hardware-based digital mixes, i.e. sealed modules with a public key present only inside the module and unreadable by outsiders, will mean no human is even involved in the process, even as a system administrator. Long chains of such mixes, operating quickly on highspeed networks, should make the task of tracing messages even more intractable. A commercial implementation of a digital mix, called MixMaster, is available; users can install such âinstant mixesâ on their Internet boxes and become remailers. This turns out to be a good example of what a simple application of strong cryptography, using PGP, can do. The Perl and C code is short and simple, and the security of the entire chain depends solely on the unbreakability of encrypted messages, on the number of hops, and on the unlikelihood of collusion between the various remailers. (If all of the remailers were to get together and compare notes, the system would of course be broken. But as the number of remailers increases, this strategy becomes less and less effective. Also, one can always remail messages through oneself, thus defeating most collusion or tapping efforts.)
Another approach to remailers is the one followed by Julf Helsingius, of Finland, who operated an anonymizing service that kept a database of mappings between pseudonyms and actual e-mail addresses. This system was easy to use, and allowed easy replies to senders. However, the database was a ripe target for civil lawsuit investigators (and criminal investigators), and Julf pulled the plug on his system in 1996. Cypherpunks remailers, by being distributed, in many jurisdictions, and robust against such requests, offer a more solid and scalable basis for anonymous remailer networks.
âDigital postageâ is needed both to incentivize remailers to operate for-profit sites (and thus expand the number and robustness of these sites) and to provide a more solid economic basis for e-mail in general. E-mail currently costs most users nothing to send; this has led to widespread âspammingâ of the Net. (Consistent with the themes of this article, what is needed is not global regulation but a market-based pricing mechanism for e-mail.) Some work on digital postage has been done, but true progress awaits wider deployment of digital cash systems.
This use of remailers is just one concrete example of the use of cryptography to alter institutions and interactions.
True Nyms
The controversy over naming and under what circumstances true names can be demanded is likely to rage for decades.
Why do we so often accept the notion that governments issue us our names and our identities, and that governments must ensure that names are true names? Governments like to be involved in identity issues because it gives them additional control. And it helps them to track the flow of money. For example, centuries ago, the rulers of various European countries forced the Jews to drop their traditional patronymic practices (âJacob son of Israelâ) so as to allow taxes to be more efficiently collected, to monitor movements, and so forth. These rulers even sold the âbestâ family names to those who paid the most, leaving others with less desirable, or even insulting, names. The same practice was repeated in the U.S. with the naming of ex-slaves and the renaming of immigrants. As Nietzsche pointed out, âThe masterâs right of naming goes so far that it is accurate to say that language itself is the expression of the power of the masters.â Governments today even give themselves the rights to create/forge completely false identities, with false credit histories, false educational backgrounds, etc. Under the guise of âprotecting witnesses,â the Federal Witness Security Program, popularly called Witness Protection, has created upward of fifty thousand fabricated identities. The major credit reporting agencies are, of course, not fooled, as these âghostsâ pop into
The Cricket on the Hearth