Did you look at the demographics of graduate students in the US lately? China has now programs to revert the brain drain, offering nice jobs to successful Chinese academics in the US and elsewhere.

Just look at the attendance list of scientific conferences. You will see at least one or two Chinese names in every big international event, in pretty much any field. The ones that are held in China has more than half the attendants from China.

And they are good. They have a tradition of working hard, and there are a lot of Chinese. Once it becomes economically attractive, it is inevitable that a lot of bright Chinese people will go into research. This is already happening in my field (astrophysics) but I feel other life sciences have similar situations.

Not that I am complaining. I am looking forward to visiting Beijing next October and start new collaborations.

Actually there is a neat solution: deny corporations having patents. They are not individuals after all. There is no reason for them to enjoy the rights of an individual (but they do) since they cannot be punished the same way (eg., cannot be imprisoned). At least do not apply same standards to individuals and corporations, eg., shift the burden of proof.

Of course this would open a huge can of worms, and it needs to be implemented w/o loopholes, but I think it would be a step in the positive direction if implemented.

The arXiv system (www.arxiv.org) already hosts just about every preprint that comes out in high energy physics, astrophysics, and several related disciplines. Access is completely free, and they currently host 400,000 papers. Needless to say, people post there for a reason: it works really effectively to get research results out to the public quickly and efficiently, and as mentioned before, it's totally free for everyone involved. Open access isn't a theoretical question taking place in a vacuum, it's already underway, and it works just fine, and can even coexist with the refereed journal system, as the physics world has learned over the past decade.