Well it looks like the Bush Administration is in fact quite serious about getting that search information from Google:

Google's attempt to fend off the government's request for millions of search terms will move to a federal court in San Jose, Calif., on Feb. 27. U.S. District Judge James Ware on Thursday set the date for the highly anticipated hearing, which is expected to determine whether the U.S. Justice Department will prevail in its fight to force Google to help it defend an anti-pornography law this fall. Although the Justice Department also demanded that Yahoo, Microsoft and America Online hand over similar records, Google was the only recipient that chose to fight the subpoena in court. After the spat became public last week, Attorney General Alberto Gonzales said: "This is important for the Department of Justice and we will pursue this matter."

I'm sure it is very important for the Department of Justice, but that is no reason to force a private party to pay for the government's investigation to provide support for a law that the current administration, regardless of party, wants to provide legal support for. There are plenty of ways that the government could have acquired the information that they need to support the COPA law, but they decided to strong-arm the major search engines into providing it. When you get right down to it, there is no ethical excuse for this given how absolutely non-critical the "Child Online Protection Act" is to our country's well-being. In fact, the majority of the reason that we have the law in the first place is that so many parents don't monitor their kids' internet use. The curious part about this is that, if what CNet says about their request be true, then the government isn't even asking for the right data to prove their case:

Prosecutors are requesting a "random sampling" of 1 million Internet addresses accessible through Google's popular search engine, and a random sampling of 1 million search queries submitted to Google over a one-week period.

Random samples of URLs that have been indexed is not going to help. What they really need instead is to buy a listing of just the search terms used in the course of a week, then pay someone to write a script or program to run each unique set of search terms through Google's search engine a few times a month for about three months in order to capture the ebb and flow of Google's index. That would be the most effective way for them to capture statistical data about the amount of pornographic content encountered. Granted, they'd have to do Google image searchs too and that would require people actually look at the search results since they couldn't just automate that process for obvious technical reasons. As far as I am concerned, though, this case is problematic because it is a government assertion that it can just force private businesses and individuals to hand over data and resources to assist the government in proving a point in court without compensation. Perhaps this is the logical conclusion to Kelo v. New London. Now the government is asserting that it can force us to work for its benefit. Didn't Hayek write a book about that?...