libeclipse
libeclipse on Hacker News
Posted by
libeclipse on Tuesday November 24, 2020 @12:00PM
from the coveryourtracks.eff.org dept.
Hong Kong is currently enjoying China's "rule of law"
You currently have a nationalist authoritarian government that is bent on curbing civil rights and is economically incompetent. It's not hard to imagine why China is doing better economically.
Users have to install a browser extension in order to participate in the study. That's a way higher barrier than the personality quizzes that Cambridge Analytica used.
It also happens at a different layer of abstraction. Cambridge Analytica extracted data through the permissions framework that Facebook itself implemented.
Facebook's interest in its users' data doesn't need further explanation after you see that most of their profits derive from their control over it. The same control that allowed the profitable mass political targeting that these researchers are trying to study.
I hope there's a mention of how Google locks out third party browser developers by refusing to let them implement widevine, their DRM solution.
> And they can't reduce their salary now because it'd be unfair on their families.
That's fucking bullshit. They make millions.
Posted by
libeclipse on Thursday September 03, 2020 @07:54AM
from the timroderick.com dept.
Peergos doesn't hide access patterns as far as I can tell.
Doesn't work for me with JavaScript enabled.
I like this prospect:
> Another unsolved problem where this approach shows promise is one of the most disturbing aspects of neural nets: No one really understands how they work. Training bits enter at one end and prediction bits emerge from the other, but what happens in between — the exact process that makes neural nets into such good guessers — remains a critical open question.
> Symbolic math, on the other hand, is decidedly less mysterious. “We know how math works,” said Charton. “By using specific math problems as a test to see where machines succeed and where they fail, we can learn how neural nets work.”
> The Facebook researchers compared their method to only a few of Mathematica’s functions —“integrate” for integrals and “DSolve” for differential equations — but Mathematica users can access hundreds of other solving tools.
> [...] it only included equations with one variable, and only those based on elementary functions. “It was a thin slice of possible expressions,”
> The neural net wasn’t tested on messier functions often used in physics and finance, like error functions or Bessel functions. (The Facebook group said it could be, in future versions, with very simple modifications.)
> Other critics have noted that the Facebook group’s neural net doesn’t really understand the math; it’s more of an exceptional guesser.
> Still, they agree that the new approach will prove useful.
Yep, one of the cool things about entropy is that it always increases. If you keep XORing more information into the seed pool, you only ever increase the total entropy.
Manjaro is a great frontend for Arch.
Maybe we should get rid of the NHS, and oxygen as well, as those things have the potential to benefit bad people.
Absolute idots. Cryptography benefits _everyone_, and this bullshit is at the expense of everyone.
With BTI can the attacker not just write a BTI opcode to the start of the jump location? Usually you would jump to a buffer that you control the contents of (i.e. shellcode)
Some employees literally doxed their colleagues. This has nothing to do with a "bubble" and more to do with being a terrible human being.
Dude if your clipboard is untrusted in your threat model then you have bigger problems. Any program that is reading your clipboard could also be logging your keys, watching your webcam, recording your screen, or exfiltrating your files.
Have a reasonable security model please
In regards to the accidental censorship of content for users behind VPNs and proxies, on my site I have set the security level to essentially off and disabled any and all integrity checking of visitors. I haven't tested the effect on those users but it should allow anyone to retrieve the pages unmolested.
There is a memory overhead associated with guarded heap allocations. The time overhead is only applicable when creating or destroying them.
For encrypted data there is virtually no memory overhead but a guarded allocated has to be created to hold the decrypted data when it is needed.
The root encryption key is stored within a special container [0, 1] that has some small but constant CPU overhead as it is continuously mutating the memory associated with the structure in the background. This overhead can be tweaked (or even virtually eliminated) by changing the interval between successive cycles.
[0] https://spacetime.dev/encrypting-secrets-in-memory
[1] https://spacetime.dev/memory-retention-attacks
Repositories are working for me but I can reproduce the issue with profiles.
I wonder if the performance gains are due to BoringSSL being faster than OpenSSL?
How about... asking them for it?
The new keys.openpgp.org service is a mitigation:
> keys.openpgp.org is a new experimental keyserver which is not part of the keyserver network and has some features which make it resistant to this sort of attack. It is not a drop-in replacement: it has some limitations (for instance, its search functionality is sharply constrained). However, once you make this change you will be able to run gpg --refresh-keys with confidence.
Do you have a link to back this up?
There are problems that arise in 4 + 1, such as how gravitational orbits aren't really stable and the force between the electrons and the nucleus is too strong to allow chemistry to exist as it does.