Rygian

Pike: strongly typed logic is great!

Perlin: stringly typed logic is great!

If each node has its own warmed-up VM awaiting from startup, there's no need to clone across nodes.

I'm looking for:

- a watch face to show the next seasons

- a watch face to upload/download data (using the photoresistor diode to receive, and the led and/or chirp-fx to send)

- how can I increase storage? I can't upload the astronomy face!

The "surprising way" is by using a derivate of LSD.

I'd argue that the surprise is rather on this: "In clinical trials, a single dose significantly outperformed standard treatments, offering hope to those who have found little relief elsewhere."

Actual punitive measures taken against entities who e.g. manipulate personal data in a negligent way. [1]

Which was much harder to achieve before.

[1] https://www.enforcementtracker.com/

The title reminds me of the single lesson I retained from a training for upcoming people managers: "You can delegate everything except accountability."

Surely that's a thermodynamic waste? Use the energy in the fuel to extract hydrogen from the fuel to them use the energy in the hydrogen to generate electricity to turn the wheels.

Is there a chemical process that achieves this with a better efficiency than just using the fuel to turn the wheels?

According to the Wikipedia article, it is still significantly worse than Just Using the Electricity (~21 to 48% efficiency just on the hydrogen production part, not counting distribution and consumption).

Check out the "Clean Hydrogen Ladder" document.

Hydrogen wastes a large amount of energy.

This sounds like the "stringly typed language" mockery of some languages. How is it actually different?

> Why would anyone care about something like this ...

Because it is a dangerous addiction [1] with recognised adverse effects on human health. Like sugar, tobacco, or drugs.

[1] https://news.ycombinator.com/item?id=46959832

The AltDrag tool on Windows includes Super+double click to maximize/restore. I find it surprising that this does not come by default on KDE.

Disclose the sale to the users of the thing being sold. Plain and simple.

Customers were sold[1] a lifetime subscription to Honest Guy's taxis, and then Honest Guy does a secret deed to sell his taxi joint to Bad Guy[2] without telling any customer about it. Then customers start getting ripped of in all manner of ways, that some of them would have known to avoid if they knew their taxis were being run by Bad Guy.

[1] Of course, the issue here is that no contracts were signed.

[2] In the specific case I was replying to, there was no malice or intent to hide from you as seller. Yet, a better outcome could have been achieved by advertising the sale to those impacted.

I don't think there is any legal support for what I describe above, but in principle whenever a user signs up for Good Thing, and then gets baitswitched to Evil Thing, the main victim is the user, and it is fair to hold responsible everyone involved in the bait-and-switch maneuver.

While assuming absolutely zero bad will on your part, I would nevertheless find it fair if you were legally on the hook for whatever happened after the sale, unless you could prove that you provided reasonable means for the users of your extension to perform their due diligence on the new owner of the extension.

This is of course easy to say in hindsight, and is absolutely a requirement that should be enforced by the extension appstore, not by individual contributors such as yourself.

By letting merchants receive payments from customers without going through Visa and Mastercard?

Granted, the FAQ entry is rather light in details:

https://support.wero-wallet.eu/hc/en-us/articles/39413057671...

It's end-of-life.

If it it still running out there, it's runningin zombie state.

The perspective that says "a whole compiler in just a few hours" is making false claims. So not a valid perspective.

I suffer from the same behavior, ever since I moved from Ubuntu to Debian.

An interactive system that does not interact (terminal not reactive, can't ssh in, screen does not refresh) is broken. I don't understand why this is not a kernel bug.

On my system, to add insult to injury, when the system does come back twenty minutes later, I get a "helpful" pop-up from the Linux Kernel saying "Memory Shortage Avoided". Which is just plain wrong. The pop-up should say "sorry, the kernel bricked your system for a solid twenty minutes for no good reason, please file a report".

I'll bite.

How was TikTok held liable for the crimes it was accused of?

> We have gotten better at grounding LLMs to specific sources and providing accurate citations

And how does the LLM know which specific sources to ground itself to?

How does the LLM know which sources can be trusted?

That would definitely work too.

What I like about my process is that there is only one source script (the python one) to keep around.

Since this process happens just a couple times per year, I find it reassuring to edit the file on the phone, check its contents, run it first in dry-run (the default), grep the output to my leasure, and only then run it in anger.

I take it as a personal preference to never delete a copy of a file I care about, until I have taken positive proof that it is copied safely where it belongs.

I have deep trust issues with commands that work on more than one file at a time.

My usecase for Termux: most pictures get backed up properly into the home NAS, but the sync process sometimes skips a few [1] (which is exasperating, but here we are still not migrated to Immich).

So I have a python script in the NAS that calculates the MD5 checksum of every photo and video, and generates a shell script that, when executed on the phone, will calculate the MD5 on the local device, and delete if it is equal to the NAS.

The generated shell script gets sent to the phone, then I execute it from within a Termux window, pointing at the DCIM folder.

I can free up tens of GB of memories with reliability in the face of a misbehaving sync algorithm.

[1] https://help.nextcloud.com/t/auto-upload-is-skipping-random-...

stdout and stderr get different colors.

You naively assumed that they would actualy disbourse 2 G€ in payments for those fines.

Reality disagrees: https://www.irishtimes.com/business/2024/12/13/data-regulato...

> Engineers in particular have expressed their desire to focus their time on engineering, without being hampered by slow process flows [1]

I wonder what correlation will exist between the set of people who end up leaving the company, and the set of people responsible for setting up those "slow process flows" in the first place.

[1] https://www.asml.com/en/news/press-releases/2026/strengtheni...

"applying regen braking in a car is the same as applying brake pads"

That's an assumption I disagree with. Brake pads will always be less smooth than engine braking. For the same braking action, I assume more brake dust and slightly higher tire wear due to brakes not able to provide fine speed adjustment.

The down-the-hill scenario is interesting, it brings new comparisons: is there more tire wear from maintaining a chosen speed, vs letting the car overspeed and then braking? How does air resistance contribute in each case?

I maintain my earlier opinion that the differences between all these scenarios are minimal and can be ignored. But if you have some physical model that helps calculate these scenarios, it could be fun to play around with.