QmS3Yg1BJM5Y6c97VjqLE2Ua9DsMfMrPPnxHcuLR3ZpeBs very_important
TOR PROJECT CHOOSES PERFORMANCE OVER SECURITY
It's no such thing.
Nothing about these changes compromises the security or privacy of users. Do you even understand the changes that were made? If not, you have no basis for making the claims you have. If you do, please explain, with details from the code, how these changes decrease the security and privacy of users.
Or fuck off, LARPer.
I was talking about browser/device fingerprinting. At bare minimum if I were to use another browser I would have to have javascript disabled 24/7 yet I still could be fingerprinted to specific browser, on specific os, connecting from specific anonymity network at certain times. And that would be a sample of max few thousand users in best case scenario. Now if I install browser plugins or use a browser that is not common, or visit sites/content that aren't common I have completely unique fingerprint.
Also browsers like firefox by default send your browsing history to OCSP servers and unique hash of downloads to google. It's a game of whack-a-mole with you never knowing how firefox will try to fuck you over next time.
Found the rust shill
XPde is a good lisp program...
but 1/2 the functionality is broken on 64 bit systems... somehow
Reminds me of the time, for years, a game engine would load the same assets on connect and mapload. 2x the time for no reason. Was eventually fixed.
I do think not verifying the hard coded DH params at boot could be a security risk, what if spooks manage to change them to weaken encryption and there's nothing checking that?
Not using 1GB keys is "trading performance over security".
You need to make a cutoff somewhere.
Learn to read nigger, absence of proof is not proof of absence.
At that point they could also hex edit the validation code to always return true.
Then your fucked, kiddo. At that point they could modify arbitrary code, and wouldn't even bother messing around with tor.