8xeh

> We realize that people don’t need to come directly to Stack Overflow anymore in order to get answers to their programming questions—the chatbots have that covered.

"LLMs make StackOverflow pointless," is quite a position to take, especially coming from SO. I guess they need to do something to bring people back since they don't make money from LLMs (probably quite the opposite).

I found the old dev story to be a nice place to host a CV. I much prefer an industry-specific site like SO to generic mass-market sites like indeed.

I'm not sure if the new site will be relevant given the above admission that humans don't use StackOverflow. I still want to like SO though, so maybe I'll give it a try.

My Mac mini m1 is a fantastic home server.

Years ago I realized running a 100W PC all the time was REALLY expensive, so I switched to an old linux laptop at 30W (about $4/month). That helped, but I moved to the m1 mac specifically for the power efficiency. It does all the things my linux server did, and it does them at 6W (75¢/month).

I've found no competitor with similar performance that can run in such a low power footprint. M4 mini's are way better at power/performance, and I imagine their price is about to come way down since the M6 mini just got announced.

Software-wise, it's different, but mostly equivalent. Homebrew or macports has a similar software inventory to debian. And Apple's container framework is a welcome improvement over colima for running most container workloads.

Neat, now put the M5 Pro or Max in a 24-inch iMac, and sell it with a reasonable amount of RAM (more than 24GB).

They'll have to "partner" with some company that's in the business of building a database of IDs and biometrics to do AI things with. Other companies in this space (Jumio) have a bad habit of ignoring privacy laws and will keep your information for years.

I wouldn't mind showing my ID to a person (in person), but there's no way I'm letting some company get a scan of my ID or passport to store in some giant database that's a rich target for hackers. Might as well give them access to all my bank accounts (Plaid) too.

(It sure would be nice if there were a national privacy law in the US.)

Also, it's illegal for companies to use facial recognition in my jurisdiction, so if I allowed them to "verify" me, they'd be breaking the law.

The article uses "robotic mobile probes," which also seems out of order.

I (native english speaker) would order it as "mobile robotic probes". But if I were writing it, I'd say "robotic probe", "surface probe", or "mobile probe". In this case, robotic and mobile mean the same thing, so using both is redundant.

And although I would order it as "private robotic spacecraft", I don't think that's correct. The spacecraft is robotic, but it's not private. It might be privately-operated, privately-owned, or privately-funded (each has a slightly different connotation). But private by itself means that a private company is somehow responsible for the mission.

So if I were writing it, I'd use something like "privately-funded robotic spacecraft" or "robotic spacecraft operated by private company XYZ".

Facial recognition is illegal where I live, both for gov't and commercial uses. Several major cities in the US have banned it (e.g., San Francisco, Boston, etc.).

I was not aware of this! That's kinda fun.

I did an entropy test on my Pi5 (according to https://rob-turner.net/post/raspberrypi-hwrng), and it (7.999832 bits per byte) has about the same entropy as /dev/urandom (7.999831 bits per byte).

However, when using it directly, it's pretty slow. /dev/hwrng is 200 KB/s, /dev/urandom is 40 MB/s.

Though, maybe that doesn't matter if it's just intended to be used to add entropy to the system entropy pool.

Oh how I wish I had a legion of junior engineers. Or even a squad.

My company has an aversion to hiring. It's so expensive to hire people, you know! There are no young engineers to teach the ropes to. There are very few senior people do everything.

Needless to say nothing happens fast, good, or cheap. And we don't ship projects, they mostly just bob up and down in the harbor.

Or better yet, buy your phone from the used market. Get a phone in perfect shape that was $800 two or three years ago for $200. Put a new battery in it.

Though I'm seriously considering going back to a $50 flip phone and enjoying the 2 weeks of battery life and general indestructibility. My current phone spends most of its time sitting on my desk doing nothing. It's hard to get excited about a newer and much BIGGER phone for $500 that will also spend most of its time sitting on my desk, doing nothing.

This approach works very well. I've had dozens of extremely remote systems hooked up this way for about 8 years. The only problem I've seen is that occasionally the server ssh process will get stuck, so you have to log in to the server and kill it. It seems to happen when a remote goes offline and reconnects without closing the old connection first.

If I were doing it now, I'd probably use wireguard, probably. This is simpler to set up and works great.

That's funny. $10 billion just to move junk around.

The Wheatridge solar/wind/battery facility in Oregon produces the same amount of power as a small nuclear reactor. $10 billion buys you a dozen of them.

It's not necessarily the organization's fault. In several companies that I've worked for (including government contractors) we are required to implement "certifications" of one kind or another to handle certain kinds of data, or to get some insurance, or to win some contract.

There's nothing inherently wrong with that, but many of these require dubious "checkbox security" procedures and practices.

Unfortunately, there's no point in arguing with an insurance company or a contract or a certification organization, certainly not when you're "just" the engineer, IT guy, or end user.

There's also little point in arguing with your boss about it either. "Hey boss, this security requirement is pointless because of technical reason X and Y." Boss: "We have to do it to get the million dollar contract. Besides, more security is better, right? What's the problem?"

