On 2/19/10 7:13 PM, Karsten Loesing wrote: > On 2/19/10 5:00 PM, Andrew Lewman wrote: >> On 02/19/2010 05:20 AM, [email protected] wrote: >>> http://metrics.torproject.org/bridge-users-graphs.html#china >>> >>> if there is no clear explanation account for the doubling of the >>> usage figure in the whole December, i would speculate that this is >>> an error in the estimation. could anyone confirm this? >> >> The best person to answer this is Karsten, and he's currently traveling. >> We await his answer. > > That's a fine question. It's already on my list. I'll let you know as > soon as I have a better answer than "probably something wrong with the > measurements."
I figured out the problem. The metrics portal had the bridge user numbers from 2009-11-30 to 2010-01-05 imported twice. This affected all countries, but was simply most visible for Chinese bridge users. I removed those days from the stats and imported the descriptors again. The corrected graphs can be found on the graphs page: http://metrics.torproject.org/bridge-users-graphs.html#china For reference, here is one of the old graphs: http://freehaven.net/~karsten/volatile/china-bridges-180d-bug-2010-03-10.png Thanks for letting us know! --Karsten *********************************************************************** To unsubscribe, send an e-mail to [email protected] with unsubscribe or-talk in the body. http://archives.seul.org/or/talk/
