chabons

They didn't even attempt to make a clean-room implementation, they just had the LLM clone it for them with access to the source.

"[…] and these interlopers are the real geniuses […]

This wasn’t the framing I was going for. My point was that industry boundaries are fluid and expand/contract as demand dictates. You’re correct that incentives are not all positive (pay, work life balance, perks), other industries contracting might force people to find work elsewhere.

All that said, I don’t think these people have “odd backgrounds”. I work in a math-heavy domain, so these backgrounds make as much sense as a traditional CS background, and I think these folks are just as likely to be retained in a crunch.

“There is a population of qualified workers […]”

In my experience, this is not true. Demand for software engineers has been so high, and pay so high as a result, that it’s pulling in workers from adjacent industries. The total software-qualified workforce is larger than the set currently working in software, and people with transferrable skills move in and out of software as incentives dictate.

A number of my current and former coworkers are from math and physics backgrounds (CFD, energy, etc…). These are folks that before might have stayed in academia, or ended up in aerospace, defense, or other engineering fields.

If everyone over hired, demand drops, and companies drop pay as a result, I’m sure we’ll see some folks in software with transferrable skills move to other industries.

Where in academia can one get a Billion (with a b) dollars to research something?

Developers are users too. If users cared enough not to use cross-platform software because it didn’t look native, then it wouldn’t get built.

"rather hard" is really underselling things. Without considering generator->motor losses, 250W is a lot of power, even for a larger (>80kg/176lbs) rider. For basically anyone who doesn't train on the bike, this is an unsustainable effort.

Even a floor of 100W would rule out smaller/less athletic riders.

Intuitively, I can understand that English Second Language students would struggle in classes other than English, but are the demographics really shifting enough to explain the drop in attainment shown in the article?

The best demographic data I can find is here: https://datacenter.aecf.org/data/tables/103-child-population...

The best data I can find on language spoken at home is here: https://datacenter.aecf.org/data/tables/81-children-who-spea...

The above shows the share of "Non-Hispanic White alone" children (who I'll assume speak English as a first language) going from 52% to 48% from 2015-2024, and the percentage of "Children who speak a language other than English at home" staying flat at 22% from 2013-2023. From 2015-2024, math attainment goes from 62% to 55%.

At a glance, it would seem that the shift in math attainment cannot be explained by demographics/language alone.

I'm missing something. What change in demographics are we talking about, and why would that influence math results?

Garmin has a market cap of ~40B USD. Inflation adjusted, that's where NVIDIA was <10yr ago. It's not missing resources to do the work IMO, but it's missing the market dominance to singlehanded set the standard for the industry.

For me, it's that the ergonomics are straightforward, and everything works out of the box. If I find myself on a new machine, just installing fish gives me an ergonomic setup without having to install too many additional tools or mess with configuration.

I don't have statistics, but given the student visa -> H-1B pipeline changes, it would seem there are a number of H-1B holders who are educated in US colleges (either at the undergrad or graduate level). This indicates that the problem is not entirely a training gap.

The author specifically stated: "Realizing this, I asked for the report to be forwarded to an actual Zendesk staff member for review", before getting another reply for H1. I read this as they escalated it to Zendesk directly, who directed it back to HackerOne.

I was thinking the same. Also, I would have liked if they discussed/measured the worst-case power draw of the device before starting to add capacitors, as it's possible a PC USB port just doesn't supply enough current to power it at steady state.

Before I knew this I had someone pull out their lighter and point the remote at it when our sensor bar died. Took me a little bit to figure it out.

Embrace them and make rules to prevent them from damaging the city they’re coming to experience.

+1, squirting the tourists who are already there does little to discourage other tourists from coming, or to solve the problem that they're competing for the same housing.

That said, I can't help but think that AirBnB(/VRBO/etc...) is just faster to respond to market conditions than traditional hotel businesses because it leverages already built private residences. If land is 5x as profitable as accommodation for tourists, then anywhere where a hotel could be built the value of the land will increase, and thus housing costs. I guess the local govt. has zoning as a lever it can use to restrict where hotels can be built though. Maybe that's the point though, that AirBnB sidesteps zoning, extracting more value from the land than other users can.

The happened for me only in Celsius. It looks like in Fahrenheit the temperature updated for me.

In general I agree, for instance an Apple Watch crams in so much hardware, that the battery frequently lasts less than a day. Compare that to my Garmin 245, which often lasts 2 weeks on a single charge if it's not using the GPS. It doesn't do as much, but that's fine. I'd love to see more "quiet" tech which eschews the pattern you describe above, and just focusses on doing the thing it's supposed to well so you don't have to think about it. EInk is another promising technology in this area.

And more generally, consumers of Windows-only software, of which there is still a ton.

This doesn't resolve the ongoing incentive to remove the LVT, creating immediate value for the current land-owners.

This assumes that the same number of vehicles use the charging station. Lower charge times means potentially higher steady-state throughput.

Here I was scanning the article for what they floated on. I had assumed it was a peat-bog type situation.

I guess the key benefit of this strategy is that the water level is high and the water is nutrient rich? Otherwise it's pretty similar to a standard canal irrigation setup.

I really doubt this would be how things play out. Most crime isn't sophisticated or targeted.

The same argument can be made about internet-connected security cameras in general, that they could allow for remotely casing a potential target, and they're generally considered to be a deterrent, and not an enabler of crime.

The videos are all shown at 2x speed, but your point stands, this is still pretty quick.

Intel Gaudi 3 has more interconnect bandwidth than this has memory bandwidth.

LPDDR5 vs HBMe2. I'm guessing there's a 2-5x price difference between those, but even so it's an interesting choice, I don't know any other accelerators which spec DDR. But yeah, without exact TCO numbers it's hard to compare exactly.

Presumably as functions, the same way as excel?

and(x, y) -> bool

or(x, y) -> bool

if(cond, funcTrue, funcFalse) -> void

Well, except we have video evidence of thieves wearing balaclavas during burglaries, not so much of them using FlipperZero's.

The way I see it, induced demand is just a way of saying that all modes trend towards equilibrium. If you invest more in trains, they become more attractive, and people gravitate towards them. The same is true of cars, busses, etc... How each mode of transport scales influences whether you want to induce demand for it.

AFAICT trains scale pretty well in terms of space and environmental impact, and point-to-point latency decreases as more trains are run (and transfer times decrease), so it's pretty nice to induce demand for it.

Interesting point. That said, the AI model space is rapidly evolving, while bitcoin's hashing problem is static. This makes it significantly more risky to make a large capital investment in dedicated HW when it's unclear if it will be able to run the next big model architecture. For instance, if this had been built + released a year ago, before SOTA models used MoE , then it would rapidly have become obselete.

AI models on FPGAs has been tried before, for instance: https://www.microsoft.com/en-us/research/project/project-cat....

They haven't been able to compete with GPU's on perf/watt. In general you end up just designing some AI accelerator for the FPGA (because the models are too big to map onto a single device all at once), but it's hard to beat purpose-built tensor and vector HW on a GPU when you're running soft logic.

Right, but the argument that the original comment was making is that the government's policy is incorrect.

My comment argues that independent of government policy, relocating an existing employee from one country to another does not consume a position that was otherwise free, unless the employee was hired expressly for the purpose of relocation.