ajb

I wonder if they need to do this because of the EU Cyber Resilience Act means older designs can't be sold any more there. Apparently that's causing a lot of design refreshes.

If your building or other construction is unsafe for the public, the local authority has the right to make it safe, if necessary by demolishing it; if you are insufficiently responsive to a notice that you must make it safe. Maybe tech companies need that kind of incentive before they will care enough not to subject their customers to this kind of abuse.

Yeah this is what pick-up "artists" call a "neg" - a negative judgement made purely so the recipient feels, at some level, that their credibility is on the line. In this case, for clickbait. I always either don't read such an article or come here to read the comments instead.

There's a lot of lobbying against open models on the grounds of a) China and b) open models mean we won't be able to prevent a hard takeoff.

It's " a major element of a £237m scheme to upgrade Oxford rail station". Most likely it is more expensive than it should be, but not to that extent

Well, it has been shown that birds have twice the density of neurons to mammals, and their brain is larger as a proportion of body mass: https://www.pnas.org/doi/10.1073/pnas.1517131113

So their brain can be equivalent to a much larger primate.

This makes the assumption that crows can identify people only by the face. It's quite likely they could recognise something else.

Well, it's not my field, but pci5.0 x4 drives are the fastest I found with a quick Google. So 24 of those would top out at ~400 GB/s. That's also where HBF starts out, for one device. The higher grades are faster, but maybe NVMe will be as fast by the time they are out.

But the win here on servers is almost certainly power. High speed signals are power hungry, and the power scales with the length of the wire. HBF is designed to go in the same package as the GPU, so the distance is much shorter than with discrete drives.

"Another notable document is a letter from Hittite generals in western Anatolia appealing to the king for reinforcements in men, horses and chariots to fend off enemy forces. "

I had assumed they had to wait for clay to dry before it could be carried anywhere, which seems incompatible with using them for urgent military messages. I wonder if there was a way to transport them while soft, or if this was just the archival copy of a message sent by other means.

When High Bandwidth Flash comes out, most likely: https://chipsandcheese.com/p/hot-chips-2026-applying-high-ba...

One of the few mathematicians not fazed by all this is Doron Zeilberger, who has given his computer (which he names Salosh B Ekhad) co-author status since 1990. He is a bit of a contrarian, but a serious mathematician. He co-invented/discovered Wilf-Zeilberger pairs. Some of his opinions:

On the Leiden declaration: https://sites.math.rutgers.edu/~zeilberg/Opinion196.html

On proof: https://sites.math.rutgers.edu/~zeilberg/Opinion174.html

Hmm having said that, the nearest borehole record shows chalk doesn't start until 26 feet down: https://api.bgs.ac.uk/sobi-scans/v1/borehole/scans/items/423... Before that it's sand, gravel, etc. Odd. Record from 1896, but it won't have changed.

That's because London is on clay. The Reading guy was mostly on chalk (and was himself an engineer). It sounds like the water table was a long way down as well. Soft clay or waterlogged sand are more like digging a hole in a bowl of custard. There are a lot of things that can go wrong - in the case of clay, even if you don't collapse the hole the clay can expand for years afterwards ("heave").

Interesting. TBH surprisingly few accounts require you to actually send from the same email; mostly they use various mechanisms based on what they sent - a URL, or ticket ID in the subject, or in the from address, etc. I have not actually yet run out of aliases in the standard proton account, but I'll bear in mind that they have simplemail too.

Yes, I was referring to their doc here: https://migadu.com/procon/#no-mailbox-2fa-yet

Protonmail doesn't have unlimited addresses for sending, which is one of his requirements. It has catch-all , which means you can receive from an arbitary number, so it's okay for one -email-per-vendor, but if you ever need to send via one of those addresses then you have to use up one of your 10 or 20 allowed addresses.

This is because like most other vendors, they do price discrimination based on number of users. Which is no doubt why the other large vendors were also ruled out.

One vendor who doesn't do this is Migadu. I don't use them as they lack 2FA support, but they are one of a small number of businesses who make their proposition easy to evaluate quickly, as opposed to making you swipe through a pitch before you get to the facts. This is so unusual that I remember them despite evaluating them multiple years ago.

There's an inherent insularity to security groups or fraud teams; they have to have a professional suspicion of everything. This can go wrong and end up being NIH or gratuitously customer-unfriendly.

Also, for finance specifically : " A sound banker, alas, is not one who foresees danger and avoids it, but one who, when he is ruined, is ruined in a conventional way along with his fellows, so that no one can really blame him." - Keynes

I guess I'm the past, rulers old enough for that also usually collected enough other health issues to be largely inactive. I can think of a few cases where countries had to wait a year or so for the old ruler to die, but not another case where they were still active while demented. Maybe Ivan the terrible? https://en.wikipedia.org/wiki/Ivan_the_Terrible_and_His_Son_...

