Subject: A reply to Lessig's _Code_ Date: Mon, 06 Dec 1999 11:56:43 -0800 From: Brian Holtz Organization: Sun Microsystems To: kaplanc@nytimes.com, declan@well.com, lessig@POBOX.COM 1999-12-03 © Brian.Holtz@holtz.org What Lawrence Doesn't Get: A Reply to Lessig's Code Lawrence Lessig's book excerpted at http://www.code-is-law.org attempts to establish the flawed thesis that, in cyberspace, software code is like the legal code, and should therefore be regulated by government. "Regulations of law" are completely different than "regulations of code". Laws are enforced by people with handcuffs and guns. Code resides in products that you can freely choose to use or not use. If you freely choose to enjoy the benefits of a piece of code, you are also freely accepting any known quality and liability limitations that come with it. If you want the code's vendor to be responsible for every unforeseen future complaint you may have about the code, then find a different vendor, or write the code yourself. If contract law were not "so eager to allow liability in economic transactions to be waived", so that code vendors could be liable for effects and usages that no one could reasonably predict, then nobody would be willing to vend code. If hammer vendors were liable when you used one to swat a fly on your window, then nobody would be willing to vend hammers. Lessig proposes that source code should be archived by the government before it is granted copyright, as a way to later help "undo the consequences of bad code" whose source has been abandoned. First, it would be a monumental task to archive all the source of every released version of software, since patches and even new releases can be generated literally every day. Second, the source is not very useful without its compilation environment -- the proper versions of the underlying operating system and other platform software. 'Software 'maintenance' may seem like an oxymoron, but it is in fact a major engineering challenge. Lessig suggests that copyrighted code is not archived because the copyright system is more concerned with profit than with knowledge. Lessig knows that this is not so, or else he would propose eliminating copyright profits. He doesn't, because he knows that only copyright protection can encourage the knowledge creation he desires. [The real copyright issue presented by digital networking is of course the looming inability of copyright holders to prevent near-effortless reproduction and distribution of their works. It is a problem that will affect every form of data (writing, music, images, video) except two: executable and perishable data. Software (executable data) can decline to function if not licensed, and live events (perishable data) can be hard to reproduce and distribute quickly and easily enough.] The sentence "code regulates all these aspects of our lives, more pervasively over time than any other regulator in our life" could have been written by Ted Kaczynski. :-) One may as well say that our lives are "regulated" by internal combustion, or by language, or by fire. If a technology is so useful as to be considered by rational people to be compelling, then the compulsion here comes not from the technology, but from rationality itself. It is irrational to decry the rule of rationality. The Y2K problem is not "environmental damage", any more than VCRs blinking 12:00 are "environmental damage". The environment consists of natural resources such as air, water, and land, not artificial resources like technology or culture. If things like computer software or human languages could be policed for "environmental damage", then this use of the phrase "environmental damage" would itself be a criminal pollution of English. :-) Lessig the legal scholar offers a grab bag of blanket justifications for government intervention. * "When private action has public consequences". Too broad: any private action that the public knows about has by definition "public consequences". * "When shortsighted actions threaten to cause long-term harm". Too narrow: threatening to cause any real harm -- to others, and against their consent -- is aggression, warranting government intervention. * "When failure to intervene undermines significant constitutional values and important individual rights". Too obscure: the purpose of government is not vaguesly to protect "values" but precisely to protect rights. The trick here is to specify those rights, rather than hand-wave at them. * "When a form of life emerges that may threaten values we believe to be fundamental". This makes little sense; perhaps it is explained elsewhere in the book. Lessig drags out a tired anti-libertarian straw man: "if we believe that government cannot do anything good". Only anarchists believe that. Lessig complains about the "environmental disaster" constituted by "the loss of privacy". In the absence of government coercion, code only threatens the privacy of those who want to a) enjoy the freedom to electronically associate while b) denying that very freedom to their electronic associates. If you don't want the grocer to know how many condoms you buy a week, then don't patronize a grocer that can remember your face or your customer number. If you don't want web sites to track your visits, then don't visit them. If you don't want your fingertip to hurt, don't keep hitting it with a hammer. :-) [The looming private threat to privacy comes not from electronic associations remembered and collated by our electronic associates. The greater threat comes from nano-technological recordings collected by our physical associates. Starting in less than fifty years, no human will be able to be confidant that a techno-savvy coworker or cohabitant is not surveilling him.]