Firstly, I want to establish that the “Linux is free” argument isn’t really a factor. Most average users are going to buy computers and not make them by themselves. And these computers they buy are likely to come with Windows pre-installed. So technically it doesn’t cost too much (if anything) to stick with Windows. It’s a sunk cost. And basic economics tells you that sunk costs should be ignored.
More importantly, most people don’t think they have an OS problem. They aren’t going to go out of their way to find a new OS that gives slightly more stability or productivity. They’re comfortable sticking to what they have. Microsoft have created a really big dip in the OS business. The biggest hurdle Linux faces in getting to the masses is not the technical issues. It’s not drivers, migration, etc. Its biggest hurdle is to be remarkable and unique enough to stand out, to redefine the mindset of what personal computing is. Currently, “PC” is already synonymous with “Windows” (look at the Apple ads if you don’t believe me). And if Linux doesn’t change that mindset – if it doesn’t completely shift the current paradigm – people aren’t going to take notice of it, and it will not reach the masses.
How? I’m not sure. If I did, I’d be doing it myself. But I think developments like SymphonyOS might be on the right track. Technologically it might not quite be there yet. But it’s different enough to have the potential to succeed, in my opinion. And that’s what Linux needs to strive for. Not to just be better (it probably already is) but to be different and remarkable.