<div><span style="color: rgb(160, 160, 168);">On Thursday, October 23, 2014 at 10:32 AM, David Fifield wrote:</span></div>
<blockquote type="cite" style="border-left-style:solid;border-width:1px;margin-left:0px;padding-left:10px;">
<span><div><div><div>In the past few months of bridge user graphs, there is an apparent</div><div>negative correlation between obfs3 users and vanilla users: when one</div><div>goes up, the other goes down. If you draw a horizontal line at about</div><div>5500, they are almost mirror images of each other. I don't see it with</div><div>any other transport pairs. Any idea why it might be?</div><div><br></div><div>I can see what could cause a simultaneous decrease in vanilla and</div><div>increase in obfs3: Tor gets blocked somewhere and users switch to obfs3.</div><div>But I wouldn't expect blocking events to look so smooth or happen so</div><div>frequently, and it doesn't explain why the reverse change happens later</div><div>(obfs3 being blocked while Tor is unblocked is less plausible). I can</div><div>also understand the overall long-term trend of obfs3 increasing and</div><div>vanilla decreasing. But I don't see why they should mirror each other so</div><div>closely over short time periods.</div><div><br></div><div>Some hypotheses:</div><div> 1. There are lots of users who have a mix of vanilla and obfs3 bridges</div><div> configured. Their tor (randomly?) chooses one of them, which usually</div><div> works. The number of such users is constant over the short term;</div><div> i.e. the sum of obfs3+vanilla is constant, but the proportion of</div><div> obfs3 and vanilla fluctuates randomly.</div><div> 2. Maybe vanilla-down/obfs3-up is caused by blocking events, and</div><div> vanilla-up/obfs3-down is caused by natural new-user churn and/or</div><div> coincidence.</div></div></div></span></blockquote><div><br></div><div>My guess is that there’s an order</div><div><br></div><div>0. vanilla</div><div>1. obfs3</div><div>2. else</div><div><br></div><div>and it *always* starts from the top</div><div>and tries going down the list until it finds one that works.</div><div><br></div><div>So when vanilla is unreachable it tries obfs3.</div><div>At a later date, when vanilla is no longer blocked,</div><div>it moves back.</div><div> </div><blockquote type="cite" style="border-left-style:solid;border-width:1px;margin-left:0px;padding-left:10px;"><span><div><div><div> 3. There is something about the way BridgeDB hands out bridges, or the</div><div> way in which users use it, that causes it to give out obfs3 bridges</div><div> at the expense of vanilla and vice versa.</div><div> 4. Some kind of feedback loop: obfs3 bridges get used and get</div><div> congested, so users switch to vanilla, which then get used and</div><div> congested, etc.</div><div><br></div><div>David Fifield</div></div><div><div>_______________________________________________</div><div>tor-dev mailing list</div><div><a href="mailto:tor-dev@lists.torproject.org">tor-dev@lists.torproject.org</a></div><div><a href="https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-dev">https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-dev</a></div></div></div></span>
<div style="border-bottom: 1px solid #f0f0f0; height: 10px;">
</div>
<br>
<div style="font-weight: bold; font-size: 14px; margin-bottom: 5px;">Attachments:</div>
<div>
<div style="">- userstats-bridge-transport-<OR>-obfs3-2014-08-01-2014-10-23.png</div>
</div>
</blockquote>
<div>
<br>
</div>