Could be related to the Yahoo! policy change, since part of the story is that Marissa Mayer ran analytics on how their VPN was being used and concluded that many remote workers weren't working much:<p>"Mayer explained the rationale at Yahoo’s “Friday FYI,” its equivalent of Google’s TGIF. ”We’ve checked and some people who work from home haven’t even logged into the VPN…” she apparently said."
This is probably related to the Copyright Alert System, or "Six Strikes" rule that many ISPs have adopted.<p>VPN is the best way around this and it costs next to nothing.
OP here. Should have posted sooner before it looks like my post got flagged?<p>I think the shock[1] is being caused by the Copyright Alert System put into place on Monday by Comcast. Here's why:<p>Comcast is the largest ISP in the US, so a change of their piracy policy (the exogenous shock in my story/model) is likely provide a large enough segment of Internet users (definitely bigger than the set of Yahoo employees or SimCity fans) with the appropriate incentives to search for information about VPNs. The logic here is simple.<p>[1]There is clear weekly periodicity in the 90 day chart. If you were to detrend the data and remove the autocorrelation, the spike would look even sharper.<p>Based on the data and the normalizations on the data that Google does, 90 day data is much better than 12 month data because the latter is not daily data anymore -- it is data aggregated weekly. Since it is obviously clear that the data has weekly periodicity, aggregating the data at the weekly-level ignores the fact that how many searches there are for "VPN" on a given day depends on what day of the week it is.<p>Calling this wild speculation, incorrect, random noise, or comparing it to weekly or monthly data is just not very good statistics in my mind.<p>EDIT: Also, for clarification Google Trends data is search volume data normalized on an index from 0 to 100 where 100 is the relative maximum over the series. Tuesday's index value was 100, with Monday's being the second highest over the 90 day period. So technically, yes, search volumes for VPN are currently spiking this week. I'm really curious to see what happens when Wednesday's data is released.
Yet drastically down compared over a larger timeline. <a href="https://www.google.com/trends/explore#q=VPN&geo=US&cmpt=q" rel="nofollow">https://www.google.com/trends/explore#q=VPN&geo=US&c...</a>
Much more drastic are searches for seedboxes <a href="https://www.google.com/trends/explore#q=seedbox&geo=US&date=today%203-m&cmpt=q" rel="nofollow">https://www.google.com/trends/explore#q=seedbox&geo=US&#...</a><p>edit: for those that don't know, seedboxes are servers you rent that just run bittorrent. Makes it easy to skirt around these 6 strikes programs
Great job on wild speculation HN!<p><a href="https://www.google.com/trends/explore#q=VPN&geo=US&date=1%2F2011%201m&cmpt=q" rel="nofollow">https://www.google.com/trends/explore#q=VPN&geo=US&d...</a>