Книга: Code 2.0
The Many Laws Rule (and the technology to make it possible)
The Many Laws Rule (and the technology to make it possible)
So what would a more balanced regime look like?
Return to the conflict that began this chapter. On the one hand, France doesn’t want its citizens buying Nazi paraphernalia, the United States doesn’t want its citizens watching “free” TV. On the other hand, France doesn’t have anything against “free” TV, and the United States doesn’t have the constitutional power to block its citizens from buying Nazi paraphernalia. It’s some way to give France what it wants (and doesn’t want), and to give the U.S. what it wants (and can’t want)?
This is not an issue limited to France and the United States. As Victor Mayer-Schonberger and Teree Foster have written, about speech regulation:
National restrictions of freedom of speech on the Internet are commonplace not only in the United States, but also around the globe. Individual nations, each intent upon preserving what they perceive to be within the perimeters of their national interests, seek to regulate certain forms of speech because of content that is considered reprehensible or offensive to national well-being or civic virtue.
Is there a general solution (in the government’s eyes at least) to this problem?
Well, imagine first that something like the Identity Layer that I described in Chapter 4 finds its footing. And imagine that the ID layer means that individuals are able to certify (easily and without necessarily revealing anything else) their citizenship. Thus, as you pass across the Web, attached to your presence is a cryptographic object that reveals at least which government claims you.
Second, imagine an international convention to populate a table with any rules that a government wants to apply to its own citizens while those citizens are elsewhere in the world. So the French, for example, would want Nazi material blocked; the Americans would want porn blocked to anyone under 18, etc. The table would then be public and available to any server on the network.
Finally, imagine governments start requiring servers within their jurisdiction to respect the rules expressed in the table. Thus, if you’re offering Nazi material, and a French citizen enters your site, you should block her, but if she is a U.S. citizen, you can serve her. Each state would thus be restricting the citizens of other states as those states wanted. But citizens from its nation would enjoy the freedoms that nation guarantees. This world would thus graft local rules onto life in cyberspace.
Consider a particular example to make the dynamic clearer: Internet gambling. Minnesota has a strong state policy against gambling. Its legislature has banned its citizens from gambling, and its attorney general has vigorously enforced this legislative judgment — both by shutting down gambling sites in the state and by threatening legal action against sites outside of the state if they let citizens from Minnesota gamble.
This threat, some will argue, can have no effect on gambling on the Internet, nor on the gambling behavior of Minnesota citizens. The proof is the story of Boral: Imagine a gambling server located in Minnesota. When Minnesota makes gambling illegal, that server can move outside of Minnesota. From the standpoint of citizens in Minnesota, the change has (almost) no effect. It is just as easy to access a server located in Minneapolis as one located in Chicago. So the gambling site can easily move and keep all its Minnesota customers.
Suppose that Minnesota then threatens to prosecute the owner of the Chicago server. It is relatively easy for the attorney general to persuade the courts of Illinois to prosecute the illegal server in Chicago (assuming it could be shown that the behavior of the server was in fact illegal). So the server simply moves from Chicago to Cayman, making it one step more difficult for Minnesota to prosecute but still no more difficult for citizens of Minnesota to get access. No matter what Minnesota does, it seems the Net helps its citizens beat the government. The Net, oblivious to geography, makes it practically impossible for geographically limited governments to enforce their rules.
However, imagine the ID layer that I described above, in which everyone can automatically (and easily) certify their citizenship. As you pass onto a site, the site checks your ID. Thus the gambling site could begin to condition access upon whether you hold the proper ID for that site — if you are from Minnesota and this is a gambling site the site does not let you pass. This process occurs invisibly, or machine to machine. All the user knows is that she has gotten in, or if she has not, then why.
In this story, then, the interests of Minnesota are respected. Its citizens are not allowed to gamble. But Minnesota’s desires do not determine the gambling practices of people from outside the state: Only citizens of Minnesota are disabled by this regulation.
This is regulation at the level of one state, for one problem. But why would other states cooperate with Minnesota? Why would any other jurisdiction want to carry out Minnesota’s regulation?
The answer is that they wouldn’t if this were the only regulation at stake. But it isn’t. Minnesota wants to protect its citizens from gambling, but New York may want to protect its citizens against the misuse of private data. The European Union may share New York’s objective; Utah may share Minnesota’s.
Each state, in other words, has its own stake in controlling certain behaviors, and these behaviors differ. But the key is this: The same architecture that enables Minnesota to achieve its regulatory end can also help other states achieve their regulatory ends. And this can initiate a kind of quid pro quo between jurisdictions.
The pact would look like this: Each state would promise to enforce on servers within its jurisdiction the regulations of other states for citizens from those other states, in exchange for having its own regulations enforced in other jurisdictions. New York would require that servers within New York keep Minnesotans away from New York gambling servers, in exchange for Minnesota keeping New York citizens away from privacy-exploiting servers. Utah would keep EU citizens away from privacy-exploiting servers, in exchange for Europe keeping Utah citizens away from European gambling sites.
