[epi0Bauqu]

I'm planning on blogging a bit on this particular topic tomorrow. If you have any questions, please let me know and I'll try to address them.

I don't think the gTrends argument works though (see http://news.ycombinator.com/item?id=4741590). Most of the results inserted are super-recent (and so I expect the recency of the trends data -- like the last 90 days -- to dominate). Of course it is all a black box so who knows.

Some of the more interesting things to me are:

--you can't reliably de-personalize (as you cited).

--the variation in results across our study was great.

--variation for signed out (even incognito) users was not much different from signed in users.

[carbocation]

It would be vaguely interesting to know from study participants how unique their browser is.[1] Mine appeared to yield the same result whether in incognito mode or not, but I can't really see Google using this approach.

Your point about Google trends is counterintuitive but evidence-supported, which makes it particularly interesting. It may well be that they are looking at outcome measures after search personalization, and abandoning personalizations that yield no results. Or perhaps a certain volume is required over time for a particular query before triggering an automated personalization trial. As you say, it's a black box, so who knows.

[1] = https://panopticlick.eff.org/