netheril96

> If your recovery mechanism completely bypasses the benefits of using the YubiKey, that recovery mechanism is what your attackers will use.

I don't use any methods other than YubiKey when I do have my keys. So attackers cannot trick me. If I see a choice without Yubikeys I immediately know I'm being phished.

I have four. One attached to the desktop, one attached to the laptop, and two carried with all my mechanical keys (one of them supports NFC and another supports USB-C).

I think the point of header generation is not to save compilation time, but to save developer time. Saving them from writing the same thing twice, in header and in implementation.

My guess is that it needs to be very pure to have zero resistance. But levitation requires less purity.

You can frame it the different way: Apple has given up improving the freedom of the Chinese people because that is infeasible. But it hasn’t given up hope on the West.

Your interests are completely opposite to mine. I welcome this pasteboard change. In fact, I updated my iPhone right away just to have this feature.

They don't mandate every keys to every door because they CAN break almost every door with force. The police can do it with a warrant and some "special forces" can do it clandestinely.

They can't break most encryption with force.

Your information is outdated. Kotlin is now a supported backend language at Google.

Every ML engineer will be impacted. Web server probably less so.

GIL has been a major pain point when training machine learning models, where every library is Python first. The data preprocessing phase must happen quickly enough and in parallel with respect to the GPU operations. The simplest option is doing data processing in several threads other than the one driving GPU operations, but that is not possible with GIL. TensorFlow and PyTorch each has their workarounds for this problem, but many difficulties remain. Removing the GIL will be greatly beneficial to this domain at least.

So is "Threads" in my opinion. But I downloaded Threads regardless.

> limited in scope.

These three words evoke PTSD in me as hearing it means promotion is no-go again.

AMD's problems with AI lie first in software rather than hardware.

Do you use Android phone? Since iOS doesn't allow phone call recording without jailbreak.

> The average user isn’t using Firefox to begin with. In fact 5% market share for Mozilla is probably pretty generous.

Is Mozilla content with that market share? If not, it must act with the assumption that its users are average.

Purely from self interest, I hope this wins over Twitter. So I installed it and made an account.

When Elon Musk laid off more than 80% of Twitter's workforce (or maybe 90%, the number is quickly increasing), many people said this would prove that a company needed only a handle of software engineers to be successful. If they are correct, it paints a bleak future where software engineer jobs will be forever cut down. I hope Twitter falls so it proves the reverse: a software company cannot survive without enough engineers.

While Facebook also laid people off, the percentage is much lower than Twitter. It will be even better if Facebook has to recruit people to win this battle.

I don't like the name "Threads" either. Probably because I'm a programmer.

I trained a WGAN on torch-directml with no issues so the software seems quite supported. But I can’t speak of performance because I have nothing to compare against.

It works for me. I have no issues training a WGAN with it. But I don’t know how much slower it is compared to CUDA on a similar priced NVIDIA card.

Just in case other people who have AMD GPU and run Windows have the same needs as I have, that is, to train or run machine learning models, please checkout torch-directml and tensorflow-directml.

You misunderstood your parent comment. What he/she meant is that the "algorithm" is only a guess from reverse engineering. The actual algorithm deployed at GFW can look significantly different.

Indeed a whole class of GFW bypassing tools are now based on masquerading as HTTPS. Trojan (TCP only), Vision (TCP only), Hysteria (UDP), just for some examples.

You were told wrong. If you uses SSH as a proxy, the connection will be slowed down to a crawl very soon. GFW distinguishes this from SSH command typing by looking at the traffic. This has been in place for at least a decade.

As an Asian, all I care about is that we should not atone for the crimes committed by someone else (e.g. the white slave owners). That AA disadvantages Asians the most is quite contrary to the whole "atonement" argument of AA.

Ethereum, the chain this bot runs on, no longer has any connection to GPUs.

In China most payments cannot be reversed. By contrast, the security of payment methods are much more stringent in China.

I was surprised when I first learnt that you could spend money online with a VISA card by only card number, expiration date and a 3(!) digit CVV. Those are hardly secure and printed out on the card that anyone can see. Then the US merchants have so many rules about fraud detection but all of them are just heuristics with many false positives and false negatives.

When Internet companies constantly invents new methods to both increase security and usability (two step authentication, pass keys, FIDO-U2F), the payment industry in the US is just stuck in the past.

Your reasoning is flawed. iPhone 14 fits the same use case as iPhone 11, but Foldable fits a different use case than normal phones. You may find Foldable useless but that doesn’t make your reasoning any less flawed.

The Fed interest rate is well above 4%. No doubt some banks are offering 4%.

The most wrong part of that previous team is to store private keys unencrypted in the cloud, not the performance part.

> Physical access will always be king

No. Your wording suggests that once attackers gain physical access, all is lost. It is not true. With a passphrase based full disk encryption, if the passphrase is strong and the machine is powered off, physical access doesn't imply data access.