>From Roger Dingledine: > On Tue, Sep 03, 2013 at 11:06:54PM -0700, Asa Rossoff wrote: >> Timeline through August 31 > > Hi Asa, > > Thanks for the timeline! Here are a few notes.
Hi Roger :) > >> * February, a dramatic, roughly ten-fold decrease in Syrian directly >> connnecting Tor users. > > This was a censoring event I believe. The vanilla Tor TLS handshake > has been unreliable from inside Syria for much of 2013. Users are much > happier using obfs2 or obfs3, and we don't have much in the way of stats > for those yet. > https://trac.torproject.org/projects/tor/ticket/5040 > >> * August 13, Report that PirateBrowser had reached more than 100,000 direct >> downloads, had 5,000 seeds on its major torrent, and was continuing to >> average "well over 1,000 downloads per hour". > >> For comparison, Tor sees 40+ million downloads of TBB a year from its > website. So careful jumping to conclusions about how many Pirate Browser > users there are based on how many downloads they counted. :) Good point. If 40 million TBB's a year = 500,000 direct user average... Let's say, generously, that PirateBrowser got 150,000 acquisitions in one way or another in those first 3 days (50,000 a day, 1.5 times the number of direct downloads) and continued to do so through the end of the month (50,000/day From 8/10 through 8/31 = 50,000 x 21 downloads = 1.05 million downloads). Let's assume a scenario where PirateBrowser users are likely to immediately start using the software and have a directly-connected to download ratio on any given day of 1.5 times the prior usual Tor user. For the sake of argument. Let's assume the usual Tor user updates their TBB 2 times per year (pulling this out of my hat). So the number of prior Tor downloaders would be 40 million downloads / 2 = 20 million. With 500,000 directly connected users, that's 1/40 of the downloaders being connected. Assuming, for no particular reason, that PirateBrowser downloaders are connected 1.5 times as much as TBB users per download, that means 1/26.67 PirateBrowser downloads are connected. Of the generous assumption of 1.05 million PirateBrowser downloads by August 31, that would have 39,375 (~ 40,000) PirateBrowser users connected on August 31. Since direct Tor connected users increased from ~500,000 to ~2,350,000 on August 31: 40,000 new users would only account for about 2% of the total number of added users. Even if I wasn't generous enough by a long stretch, and a million PirateBrowser users are connected -- and that really seems a stretch -- that would only account for somewhat over half of the new users. Asa -- tor-talk mailing list - tor-talk@lists.torproject.org To unsusbscribe or change other settings go to https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-talk
