Different fields are affected by patents differently. If I had to choose the current system and no patents, I'd choose no patents, but I see how there is space for other things.
In Software, patents are silly because everything is derivative, and it's very difficult to separate the obvious from the non-obvious. It's not as if most patents are coming from many years of research: Enough eyeballs often would have arrived at the same conclusion.
We'd also still build software without patents: It's just too useful not to keep working at it. Copyrights are more than enough.
So now let's look at a very controversial one where patents make all the difference: Soybean seeds. If I spend 15 years developing a seed that yields better, copyright will not help me. But the minute I sell a seed, you can make as many copies of it as you want. You don't even have to buy the first seed from me, since the seeds themselves are what the farmer sells. Therefore, investing in seed quality becomes something that is extremely hard to do in a for-profit organization, because you have to recoup your entire investment in the first year, as in the second one, anyone that has a field can be your competitor. Patents in the field are also rarely dependent on each other: Different genes can be stacked on a plant, but unless you developed the entire suite of them all at once, it's not as if it's likely someone is going to find a little refinement that they'd love to sell, but depends on a different party's patent to also be on the plant.
So in that field, no IP protection completely changes the game. Investment in soybean seed research becomes a purely government task, as it's impossible to get any ROI out of it.
Now, one can still be against life sciences patents, but it's easy to see how removal of said patents would lower the effort spent on research, while on software, if anything, it'd make it go up, as we'd stop being afraid of doing something that we find obvious, and that someone else found patentable.
So on each field, patents have different dynamics, and should be evaluated independently. I don't think you'll find many pro-software patent people that aren't just pro-patenting everything. And those really don't make any sense.