paulgb

I don’t think that’s what’s implied here. Possibly they are trying to get API tokens of providers to distill their own models.

That's a great example because it also acts as a mnemonic to help people remember the content, rather than distracting from it.

I wish it were that simple.

    ~: sudo chmod -x /Applications/Music.app       
    chmod: /Applications/Music.app: No such file or directory
    ~: sudo chmod -x /System/Applications/Music.app
    chmod: Unable to change file mode on /System/Applications/Music.app: Operation not permitted

(Mac OS Tahoe 26.5)

I remember following this at the time! FYI it was "shut down" (kept online but no further development) two years ago (https://x.com/KennethCassel/status/1620500575183073280)

> Decided to shutdown Diode. (For now the site is live, I'll likely open-source the code in case anyone wants to take the baton and run)

Kenneth is now working on RMFG (https://www.rmfg.com/)

I like betting because I find it fun to quantify probabilities of future events, and having a ~small amount of money at stake keeps me motivated to follow through on whether I was right.

I do think there are positive EV bets to be had on prediction markets though, they are mosty not efficient.

> an unrealistic maladaptive cope

I like betting myself, but I don't think abstaining from betting is maladaptive at all. Most of the bets we encounter in the real world are negative EV.

I'm mostly surprised that someone can so consistently and repeatedly demonstrate an inability to filter information he receives and still be trusted with LPs' money. It's another form of Gell-Mann amnesia.

I would bet in the direction of this being a bug on big corp's side rather than Cloudflare's.

Enjoyable read, but kind of ironic that it interrupts your reading half way down to nag you to subscribe. Everything is an overcrowded airport lounge, indeed.

Cars are only “heavily regulated” in the sense that you pass a test once when you are a teenager and then never have to pass a test again, just pay a nominal fee to renew your license.

I am curious what data you are looking at that gives you the impression pedestrians and bikers are the root cause of most accidents. As a frequent pedestrian / biker here, I see a car doing something unhinged about every mile I walk. On Wednesday I almost got hit by a car flying the wrong way down a one-way street and then running a red.

This is by Steve Ruiz of https://tldraw.com, in case anyone noticed the similarity between the two.

This is great! Thanks for writing it.

One nit, the unquoted quotes in this file seem to be a parse error (I replaced the inner ones with single quotes and it ran) https://github.com/anordin95/a-conceptual-overview-of-asynci...

I had a trip with my family and used v0 to create an itinerary app with a timeline view of our flights, hotel/airbnb bookings, activities, etc.

It was the only thing I’ve 100% vibe-coded without writing a line of code myself. It worked pretty well. In an earlier era I might have used a shared google doc but this was definitely a better experience.

If you’re looking for things to use lovable/bolt for, I’d say don’t use it for software you otherwise would have written by hand; use it for the software you would never have written at all.

I've said it before but really feels like a flaw that the halvings are discrete and happen suddenly every four years, instead of gradually each block. As far as I can tell the only advantage to it is that it makes the math simpler. The disadvantage is that it creates weird market dynamics in which large amounts of mining capacity are plunged into unprofitability in one instant. If I wanted to run a 51% attack, I'd look to buy up suddenly-unprofitable capacity immediately after a halving.

That's a dated rule at this point, ChatGPT has been able to use its Python interpreter as a calculator for a while and in my experience will opt to do that for back-of-the-envelope calculations.

I think the intent is less about the CO2 emissions as about the air quality that people have to breathe (hence a stricter standard in some locations).

I don’t know about measurable effects but I hate when I pass a long-idling truck and can taste it in the air.

I see where you’re coming from, but the alternatives are either that the law isn’t enforced, or the state ramps up its own surveillance, which is more dystopian to me.

I see this as in the same vein as SEC whistleblower awards, which I’ve never heard described as dystopian. Businesses just don’t have the same expectation of privacy that individuals do.

It’s 75% lower off-peak though, so there’s still an incentive.

> Whether someone is driving because of speed, or comfort, or some other factor, the cost has to exceed their personally calculated benefit.

It's a dynamic system though; as some drivers opt not to drive, the utility of driving for those other drivers increases. Yes, the market will find an equilibrium somewhere where some people will still drive, but that's kind of the point.

> That's the fundamental argument against the assertion that traffic speed increases will offset the costs. It cannot be true, or people would choose to drive.

I think the mistake you're making here is assuming that the value of driving and the cost of congestion are the same to every driver.

For some people, driving is an elastic decision. They mode shift, or time shift to off-peak, or carpool, or combine errands in the city into one trip instead of multiple.

For other people, driving is necessary. They'll benefit from fewer of the first type of person being on the roads during peak hours.

It was not just an idle threat either, they tried to do it until they were blocked by a judge https://www.nytimes.com/2025/05/27/nyregion/nyc-congestion-p...

> Literally everything we buy and use in the city gets more expensive because of this law.

Congestion was already priced into all goods and services in NYC, it just came in the form of a deadweight loss (paying delivery workers / tradespeople / professionals to sit in traffic) instead of a tax that at least ostensibly will fund better transit.

My experience is that the best LLMs can produce syntactically valid strudel but don't understand the actual semantics of the "mini notation" language, unfortunately.

> Their determination? I wasn't entitled to a $3 refund.

Frustratingly, Lyft’s position on this is that if you don’t like the car that arrives you should reject it when it arrives, otherwise you’re not entitled to a (even partial) refund, even when they know on their end that the car they sent doesn’t match what you paid extra for.

Part of what makes it seem shady here is that airline ticket prices are pretty opaque. If they advertised it as a group discount, it would be received differently.

I think this is fair play, they can charge how they want (within reason) and it’s not too different than other bulk discounts.

But someone should totally make a site for finding strangers to book the same flight with :)

Largely, yes -- one of the big issues with using other people's random MCP servers is that they are run by default as a system process, even if they only need to speak over an API. Remote MCP mitigates this by not running any untrusted code locally.

What it _doesn't_ seem to yet mitigate is prompt injection attacks, where a tool call description of one tool convinces the model to do something it shouldn't (like send sensitive data to a server owned by the attacker.) I think these concerns are a little bit overblown though; things like pypi and the Chrome Extension store scare me more and it doesn't stop them from mostly working.