> 1. NEVER EVER login from an email link. EVER. There are enough legit and phishing emails asking you to do this that it's basically impossible to tell one from the other. The only way to win is to not try.
Sites choosing to replace password login with initiating the login process and then clicking a "magic link" in your email client is awful for developing good habits here, or for giving good general advice.
:c
In that case it's the same as a reset-password flow.
In both cases it's good advice not to click the link unless you initiated the request. But with the auth token in the link, you don't need to login again, so the advice is still the same: don't login from a link in your email; clicking links is ok.
Clicking links from an email is still a bad idea in general because of at least two reasons:
1. If a target website (say important.com) sends poorly-configured CORS headers and has poorly configured cookies (I think), a 3rd-party website is able to send requests to important.com with the cookies of the user, if they're logged in there. This depends on important.com having done something wrong, but the result is as powerful as getting a password from the user. (This is called cross-site request forgery, CSRF.)
2. They might have a browser zero-day and get code execution access to your machine.
If you initiated the process that sent that email and the timing matches, and there's no other way than opening the link, that's that. But clicking links in emails is overall risky.
1 is true, but this applies to all websites you visit (and their ads, supply chain, etc). Drawing a security boundary here means never executing attacker-controlled Javascript. Good luck!
2 is also true. But also, a zero day like that is a massive deal. That's the kind of exploit you can probably sell to some 3 letter agency for a bag. Worry about this if you're an extremely high-value target, the rest of us can sleep easy.
If you interact with internet comments and discussions as an amorphous blob of people you'll see a constant trickle of the view that models now are useful, and before were useless.
If you pay attention to who says it, you'll find that people have different personal thresholds for finding llms useful, not that any given person like steveklabnik above keeps flip-flopping on their view.
I wrote that patch. It's not actually used for MI50/MI60 in any of the Debian system packages, since Debian builds for gfx906 rather than using the gfx900 fallback path that patch provides. Debian is not relying on any special patches to enhance gfx906 support. That architecture is the same as upstream.
Now, for some other GPU architectures, you're absolutely right. There are indeed important patches in Debian that enable its extra-wide hardware compatibility.
"brand safety" is far removed from what they care about and would likely exist without that sphere's input given corporate trust & safety teams' history.
It has certainly negatively polarized people against any form of safety near AI.
This reminds me of when I played around with laser diodes as a kid (of course to burn random things).
I ordered a powerful green laser diode from eBay, wired it up and pointed it at some black paper, hit record on the camera, excitedly connected the battery and... nothing. I checked the wiring, all was good. I looked in the end of the diode, I could see a faint red glow inside, but nothing else. I must've got a dud unit.
Later I looked at the recording and my heart sank. When past me connected the battery, the room immediately lit up with a bright white glow. The diode emmitted an intense beam of infrared light... and I pointed that thing directly at my eye.
There wasn't a wider point to that, this just reminded me and I wanted to share. I suppose be careful of what you can't see.
I got lucky. That sort of thing can cause big problems. Especially those that stay unnoticed until old age. Hearing is also one of those.
Can the laser in OP’s laser burn things? No? Point of my comment is to console OP’s concern with long term damage to his eyes. Some Windows Hello’s IR emitter is powerful enough I can see bright red blinking.
I have a friend who built one of these as their thesis. IIRC they were telling me that the air through which the sound beam propagates acts as a low pass filter, so if you're at the correct distance from the device the high frequency energy should have dissipated.
Interesting stuff, I wish I had more time to learn about what they where doing.
Not sure if there are standing waves involved, or resonance. I presume it is very similar to a phased array [1] for beamforming in antennas, except that then anisotropic properties of medium may not be negligible to construct the wavefront (temperature gradients & wind), which is probably also why these devices do not sound great. To produce a high quality waveform at the receiving end the physics probably becomes quite involved rather quickly.
No, we don’t hear ultrasonic frequencies because our ears do not resonate at those frequencies.
We hear sounds when the cilia in the cochlea resonate with the incoming sound. We don’t have cilia of the length required to resonate with ultrasonic sounds, so there’s no danger of hearing loss.
Animals may get their hearing damaged, if they are in the path of the sound, are close enough that it’s still ultrasonic at their location, and are sensitive to the frequency used, I believe. Maybe someone who knows for sure can say for sure.
I'm not an expert on this, but there don't seem to be any reported cases of hearing loss from sounds above 30 kHz, but there are documented cases of unpleasant effects. In any case, I'd keep some distance, just to be safe.
People want them not to be advertised at official Nix events, not to bar them from using it. That wauld be impossible due to licensing so isn't on the table.
Would you require the FSF to accept a sponsorship from anyone and to advertise them in return?
conan makes it very hard to build a project offline and with a fixed set of inputs, it makes things worse in important ways if you care about reproducibility.
`device_grouped_conv2d_fwd_xdl_ngchw_gkcyx_ngkhw_f16_instance`, what are you doing to our poor friend clang?