‘Code-is-law’: a primer
In traditional Internet computing, users of a program must trust that the creator, who has complete control over the servers on which that program runs, will act in good faith. The users hope, but can never be sure, that their stock exchange app won’t front run their trades, or that their file-sharing service won’t suddenly deny access to important documents, or that the content of their private conversations won’t be observed or modified.
This power imbalance prevents the Internet from being a truly political space with shared control over important decisions. A vote on changes to Facebooks terms of service, for example, could never be legitimate when Facebook engineers could simply adjust the vote tally.
With decentralized computing, such trust is no longer necessary. Engineers can write code that, once deployed, no single individual can modify. This has made it possible, and therefore worthwhile, to try and solve difficult governance, social, and legal problems by designing digital mechanisms.
One of clearest expressions of this dynamic is a concept called “code-is-law,” which generally describes the notion that a program or set of programs, designed around the right incentives, can play the role (or at least part of the role) of a legal system amongst the community that uses that program. As a simple example, say you built a program that holds funds, which the program can invest. Participants can contribute money in exchange for shares, each of which gives them one vote on how to invest the entire fund (perhaps by purchasing and holding Bitcoin). The vote will liquidate its investments and return a proportional share to the participants with a majority vote.