Phoenix does this:

https://www.scientificamerican.com/article/to-beat-the-heat-...

> Experts say road temperatures in the Phoenix area can rise to 180 degrees on a hot day

I guess being only 160°F means the gray is working. I wonder how well it would do if it were painted white.

Please make speed limit signs broadcast speeds by radio!

That would be incredible. If the road noise outside my house gets too loud, I can transmit a lower speed limit instead of turning up my TV.

Without significant whitespace, you'd need at least 5 more comments consisting solely of "}" for this reply chain to be syntactically correct.

This is a way to show off someone's mathematical work to a non-mathematical audience and to promote the college. Would you prefer "Our scientists wrote a paper about some math they did called 'An Improved Calendar Ring Hole-Count for the Antikythera Mechanism'", you should read it." ?

That doesn't seem as likely to grab the attention of people who watched Indiana Jones and the Dial of Destiny.

The antikythera mechanism is a (bad) predictor of planetary locations. People have recreated it in its entirety using legos. There isn't much mystery about what it does.

However, a new high resolution X-ray of the device inspired some scientists to do some neat math on it. I read the paper, it's good work. I'd love to a chance to get an article published about one of my papers, even more if regular people had even the slightest chance to understand it.

Cold batteries do struggle a bit to deliver power and charge when they're cold, but they're just fine being stored in low temperatures. LFPs are really good at maintaining their charge. Even below -5°C, they should be fine, particularly if they're not being used.

Also, I wouldn't worry about the BMS too much. Unless it's really dumb, when there's no charging or discharging happening it will put itself to sleep for most of that time. Technically it will use power, but likely only a few hundred micro watts to a few milliwatts (nothing that will make a difference on a 5 kWh+ battery). Then when the sunlight comes back in the spring (or whenever the snow melts off the solar panels), it'll see some input voltage and start feeding the battery again.

LFPs are indestructible and last forever.

For general advice, they like it cooler[1] than other kinds of lithium batteries. Try to keep them between 15-25°C. Other than that, charge LFP batteries to whatever percentage you want whenever you want.

They are simply better than every other kind of lithium chemistry. Their practical downside is lower cell voltage (3.2 vs 3.8), and slightly lower energy density. I've spent the last 7 years building LFP-powered lawnmower-sized robots, and in practice, the lower voltage and doesn't matter. And, unlike other chemistries, they don't catch fire when you stab them.

For other kinds of lithium batteries, to maximize lifetime, keep them between 40-60%, avoid big charge swings (10-90%), and keep them warm (above 25°C).

[1]: At 35°C, tests show they lose about 5% more capacity after 2,000 cycles compared with 25°C. But even that isn't really a problem, because they don't get to 20% capacity loss until 4,000 cycles. See: https://iopscience.iop.org/article/10.1149/1945-7111/abae37, fig 3.

Another that I've used: https://github.com/posva/catimg

Most of the leaf blower noise in my neighborhood is the one or two guys on the crew who's job it is to run the leaf blower while the other guys are doing the lawn mowing or edging.

If it was that guy's job to use a rake and broom instead of a leaf blower, the same amount of cleaning would be done.

I've been looking at that particular rabbit hole since a professor of mine mentioned it in 2003 or something. Once or twice a year, I'll read about some theorem or something and think it can be applied to Collatz somehow and dive back in.

I've actually proved it several times...except for the insignificant detail that I glossed over that didn't seem important but tanks the proof.

Someday I'll have to publish my "book of lemmas that don't prove the collatz conjecture."

You _can_ do it, but EV charging is a big load for a few hours. That's not really what solar is good at. Solar likes smaller loads spread throughout the day.

If you go super cheap (the "$1000" system in the article), you'll be able to put a couple miles into the car in the afternoon on "level 1" charging. It's not nothing but it's not really practical for a car (but probably fine for an electric bike/moped). Also, most of the available power is unused since the battery is so small.

So you get a battery big enough to hold everything your panels can generate. You get a more powerful 240V inverter, a nice 8-10 kWh battery, and move to Southern California. With the best possible circumstances in July, you can fill up that battery every day, and add 19 miles to your car overnight.

Now you're at $10,000 for about 6,500 miles per year.

If the battery lasts 8 years (LFP can do 3000 cycles, 1 cycle per day = 8.2 years), you're paying 19¢/mile (and only if you use all of the power you generate).

Compare that to just plugging into the wall, which is 3-6¢/mile.

First they break lolwut (https://github.com/redis/redis/issues/12074) and now this.

See: https://news.ycombinator.com/item?id=38841284

The old Mac Plus had a screen resolution of 512x342. His screen is 512x384 (4:3), which is why there's that extra blank space at the bottom of 3D Checkers and Mac Paint. I think the Color Classic (1993) was the first compact Mac with a built-in screen to have a normal 4:3 resolution of 512x384.

But I guess I can let that one thing slide considering how incredible the rest of the build is!