What’s the big deal about the DOJ’s request for a random selection of ‘anonymised’ searches?
The US Department of Justice asked Google recently for a selection of the searches people made over a two month period, with the identities of the searchers removed. Google, to its credit, has resisted this while the other major US search engines have not. John Battelle and Danny Sullivan have further analysis of this issue.
There are four issues that arise from this which have not been prominently dealt with in discussions I’ve read:
1) If Google hadn’t resisted this request, would we have heard about the fact that AOL, Yahoo and MSN immediately complied?
2) How many people realise that Google (and others) frequently do keep track not just of what people search for but who searches for what?
3) Mightn’t there be some searches that tend to reveal the identity of the searcher even if the searcher’s IP address is concealed? Particularly if you examine them in time order? Update: Search Engine Watch just produced a lengthy analysis of this issue which suggests that this is probably not a big issue.
4) The DOJ doesn’t seem to have asked only for searches by Americans. Does it have any business looking at searches from international users who might (for whatever reason) have chosen to use yahoo.com instead of (for example) yahoo.co.uk?
Update: Tim Wu of Columbia Law School writes an excellent analysis of the whole issue on Slate, arguing that it’s simply excessive for search engines to collect these reams of personal data on what people are searching for.