This is the third in a series of blogs.
To recap: In 1996, during the Clinton administration, Congress passed a law (CDA) intended to prevent children from being exposed to harmful material. The Supreme Court struck it down as unconstitutional on First Amendment grounds. Congress came back in 1998 and passed a new law (COPA) with narrower parameters. The ACLU immediately asked for and received an injunction. The Supreme Court upheld the appeals court's injunction against COPA in 2004, again on First Amendment grounds. This time the Court suggested that the Bush administration either come up with another law or prove that COPA would be more effective than filtering software in preventing material that could be harmful for children.
Instead of shepherding a new law through Congress, the Bush administra- tion chose to take the latter route of proving that COPA is the only viable way to combat harmful material. According to CNET News, the Department of Justice plans to argue in court "that filtering software is not a realistic alternative to a federal criminal law because the concept of filtering is flawed and unworkable in practice." In other words, the Bush administration believes that even an innocuous search can inadvertently turn up material inappropriate for children.
And, instead of doing the work to prove their contention themselves--by doing searches children might do on the Internet--in August, 2005, the Department of Justice (DOJ) under Alberto Gonzales subpoenaed four search engine companies for all their URLs and a month's worth of the searches done on their engines.
The recent DOJ motion indicated that there was significant negotiating before the government agreed to narrow its request for information on the grounds that a simple list of all the URLs would be untenable and a list of all searches done would be tainted by the web-crawlers doing searches in order to rank websites. Privacy issues were also raised. Therefore, the government narrowed its request to a million random URLs and the search string (text only) requested by users for one week without identifying the users. At that point, it appears that AOL, MSN and Yahoo turned over the requested material.
According to a government motion filed last week, Google listed the following as their reasons for not complying (included are my comments):
1) Relevancy: Google objects to the implication that their "search-engine database is reflective of the entire world-wide web." More relevant information could be obtained by the DOJ by simply doing their own searches on the Internet.
2) Privacy: "Google next objects that its compliance with the request would require it to produce information identifying the users of its search engines."
What's to stop Gonzales' crew from coming back and saying, "You've already given us the searches, now give us the users"?
3) Redundancy: MSN, Yahoo and AOL have already turned over their information. What's the point in having data from Google as well?
4) Privilege: Google does not want to share trade secrets. Google is apparently afraid that, with a large sample of data, competitors could reverse engineer their way back to the algorithms Google uses in its search engines. Those algorithms are a closely guarded secret.
5) Undue burden: The request would require too much effort on the part of Google
Remember--this is not national security. The government is not trying to find terrorists and has not yet proven that this information is its one viable means to protect children from harmful material online.
CNET News has been conducting an online poll. Of the 1,700 respondents, 83% believe Google was right in refusing to hand over the requested information while 14% are uncertain. Only 3% believe the data should have been given to the government.
What happens next? Google must answer the DOJ's motion. Usually that would take about two weeks although the deadline might be extended. Some legal experts believe that the Bush administration is going to have an uphill battle in proving their urgent need for this data.
Stay tuned . . .