This structure, in effect, is precisely the structure that is already in place for regulating interstate gambling. According to federal law, interstate Internet gambling is not permitted unless the user is calling from a gambling-permissive state into another gambling-permissive state. If the user calls from a gambling-restrictive state or into a gambling-restrictive state, he or she has committed a federal offense.
The same structure could be used to support local regulation of Internet behavior. With a simple way to verify citizenship, a simple way to verify that servers are discriminating on the basis of citizenship, and a federal commitment to support such local discrimination, we could imagine an architecture that enables local regulation of Internet behavior.
And if all this could occur within the United States, it could occur between nations generally. There is the same interest internationally in enforcing local laws as there is nationally — maybe even more. And thus in this way, an ID-rich Internet would facilitate international zoning and enable this structure of international control.
Such a regime would return geographical zoning to the Net. It would re-impose borders on a network built without those borders. If would give the regulators in Hungary and Thailand the power to do what they can’t do just now — control their citizens as they want. It would leave citizens of the United States or Sweden as free as their government has determined they should be.
To those who love the liberty of the original Net, this regime is a nightmare. It removes the freedom the original architecture of the Internet created. It restores the power to control to a space designed to avoid control.
I too love the liberty of the original Net. But as I have become skeptical of short-cuts to the policy I like — short-cuts, meaning devices that produce a particular result without effective democratic support — I’m hesitant to condemn this regime. Of course, no democratic government should permit the will of a nondemocratic government to be reflected in a zoning table. We shouldn’t help totalitarian regimes repress their citizens. But within a family of democracies, such a regime might help promote democracy. If a restriction on liberty is resented by a people, let the people mobilize to remove it.
Of course, my view is that citizens of any democracy should have the freedom to choose what speech they consume. But I would prefer they earn that freedom by demanding it through democratic means than that a technological trick give it to them for free.
But whether or not you, or I, like this regime, my argument at this point is predictive. This regime is a natural compromise between two results, neither of which governments accept — governments will neither accept a world where real space laws don’t affect cyberspace, nor a world where the rule of one government, or of a few large governments, controls the world. This regime gives each government the power to regulate its citizens; no government should have the right to do anything more.
This balance is already being struck privately on the Net — though there’s significant resistance and unease about it. As I’ve already described, in January 2005, Google announced that it was giving something to the Chinese government it has refused to give anyone else in the world — a version of the Google search engine that blocks content the Chinese government doesn’t want its citizens to see. Thus, if you search on “democracy” or “human rights” on Google.cn, you wouldn’t find what you’ll find if you search in the same way on Google.com. (Wikipedia now keeps a list of words blocked by search engines in China.) Thus, Google would effectively remake the Internet for the Chinese according to the values the Chinese government pushes.
I understand the motive (profit). I certainly understand the justification (it will speed China to a real democracy). But whether or not you believe this balance is right in the context of Communist China, it certainly has more justification when we’re describing agreements among democratic nations. What the Chinese do to its journalists is, in my view, wrong. If a Chinese publisher offered to publish this book in China only on the condition that I omitted this paragraph, I certainly wouldn’t. But I have a different view about rules imposed by France or Italy.
One important consequence of this architecture — indeed, perhaps reason enough to oppose it — is that it will make regulation easier. And the easier it is to regulate, the more likely regulation is.
Yet this is the trade-off — between cost and the willingness to regulate — we have seen again and again. Cost for the government is liberty for us. The higher the cost of a regulation, the less likely it will be enforced. Liberty depends on the regulation remaining expensive. Liberty comes with friction.
When it becomes easy or cheap to regulate, however, this contingent liberty is at risk. We can expect more regulation. In these cases, if we want to preserve liberty, we will need to develop affirmative arguments for it. We will need these affirmative arguments to prevent identity-based regulation of the Net. As I explain in the balance of this book, there is both a surprisingly great desire for nations to embrace regimes that facilitate jurisdiction-specific regulation and a significant reason why the costs of regulation are likely to fall. We should expect, then, that there will be more such regulation. Soon.
The effect, in short, would be to zone cyberspace based on the qualifications carried by individual users. It would enable a degree of control of cyberspace that few have ever imagined. Cyberspace would go from being an unregulable space to, depending on the depth of the certificates, the most regulable space imaginable.
- 6.6.4 sigwait and sigwaitinfo
- Glossary Acronyms and Tools
- Chapter 6 Creating and Managing Virtual Machines
- The Windows Driver Model
- Message Routing and Delivery
- 6.5.1 User and terminal identification
- Introduction to Microprocessors and Microcontrollers
- Serial and parallel transmission
- 6.4.2 getchar unlocked and putchar unlocked
- Delphi, Kylix и Borland C++ Builder
- 11.7 Soft Timers and Timer Related Operations
- An 8-bit microprocessor – the Z80180