And because § 101 of the Patent Act says that it precludes patent protection for laws of nature, and Math and Logic just exist *AS LAWs OF NATURE*, NO implementations or expressions that use Math or Logic should be patentable.

I have trouble seeing every piece of software in existence as a law of nature. You jump from saying a law of nature not being patentable to everything using certain laws of nature (math and logic) being unpatentable. But doesn't every invention use laws of nature? A mechanical invention is mechanical as much as software is math. An optical invention is optical as much as software is math. A chemical invention is chemical as much as software is math. If software being math is an argument against software patents, isn't it an argument against all patents? That obviously is not the intention of the law.

To me an important reason that software patents don't make sense has to do with obviousness. An invention must not be obvious to someone with ordinary skill in the art to be patentable. How obviousness is actually tested in courts is, judging from the bits and pieces I picked up about it here and there, a procedure with a high likelyhood of being completely disconnected from reality and therefore problematic. Ignoring that, there are I think about 900,000 software engineers in the US. Assuming their skill, like intelligence, follows a normal distribution, the term "ordinary skill" seems to suggest that less than 450,000 software engineers should think the invention is obvious. But not too much less, otherwise "ordinary skill" is a nonsensical term to use. But let's take an invention that is obvious to only the most brilliant 1% of software engineers. That's still 9,000 people. Most inventions happen because the field has advanced to a level where some limitation starts to be felt like problem. All it takes is someone from this group of 9,000 people to be in a place where this problem is felt and the invention will be done. A much larger group of software engineers will not think the solution is obvious, but they will still be capable enough to get there. And many of the 9,000 or the larger group will be able to find alternative solutions. Solutions for the problem will be found in these conditions. Patents add no value at all (and it's clear that they do a lot of damage). The same will be true for any field where so many people have expertise and which can be practiced without having to do the kinds of investments only the largest corporations can afford. The criterion should not be obviousness, and certainly not to someone with "ordinary skill" in the subject matter, but the likelyhood an invention will be lost to society without patent protection. For the large majority of "inventions" in software that likelyhood is zero.

[ Reply to This | Parent | # ]