I'm not arguing that the scenario I painted was the usual scenario, or even common. I was just putting an example together to show how his defense would not work in principle. That is, a legitimate patent scenario where the defendants were infringing without any knowledge of the patent existing.
Basically, I was trying to point out that wide infringement without knowledge of a patent does not a priori mean a patent should be invalid.
Actually, I think you're right about the way invention usually happens. Software patents are by far a net loss for the industry, and I'm against them. I just like to reply when I see what seems to me to be a flawed argument, whether I agree with the conclusion or not.
The purpose of a patent is so that the original inventor can make money from his invention whilst not having to keep the actual invention a secret. Patents have a time limit (not sure what that is) in which the inventor gets to make his money before the invention is open to society.
Maybe we should ask to change the law to say that because of the nature of software and the rapid change of growth in emerging technologies, the length of time a software patent owner gets to generate money from his invention is limited to a much shorter length of time, say 1 or 2 years. This way, legitimate patent owners as per your example still have time to generate income and establish themselves as market leaders in whatever technology, but there is much less incentive for people to patent troll, since the patents aren't intrinsically worth as much.
Also, perhaps you shouldn't be allowed to sell patents on to other people. That seems completely against the spirit of the whole thing - I should only be allowed to license you my patented idea. If I sell you my patent, does that mean you came up with the idea, suddenly? Buying and selling patents is surely only for evil purposes...
Basically, I was trying to point out that wide infringement without knowledge of a patent does not a priori mean a patent should be invalid.
Actually, I think you're right about the way invention usually happens. Software patents are by far a net loss for the industry, and I'm against them. I just like to reply when I see what seems to me to be a flawed argument, whether I agree with the conclusion or not.