Perhaps approaching gaming as a streamed service in the manner proposed by OnLIVE (http://www.onlive.com/) could be a viable means of reducing piracy. It seems possible for a sizeable games publisher to move to a subscription based business model.
Except games can't and will never be run off-site as a service - the technical capabilities for OnLIVE are not and will never be available in the US.
More and more games will be moved to models that require a persistent internet connection though - the article does a good job of pointing out that winning the battle for just two months is nearly "good enough".
It's very do-able today. The key is just having the server located close enough for latency to not be a problem. ISP partnering would be their best bet. Run the service and let ISPs resell it -- locate the servers in their headends/datacenters. If you traceroute something and look at your first few hops it's easy to see how it would work. <10-20MS should be fine.
I don't think there are technical limitations holding this back. You can already get the necessary low-latency video encoding as open-source software, even:
http://x264dev.multimedia.cx/?p=249
Gaikai seems to have pretty good business model to make it happen, too - better than OnLive anyway.
It's difficult to look at the computer world, the rise of the internet, and onlive itself and not see that games are really no different than other aspects of computing, which have already migrated to the cloud...
How many "nevers" have been obliterated by advancements in technology?
