I found the China-Google Censorship article interesting. But I wonder how much of a problem is censorship really. The Internet is inherently decentralized. That makes censorship very difficult in reality. The reason China is able to get away with it is because the endpoint of the content sits in a single place. That is, if you want to search for "illegal material" on google which China is censoring, you have to go to google's home page. This is a single point of access which is easy to censor.
But think of all the technologies out there which are used to download copyrighted material. Peer-to-peer technologies came out almost a decade ago, and what's out there now, I'm sure is far more sophisticated (i2hub, gnutella, kazaa, bit torrents). Now I don't download either legal or illegal audio or video on the Internet. So, I'm not familiar with how any of these technologies work. But I'm sure search engines might be able to learn a lesson from the success of some of these technologies.
Here's one idea I had. After reading the article, it took me a few minutes to think of this. I'm sure there are much better ways to circumvent censorship. Suppose I make essentially a front-end to google, and I call it theroaminginternet.com. The first thing I would do is encrypt the entire site. That way it's impossible to filter based on content. The government will still know who's using my site, but not what they're looking at.
OK, but if theroaminginternet were to become popular, it would probably just get blocked altogether. Fair enough. The next step I would take is to set up a simple API so other sites could easily access my own. They would become a front-end to my site, and I would pay them a percentage of the money I earn to encourage them to do so. If I write all the software and make it easy to set up, it would only take a few hours of their time and about $6 for a domain. Now I have decentralized google and made it nearly impossible to track or filter in any way. Anyone could set up a new domain at any time for very little money.
Of course there is still an obvious problem. How are people supposed to know about these new domains that are coming and going? You would have to make a web site that lists the domains. There would have to be software that would continually update the web site, but that's not too hard to do. There might also be user feedback, so it would be easy to know which sites were being blocked at the current time. Of course this site could be easily blocked; so it too would have to be accessible through many domains.
You may be thinking this is just too complicated. No one is going to put the effort into finding the lists of unblocked domains so they can find a front-end to theroaminginternet so they can perform an uncensored search. But consider if even .1 to 1 percent of the population is willing to put in a little effort. This is millions of people. There is no way the government would have any hope of tracking and blocking such people. According to the article, much of the current censorship is done via self-censorship by using intimidation and fear. They simply don't have the resources.
All it takes is a small percentage of people who know how to access information. If a less knowledgeable person wanted to do an uncensored search, he could just ask a friend. And such information tends to spread almost instantly on the Internet. Not to mention filtering web sites would become very difficult if too many domains were set up. Every single packet that passed through "The Great Firewall of China" would have to be compared against thousands or tens-of-thousands of blacklisted domains. And it would be a constantly growing list. It could create serious delay problems for the Chinese Internet.
Anyway, my point is, censorship is only an illusion. Anyone who wants to find information should be able to so without leaving the country and without Chinese officials knowing what that material is.