuals
to separate themselves from their fellow man and to filter out the information
that they do not want. It is not as though AT&T is replacing the open Net.
It is giving an alternative avenue to it.
Larry then talks about the issue of anonymity, and there again we have to
look at both sides of the question. Some people prize anonymous speech: The
Federalist Papers were written in that voice. But no one prizes anonymous
threats or thefts; so the question is whether individuals will trade in liberty
for security in the same fashion that Locke asked in dealing with the social
contract. Nor is there anything that says that the choice of anonymity or
identification has to be made once and for all. Credit-card companies
frequently collate data in ways that allow for targeted mailings by
advertisers. Most people welcome this because they know that the address can be
used by an advertiser without being given to him: The company prepares a list
for third-party use and guarantees that the names fit certain protocols. Anyone
who wants out can click a button to escape. Here, a good technology allows
choices to be heavily individuated. A technological and design issue that
expands freedom in sensible ways. It is not that the Net is "architected" as
though it were an all or nothing issue. It is the extent that any portal to the
Net architects itself and gives notice of its deep structure to the rest of the
world.
On a similar front, several high-school students at the University of
Chicago Laboratory Schools designed a program that allows individuals to cover
their tracks when they surf the Net. Here again the private response to the
systemwide structure helps improve matters. Cookies are subject to the same
analysis. Let Netscape track me so long as they disclose what they are doing.
But I think that they would lose market share if they did not allow users to
disable the system with a click of the mouse, which I take it they do. So again
individuation works. Taken as a whole, then, I see no reason for gloom that we
have changed the Net as its user population changes.
Larry promises us that more is in store: The older Net was "end-to-end," so
that all the action was in the user applications. That is fine in principle.
But what if the user wants to put his own private network at the end? I see
nothing wrong with this approach, and it would allow Harvard to have, as Larry
notes, a heavily regulated internal computer culture, while the University of
Chicago is more open--a good reason for Larry to return to his intellectual
home at Chicago. But does the new network add-on violate the old principle? Not
as far as I can see. It is the gated community example all over. The same open
Net carries the same traffic as before. It is not as though the size of the
commons is diminished because of the subdivision on its side.
But what about AT&T? Let it build its own system so long as it does not
shut down any other. But now they want to make sure that they influence the
traffic over the Net. So they will get less for the service they supply and
others will remain part time on the public highway. Or, if they have acquired
too many direct competitors, then we have a situation where there are possible
earthbound antitrust issues, a point on which I would be skeptical given the
ease of entry through other portals on the Net. Its power to discriminate could
add value to consumers; and if it does not, it will become the Gandor Airport
of the 21 st century.
Maybe now we have an explanation for the larger puzzle. Why are
cyber-libertarians slow to recognize the dangers to the Net from the new set of
uses? My explanation is that of optimism itself. I don't think there are
dangers that are anything close to the enormous possibilities that are
introduced by the enormous expansion of its use. That one Net can support
millions of Web sites dedicated to all sorts of different ends is a tribute to
the original model of how private diversity can flourish on a public grid. What
is truly optimistic about the Net is the sense of ratios. The amount of space
that has to be kept public for all to gain access is relatively small compared
with the amount of land that has to be used to maintain a coherent highway
system. So I quite agree that we should pay attention to these changes. But I
disagree that we should view them with suspicion. The blurb on Larry's book
speaks of it as a "dark, exhilarating work." I see the cause of exhilaration
but don't have Larry's dark forebodings. So long as we keep our intellectual
feet firmly on the ground, we shall not lose our way in cyberspace. Over to you
Larry, to explain why this happy estimation of the future is misguided, and a
gloomier response is warranted.
