animalCrax0rz

5 ms·

1.
by animalCrax0rz 6y ago
Given Amazon's willingness to spy on their business customers (traffic data in this case), should I be worried to deploy code on AWS that has high IP value in source code form (JS, Python, etc) or bytecode (Java/C#, which can be e
2.
Scientists uncover principles of universal self-assembly(phys.org)
8 points by animalCrax0rz 6y ago|0 comments
3.
WHO warns that few have developed antibodies to Covid-19(theguardian.com)
2 points by animalCrax0rz 6y ago|0 comments
4.
by animalCrax0rz 6y ago
Totally. The general problem is indeed difficult to solve, but I've seen lots of great work in this area.
5.
by animalCrax0rz 6y ago
In addition, the wording "test positive for Covid antibodies" may lead people to think that 30% of the population in MA is immune to Covid19. It obfuscates the fact that the mere presence of antibodies does not imply immunity. The
6.
by animalCrax0rz 6y ago
Nope, I was simply countering that it's not as simple as they suggested. I added the link to the paper to show one approach to given them some context as to what might it look like.
7.
by animalCrax0rz 6y ago
Google is aware of this problem in their search approach. It's a business problem, not a technical one. You're saying the same thing in suggesting they base their decisions on some measurement set. If solving the problem adds comp
8.
by animalCrax0rz 6y ago
Google is in the business of producing quick results to sell ads while keeping the cost low. If anyone does better, they will likely do it using a costlier algorithm, and if there is more margin in reselling tangible products than ads then
9.
by animalCrax0rz 6y ago
I keep trying to talk about the general problem and people keep focusing on this one example. I give up. :)
10.
by animalCrax0rz 6y ago
Yes, I was just countering that it’s more involved than recognizing the importance of “no” in a sentence.
11.
by animalCrax0rz 6y ago
Which part of "The paper I linked to below describes a technique they call "deep parsing." Check it out for more context." could you not parse? I mean if us humans have a difficulty parsing each other's statements,
12.
by animalCrax0rz 6y ago
"there is no evidence of cancer" and "there is evidence of no cancer" are two different statements with different meaning, so it's more complex a task than just understanding the importance of "no" in a se
13.
by animalCrax0rz 6y ago
There is this, too, from 4 years ago, which seems reasonable in my not-very-well-informed opinion (speaking of which, I'm not sure the work referenced here can deal with hyphenated negation, but it should be simple to include) https:&
14.
by animalCrax0rz 6y ago
A recent study says that 6 new types of coronavirus were recently found in bats https://www.zmescience.com/science/six-new-types-of-coronavi...
15.
by animalCrax0rz 6y ago
It’s time to go deeper? When faced with an existential crisis, we tend to jump into fix it mode. But fix what? The symptoms or ourselves? Let’s start with ourselves: we have failed miserably at being stewards to this earth. We have failed (
16.
by animalCrax0rz 6y ago
That's mostly unintended FUD (but still FUD) or problems that can be worked around: << - Is unstable because if a dependency/host goes offline, all users are broken instead of an internal build being broken (think leftpad bu
17.
by animalCrax0rz 6y ago
It's not about choosing a different registry for all your modules, which will almost always be a commercial for-profit entity due to the cost of running and maintaining a centralized repository of that scale. AFAIK, or as far as I can
18.
by animalCrax0rz 6y ago
I think there's some confusion here. NPM is both a massive repository as well as a package manager. Deno will (soon?) have a package manager, but it won't be tied to one central repository run by a private company, now part of a m
19.
by animalCrax0rz 6y ago
This brought up in my mind the thought that while Deno is still WIP (for example, packaging of Rust plugins is not yet resolved) and the ecosystem around it barely exists it was designed to have no dependency on 3rd party tools like npm and
20.
by animalCrax0rz 6y ago
HN wouldn't let me nest another response. This is in regards to "Timers aren't necessarily even a requirement" <<you can busyloop and count iterations of that instead as a “timer”>> Assuming you're the o
21.
by animalCrax0rz 6y ago
the minimum essential behavior to implement a feature is one that takes into consideration keeping the user safe from attacks... you could call that bloat, but I wouldn't be sarcastic about it: if you can make the mitigations more conc
22.
by animalCrax0rz 6y ago
performance.now resolution in Chrome is between 1 and 2 ms, I believe, with jitter. If they have a working POC for Chrome why not demonstrate the full exploit and force the Chromium team to rethink their mitigations? Lots of people talking
23.
by animalCrax0rz 6y ago
"We were not able use these techniques in Firefox, as they recently reduced the timer resolution to 2ms. The techinque presented here is potentially relevant even for this timer resolution, but some parameter tweaking is required.?&quo
24.
by animalCrax0rz 6y ago
Cool. I wasn't aware Chrome re-enabled SABs, with mitigations. https://github.com/tc39/ecma262/issues/1435
25.
by animalCrax0rz 6y ago
postMessage cannot provide a reliable timing signal since it goes on the task queue on the receiving end (in the main thread) along with other pending events, and even if there were no other events, there is latency noise in postMessage due
26.
by animalCrax0rz 6y ago
Again, JS can no longer perform this exploit. Browser vendors have disabled (made inoperable) high resolution timing. It's now at 1ms resolution. Not enough timing resolution to mount the attack.
27.
by animalCrax0rz 6y ago
That depends on high resolution times which have been disabled in all browsers. AFAIK.