Tuesday, October 16, 2012
Andrew Stumpff on the law as a fractal. I particularly like the explanation of Mannings "Law of Conservation of Ambiguity". a profound insight....and then there is this: "But despite a lawyer’s hopes, the rule-writer cannot provide every answer in advance. That a regulation may fail to address all the questions practitioners might have is not a flaw in the regulation, or a mistake by the regulation-writer. It is a feature of reality; part of the fabric of the universe." A fabric of the universe indeed. See Larry Tesler's Law of the Conservation of Complexity in which my ITWorld article from a few years ago is quoted re business process complexity. There is a lot of conversation going on at the moment about how similar laws and source code are in may respects. Clay Shirky's recent TED Talk for example. This is all fine and valid but for me, the most powerful similarity is the correspondence between unit tests and caselaw as a means to deal with the specificity problem Andrew Stumpff writes about in this paper. Simply put, when a question arises about what is "correct" in software, we write a unit test to test a specific case. The unit test corpus gets bigger and bigger over time, serving as a regression test suit for future questions of correctness. In law, this is called precedent. Fundamentally the same thing.