tehnoslow

I know this situation so well... What helped me was to first start doing household chores, at least a little at a time (once you start, it becomes easier to continue). Then start going out into the fresh air for at least an hour a day and allow myself not to think about what scares me so much. And then at some point just grit your teeth and stop putting it off. Just get it done, despite your fears, and don’t let yourself dwell on it any further.

I like this idea

I couldn't do it. I’m not sure if this is even possible.

I like this concept

I’m not a fan of the chronological approach, but I like this project. It’s beautifully done, I found it interesting to explore

Thanks, now I know what I'm going to read next

I don't think that's the best idea for starting a business. Demand is too low, and the required investment is too high.

I am inclined to agree with the view that AI is merely a tool. And, incidentally, using it rationally isn't always easy.

I don't yet see open models exerting significant pressure on the closed API market. Access to the most significant models is still quite expensive, and demand for them appears to remain high.

I didn't know the ceiling fan market was in such bad shape

Damn, that’s heartbreaking. Rest in peace.

It seems to me that it's like this everywhere now

The idea is interesting, of course, but I think the user experience won't be much different from a classic dating app, plus it inevitably requires an analysis of many nuances

I have to agree

Fun game, but I agree with senectus1, it's a bit clunky on desktop

It’s not yet clear how all this will be implemented. But I have plenty of doubts.

Actually, yes, that would be at least interesting

"Nicotine is just caffeine" is too clean given the tumor-promotion evidence. To a large extent, it comes down to the method of delivery: patches are difficult to get hooked on because they do not cause a rapid spike in nicotine levels and are not associated with any ritual; cigarettes are hard to give up due to that spike, the accompanying ritual, and the MAO inhibitors found in the smoke; while vapes occupy a middle ground—which is precisely why they help some people quit smoking but foster addiction in others.

A good project. It’s great that the tasks are generated deterministically, without using AI-generated answers

A fair criticism of the article is that it lists SAML's vulnerabilities but doesn't do the same comparison for OIDC. OIDC has its own problems too: JWT algorithm confusion, none algorithm attacks, missing audience checks, and bugs in JOSE libraries

Thanks for Axiom, looks useful. Does it already cover the new 27.1 beta APIs?

The claim that the IAU vote was procedurally invalid is interesting

The real-time stats on the website are fascinating. Did you set those up to satisfy your own curiosity, or to give remote callers a sense of "presence" at the playa?

That's some crazy stuff. I knew LG and others liked to push ads on their smart TVs, but I didn't realize it was this bad...

Interesting breakdown. The author clearly put in the work to test this theory. I like that he's not trying to force the data to fit the result but instead explains why even apparent matches could just be coincidental.

Cool idea, but it feels more suited for pedestrians or cyclists moving at a slower pace. Not sure this format would work well in a car, since you can't really see the whole road ahead. At 100 km/h, you need to grasp the route and upcoming turns quickly and this view might not give you enough time to get in the right lane. Still, props to them for trying something new.

Cool use case with the agents, hadn't thought of that

Worrying news, but honestly not surprising anymore. What's interesting is that El Niño hasn't even peaked yet, if it gets as strong as they're predicting 2027 is gonna be even hotter.

Also kinda depressing to realize that no matter how much one person does, it won't make a dent globally. And what really stings is that even when decisions get made they're either too late or half-assed. Guess all we can do is hope we don't live to see things get really bad...

Honestly, this is the most practical take on prompt engineering I've come across lately. Makes me wonder why spend all that time crafting a 500-word essay to tell Claude to stop being Claude, when the real move is just to chain a couple models together and call it a day

The more I think about it the more autonomous cars look like extremely demanding edge computers. You have limited power and cooling, unreliable connectivity and very strict latency requirements. It's probably one of the few places where designing the hardware around the workload makes more sense than trying to make a general purpose platform do everything