Via today’s nytimes.com (do continuously updated websites have “today” editions), there’s the AP story on the government asking Google for some data, including:

a request for 1 million random Web addresses and records of all Google searches from any one-week period […, and sic: “random web addresses?”]

which the DOJ says it needs to

determine how often pornography shows up in online searches as part of an effort to revive an Internet child protection law […]

What I wonder is why the government didn’t just ask for the aggregate statistic directly.  After all, Google publishes statistics all the time on aggregate data (and those are depressing statistics, btw).  They must have a methodology in mind for what they would do with the data, which presumably doesn’t involve Potter Stewart looking at a million web pages to do, so they could just ask Google.

It doesn’t do anything to address the privacy concerns with Google having that data, but it does address the privacy concerns of the DOJ having data on private citizens (although it’s hard to tell from the AP story whether any private data would actually be disclosed in the subpoena).