Splitting hairs here. Nobody can afford to pay licenses for software patents. Any one sufficiently large piece of software is covered by tens or hundreds or patents. The only way you can afford to license them all, is if you're prepared to publish at a loss.
So, because you can't negotiate proper licenses, you shouldn't have to pay for anything? That's like saying that the price of a CD is too high, so I shouldn't have to pay for any music, anywhere.
If we would truly negotiate for and license every single patent major software products violate, the software industry would grind to a halt. The patent system is ill-suited for the software industry as a whole. And from all possible ways to handle software patents, the US in particular has chosen about the worst one.
Whether it is obvious or not is beside the point.
I thought it was central, because if the invention is obvious, then you've got a justification for claiming any patent is invalid and therefore using it without paying isn't infringement of a valid patent.
But, okay - given an invention that is novel and not obvious, what's your justification for still not paying for a license?
Software patents shouldn't exist in the first place, is what I was getting at.
They demonstrably (seriously, there have been studies) slow down invention in the software space. So why put up with them?
Except that there haven't been studies... There's been sort of meta studies based on ancillary data to sort of argue in a roundabout fashion that patents slow down software innovation, but not any actual studies, because software has been patentable in some form for as long as software has existed. It'd be like a study purporting to claim that due to the economics of fossil fuels, using electricity from any grid that includes non-renewable sources slows down innovation in the software space. Uh, okay... There's no real data, and it's mostly hypothesis, and you certainly can't compare it to anything since there isn't a society out there that doesn't do it, so we'll take it with a grain of salt.
At the very least, litigation is costing major software companies more than they are actually gaining from licensing. We do have something to compare to btw, an entire software industry with much less patent proliferation up until the early 80s.
But we can also ask ourselves: how many software patents really apply to inventions that would otherwise would not have been worth coming up with? You think nobody would've gone through the trouble of coming up with MP3 had software patents not existed? Software design is not the same as drug design, and simply shouldn't be judged under the same rules.
Patents are not there to make you rich. They're there to promote innovation. They're clearly not necessary, and so most likely to be an impediment instead. Which studies do back up, no matter what you seem to think.