Troll estate sales. I bought one last year for $10. It's a spare, as I have a half dozen, including one with an integrated mouse stick, a narrow version without a keypad, and the monster honkin' version meant to be used with the weirdo PC-based 3270 client. I think it weighs 10 pounds.

What does this have to do with "disposable computers?" For context, I'm still using a slim iMac to run some old Firewire audio gear. It works fine.

The windows equivalent would be Microsoft telling people to stop using some janky encryption overlay on top of FAT32, and switch to native NTFS encryption.

github burn /dev/cdrom0 https://github.com/Octocontrab...

I can go over to my friend's house for free. Our neighbors come over to our patio for free, too. Sometimes we make pies. If you want to get really technical, though:

Nominal cost for a nice pie on the patio:
Tub of raspberries on sale (in season) at Costco: $6
Sugar, flour, salt: I'd estimate $1, but I'll be generous and say $2
Coffee - 2lbs bag from Costco usually costs around $16 and makes about 15 big pots of coffee, so $1
So that's, rounding up, $10 to entertain about 8 people, or about $1.25 a person. I'd say that's doable for most people, it just takes some time.

A good point, but AI-written.

There are numerous interventions that have been shown to do just that, even non-invasively. GPL-1 agonists and psilocybin are two examples. The opposite can happen as well: sudden addictions in frontotemporal dementia, for example. Addiction is a well-understood mechanism modulated by the nucleus accumbens, which is what they targeted. An analogy is treatments for Parkinson's disease or essential tremor: you walk in with a tremor, right after the procedure you have no tremor. You're right that physical withdrawal will happen, and it's very uncomfortable, but that will go away after a few days, or it can be softened with methadone.

You're correct about late-stage Alzheimer's disease (AD), where there is widespread cortical and hippocampal degeneration - those areas are not coming back. However, early-stage AD can still benefit from these therapies.

Parkinson's disease (PD), though, is actually the first disease deep-brain stimulation was developed for, in the 80s. That's because PD is a degenerative disease of only a specific brain region (the substantia nigra pars compacta) and the downstream effects are a network dysfunction. Motor symptoms are treated with exactly these kinds of therapies - deep brain stimulation (DBS) of the subthalamic nucleus and globus pallidus internus, as well as high-intensity focused ultrasound (HiFu) ablation of those same areas. Same for essential tremor and dystonia (different targets). None of those therapies treat non-motor symptoms, and they don't even treat all symptoms of PD equally, but they are very effective.

DBS for OCD has been FDA-approved for over a decade now, and there are many trials going on for DBS of almost every psychiatric disorder you can think of, starting with the biggies, major depressive disorder and generalized anxiety disorder. Many of those target the nucleus accumbens, like this therapy.

This technology (MRgLIFU) is less than a decade old and is being tried for all of the indications that DBS and HiFu have been used for, but so far it's all trials, none of the therapies have reached FDA-approval stage. It's a very promising technology, it's very real and very much applies to movement disorders like PD as well as psychiatric conditions, and even early-stage AD. However you're right that we're probably not bringing back the memories of people with late-stage AD.

Engineering car safety is notoriously hard. Almost every design choice is a compromise of one thing or another

Crumple zones make cars a lot safer without adding significant weight and reducing fuel economy. It means, though, that even minor collisions can total a vehicle, or cost multiple thousands of dollars to fix.

The most dangerous accident for a car is a rollover. One of the reasons is the roof can cave in and hit your head, injuring or killing you. One way to fix this is to reinforce the roof, which adds weight high on the chassis, and makes rollover accidents more likely, increasing the likelihood of other injuries from rollovers.

Lastly, you have legal pressures complicating everything. If you go ahead and reinforce the roof, you'll be sued for increasing the likelihood of rollovers. If you reduce reinforcement and the roof caves in, you'll be sued for that.

It still is, for successful companies, and when it's not, you go work for Sony or Dreamworks :)

This is the underlying problem. The CEOs don't trust their own company. When the studio was churning out hits in the 1990s, the should have thrown everything they had at the studio team. Instead they cut the budgets and strangled wages, so half the people went to work for Dreamworks and Sony. Same thing happened with Pixar. Brad Bird (The Iron Giant, The Incredibles, Ratatouille, etc...) wanted to direct a live-action movie. Disney eventually nixed the project, so he left for a few years.

If you trust the people working for you, you pay them well and fund their projects. If you don't trust them, you keep buying other companies hoping to fill in the creative void.

This is factory zero, a low volume plant that makes a few electric trucks and SUVs. Even with the layoffs, it has around 2,500 employees, which is a lot compared to some other plants. The Orion general assembly plant only has a thousand, as does Bowling Green.

Did it on my home LAN. It's not too hard on a Synology router. The only thing allowed through is a sprinkler controller's web interface facing inwards. The Synology also has a halfway decent IDS that looks for weird outgoing traffic.

charging works fine.

The restrictions are about the same as NYC and for the same reasons. DC is covered by the flight paths for two extremely busy airports, along with all the random military aircraft flying around DC in particular. Our main airport is next to a small suburban town, and you can't fly a drone there, either.