please point out ANYWHERE that I have said that patents are a good idea for software?
The conversation above was about drug research, where patents tend to be fairly narrow ('use of this compound', frequently further limited with 'for this purpose')
I think that the 'public pays, private patents earn rewards' is fundamental abuse of the system, and all public grants should include a clause stating that the results of the research should be publicly available at no cost (I would say cost of replication, but in today's world that's so close to zero that it's better to just say 'no cost' rather than leaving the door open for abuse)
People are not advocating the end of private research, but they are advocating the end to the way that private research pays off. Unless other reward mechanisms are created, that's effectively the same thing.