A new view of intellectual property and software

this article spent considerable effort attempting to answer this question. After a thorough study of the issue we realized: We don't know. What's equally important here is neither do you. There is an interesting, important, unsolved question here that is open to empirical investigation: What is (are) the metabolic rate(s) of the software marketplace(s)? How long is long enough for the market to establish the worth of an innovation? The issue need not be the stuff of invective and opinion; we have some real homework to do to come to understand our industry. The authors suspect the right number is far closer to about 2 to 5 years than it is to the 17 years of a patent or the 75 years of a copyright. But suspicions are not the issue, real research is. If we are to provide lead time by restricting behavior clones, we will inevitably have to make the judgment call of how similar two programs must be in order for one to be a clone of the other. That is, what is similar enough to precipitate market failure? We have developed a metric based on three properties: 1) what was cloned (the extent and significance of the behavior and design overlap); 2) how it was created; and 3) the proximity of the second-comer's product and market. Market failure is clearly more likely if 1) the amount and significance of the behavior and design taken is substantial (a near complete overlap) ; 2) the second-comer's development effort is rapid, easy, and highly dependent on the first-comer's product (based on extensive black-box testing of the innovative product); and 3) the products are nearly identical and are to compete in the second market. The converse is also true: The smaller the taking, the less dependent the creation, the less similar the results and less proximate the markets, the less li...