This post indicates remote in Canada but the linked postings say United States under "Work Eligibility", can you please clarify this requirement?

Invariably seems like a strong word for the output of an LLM. If we can prompt a language model to double-check it's work (and that means it always gets it correct), why is that not part of the base prompt?

It’s less double check and more a request to go check sources online, which it does, provides links to those sources and then summarizes.

This is pretty much exactly what I have been looking for, I do this with paper/pencil on our fridge.

If you don't open-source it, where is the data stored?

Right now just unmanaged postgres instance hosted on fly.io

Agreed. I was quite happy with my Pixel 6 until the 7 was released, which I got for my wife. Immediately noticed that both phones cameras were worse.

In particular, shots of light-eyed people show as fine initially and then when the software processing finishes they have almost black eyes from all the contrast they add. If you choose raw you need a whole other processing pipeline, not to mention you get to see how shitty these sensors and lenses are without software doing a herculean lift.

I was able to work around it on my 6 with a GCam APK but couldn't find one for the 7.

Thank you for this walk down memory lane—2advanced was a huge inspiration for me when I started web design and learning flash, but I couldn't remember the name of the site until now. Their 2003 design is burned into my memory.

FYI this article was published on April 1 and is labelled as "April Fool's".

hhhahaahha woa I missed that completely

I'm pretty sure i'm not the only one who took it seriously. Thanks! I wonder if mr LeCun made the same mistake, it's just... almost believable! Good April fools joke I gotta admit.

EDIT: it's now 5 minutes later and I'm still laughing. I believed for months that Yudkowski had started a death cult.

It wasn't entirely in jest. See this article, published a short while later: https://www.lesswrong.com/posts/uMQ3cqWDPHhjtiesc/agi-ruin-a...

And more recently: https://time.com/6266923/ai-eliezer-yudkowsky-open-letter-no...

Sure, but the thing about starting a suicide cult was.

Love deep dive troubleshooting like this. I haven't heard of systemtap before; looks nice. When I had to troubleshoot a kernel bug [1] I used perf [2] probes which are also really nice for this kind of debugging.

[1] https://www.spinics.net/lists/xdp-newbies/msg01231.html

[2] https://www.brendangregg.com/perf.html

This is a welcome bit of perspective, so thank you for that.

I often find myself complaining about things at work like competing priorities or wearing too many hats and it helps me to ground myself by thinking about past jobs at minimum wage or even watching what people making less than a fifth of my salary have to put up with on a daily basis.

This same site has recommendations on pencils. If you're looking for mechanical pencils I'm a huge fan of the Uni Kuru Toga Roulette, which came in second [1]. I like it because it rotates the lead as you write. Some people find the little bit of squish it has off-putting but I can't use other pencils without snapping the lead constantly.

1: https://www.jetpens.com/blog/The-Best-Mechanical-Pencils/pt/...

This isn't necessarily a bad thing—studies[1] have shown that adoption discovery between birth and two years of age may increase life satisfaction and decrease stress.

The same applies to other cases where parents aren't genetically related to their children (i.e., donor egg and/or sperm).

[1] https://www.theatlantic.com/family/archive/2019/07/adoption-...