Personally, I am thoroughly sick of people who automatically assume that patents, a massive government interference in citizens minding their own business, will encourage innovation in every area of human endeavour when it's quite clear they don't. If they're going to argue for this government interference then they should at least have some scientific evidence that it is a net positive in any particular area. Not the usual childish handwaving about how an inventor won't invent without patents (history shows this is nonsense) and that patents have no harmful effect on society and the free exchange of ideas (also nonsense).
Nobody suggests that all inventors will stop inventing without a patent system. The idea of a garage inventor is reasonably argued as not significantly motivated by the hope of obtaining a patent. I argue against this position, for I know if I were such an inventor and saw an invention I was the first to invent in Walmart without any compensation to me, I would feel wronged.
However, thinking about areas of science and technology where the cost of entry is high, the need for patent protection becomes clearer. Without sufficient protection, products that require substantial R&D expenditures are less likely to be developed because the developed product can then be produced by competitors. This is most evident in the field of pharmaceuticals, with the constant push-and-shove between developers and generics.
A patent stops billions of people from using an idea that's probably going to be independently re-invented many times so that one (1) person can have additional incentive to invent something. Explain to me why this is a net positive?
I'll gladly respond. First, your presumption that an idea that is going to be independently re-invented is already addressed in the patent code, specifically 35 U.S.C. 103, not allowing patents that are obvious. Given the expansive interpretation given to this section by the Supreme Court in KSR v. Teleflex, examiners are given the ammunition to reject claims for inventions that are likely to be independently re-invented, as these inventions are likely obvious. This is so because inventions that are likely to be independently re-invented are just that because there is a market demand for an invention, there is an associated motivation for an inventor to satisfy that demand. Hence, obvious.
Particularly for software industry where the entry cost is so low?
Fair enough. However, the cost of entry is not considered in the patenting process. A fair remedy could be to have different standards of non-obviousness for some areas where entry costs (can) be low, i.e. software and mechanical applications, and another standard for areas where entry costs are high, i.e. pharmaceuticals.
However, I would argue that this is unnecessary. If an examiner is truly an expert in the field, as they are presumed to be, they will be able to recognize those pieces of software which are obvious and those which are not, keeping in mind the inherent standards of obviousness for their area/art unit. I guess you call this hand-waving; I call this understanding the patent prosecution process.
Every new patent is a new law; another opportunity for a lawyer to make money at the expense of the wider community.
If you feel that way, I encourage you to write your congressional representatives for a repeal of the patent system. Don't be too surprised if your view doesn't gather too much support.