Actually most libertarian are strictly against any form of IP and patents. They prefer trade secrets over state regulation.
But then again I don't know if you mean real libertarians or just crypto-statists calling themselves libertarian because they don't know better...
The objectivists just find it to be to their advantage, these days, to make it appear that they might. The controversy that you suggest is as manufactured as the controversy about anthropomorphic climate change.
And I'm aware a sizeable chunk of anarcho-capitalists think IP as a concept is bullcrap.
Then there's the shift in power that comes from control over weapons and infrastructure ... but AFAIK the coordinated masses still have power on their side - partly from the assumption that given enough "external" disagreement, some people on the inside are likely to cross over.
Indeed, one could argue that the most terrifying part about the Snowden disclosures on mass surveillance, is that the current government is poised to deploy counter-measures long before any opposition consolidates and organizes. This might take the form of buying out individuals - or "dissappearing" or taking out individuals that are perceived to be in the process of becoming important.
Society/the government is ostensibly afraid nobody will invent if they are not artificially advantaged. Those with working intellect consider that trade secrets are enough of a protection against competition.
The article went further into how trade secrets allowed a company to have its employees sign non disclosures and actually sue the employee and whoever else used the employer's "stolen" knowledge in criminal court, while patents changed this to be only civil and prevented employees from being sued at all.
As everything in life, a balance between the two is probably the most optimal. So I agree patents might have started going too far one way, but I doubt trade secrets would be any better in addressing OP's complaint that patents prevent the free transfer of knowledge and limits innovation.
While I never lived when trade secrets were the norm, I believe most people supporting it in favor of patents suffer from the same problem a lot of younger programmers suffer from. We keep reinventing the wheel and making the same mistakes over and over. It's all a cycle, we'll go back to trade secrets and then their issues will become more "apparent", people will have forgotten the inherent issues with patents and we'll move back to patents again, repeating the cycle.
So I personally disagree with both, you rarely solve issues by reverting back to older ways which the people before us had already tried in vain. We have to try something else, maybe a hybrid, maybe something totally different.
He didn't work alone, did he? I think he had a company, but you can bet he got a sweet deal with his employees on those patents.
I think there's something valid about a government protected temporary monopoly. But in the case of copyright, for example, it's life plus 70 or more years. It is ridiculous, as in, worthy of ridicule. Medical patents are 20 years, which itself is too long, but it's made worse by rather permissive ways in which they can be effectively renewed by barely changing the medicine.
You know what it is? It's a way to get white collar work to continue to pay out more than, and longer than, the actual amount and duration of work on a thing takes. It's an incentive to do that kind of work. But I agree that it's grossly lopsided when you can't do repair or research on that same material - that's one of the original points of patent and copyright law. Put your ideas on paper so that you not only get protection for a while, but others can create derivatives that are better, and thereby we learn more about it all.