I don't think DPI is the answer here: o In the case of Turkey, they had more than 10k direct users, the DPI-based censorship yielded a 50k spike, that's a 1:5

I don't think DPI is the answer here: o In the case of Turkey, they had more than 10k direct users, the DPI-based censorship yielded a 50k spike, that's a 1:5 ratio, whereas in the case of the UAE it's a freaking 1:30 ratio and it doesn't stop from increasing. o In the case of Turkey, the spike was followed by a spike in bridge use, most using the obfs4 pluggable transport. There's however no such apparent spike for UAE, in fact it's only a nearly constant 300 obfs4 users https://metrics.torproject.org/userstats-bridge-combined.html?start=2017-01-13&end=2017-01-22&country=ae https://metrics.torproject.org/userstats-bridge-combined.htm...