Oh, very interesting. Not sure it's working correctly though: set everything to ambition level 4, which takes us to net zero minus 4. Then, the following knobs have no effect: efficiency, nuclear power, carbon intensity, solar (!) and wave & tidal. That can't be true, so I think it's buggy

Great story! Was expecting it to be Bernie Madoff's, not Sears lol

Way back there was a news story about one of those, in which the owner didn't pay the licence fee for the software, so the software company bricked their parking garage with loads of people's cars inside.

City of Hoboken, NJ. Plenty of more recent stories about Hoboken using robots, so I guess it didn't put them off.

Oh yeah, here it is: https://m.slashdot.org/story/70742

(Edited to add: that account is disputed, in some accounts the city stopped maintainers working on it without realising it wouldn't keep working)

There's a reason why an interviewer might not say what they were looking for: when a candidate raises them unprompted, that gives more confidence that they have actually done the job, rather than just memorized some bullet points.

It's a hard thing to do right though. Looking for some very specific topic may not occur to the interviewee that it's what you're interested in, so it's better to just try to give them more general opportunities to show their thinking. The ideal is to get them into the frame of mind where they feel they can reminisce about some difficult task, rather than just "answer the exam question". You get more signal that way, but you have to be prepared to give credit for whatever important things they were able to talk about.

Interviewing is hard, honestly. It's not most interviewers main task, you can go for stretches of your career without any practice, and there are many reasons other than being an arsehole that people might not do it well. And when you do it, you have a fixed amount of time, and can't easily "try another way". As a job seeker one has to realize that a largish fraction of rejections are because an interviewer "couldn't get it to work in 1 hour", not because of you. Although frustrating, as it increases the amount of time you are without a job, it is liberating in a way to realize that interviewers are just people muddling through, not perfect judges whose rejection you need to take as an accurate prediction of how others will see you.

Competition is not one thing. For a producer of widgets, competition means that there is immense instantaneous leverage if you can reduce the price, because buyers will switch to you, increasing your market share until your competitor can match - if they can.

A landlord is usually in a different position. If their personal costs drop (eg, they paid off their mortgage) but the market price stays the same, how do they benefit by reducing the price they offer? Only by reducing the time it takes to rent out, which isn't a significant factor in exactly case when tenants would most like rents to go down - when prices are high because of demand.

The details of these mechanisms matter. Market can be out of equilibrium for a long time; maybe indefinitely.

A transaction can occur if the price is >= the seller's cost, and <= the value to the buyer. If it's at the minimum, the seller must pass on the tax or not trade, at the maximum (down to the max minus the tax) , the seller can't pass on (all of) the tax and still trade. We observe that landlords make profit, so we are not at the minimum. Are we at the maximum?

I think there are cases where we are in between for significant periods of time. Consider a positive shock to wages. Can landlords put up prices overnight? I don't think so, unless the landlords all co-ordinate to do so, prices will be sticky as there are always some properties in the market, so it's difficult to be the first to increase rent. Unlike a purchase, a landlord who waits for a buyer at the right price is forgoing income during the void. So the market price can be a Schelling Point.

A tax increase, however, happens at the same time to all landlords. All of them would prefer to pass it on, and they know that so they can assume all other landlords will try to pass it on. If the rent is currently below the maximum possible, they will succeed.

How long it takes prices to adjust I don't know. It is an empirical question, but I don't know what data would answer it. But it seems like it took decades for landlords to capture the two-income surplus after it became common for women to have a career.

In the UK, the state has a conflicted relationship with mobile vending. On the one hand it is controversial to restrict commerce in a free society; on the other, there is an increasing aversion to being called to the door, or offered goods in the street. Technically you are would be required to have a "Pedlar's certificate", but there is a right to obtain one unless you are known to be of "bad character". "pedlar" is a very old word today hardly used except in the phrase 'drug pedlar'; and you cannot obtain such a certificate except by application at a police station, a rule unchanged since 1870. There are only a few thousand people with this licence and most street sellers probably don't have one (this does not include sellers at a fixed "pitch", who need a different (and more expensive and scarce) licence from the local council).

This article shows the biasses of the press. Everything is about public perception, so the problem is that the public don't perceived that there might be a war. This is stupid, public perception can shift very quickly. The actual problem is not maintaining investment and critical industries. The UK was able to shift commercial manufacturing to wartime production for world war II. That doesn't work if you've offshored everything.

Rather unfortunate if so, as aphantasia has been known about since Galton discovered it in 1880.

I'm not saying you're wrong, I'm saying that's a case you have to make, not one you can just derive from an accounting identity. Also it doesn't have to be increased revenue- it can be reduced costs. This is more likely to be their motivation. Of course, it's one that may be more problematic for the economy as a whole.

In general, political discussion suffers a lot from far too much of it being simply "I am obviously right, you are an idiot and/or malicious if you believe the other guy". This is not a dig at you personally, look at any political discussion and you will see that 99% of it is like that. Our public sphere would be in better condition if more people pushed back on it. So no, I don't think anyone should accept "proof by ridicule/slur".

That assumes that their use of AI does not generate cash flow. Trivial example: the ones reselling tokens.

You can't analyse expenditure assuming income is constant, because all expenditure decisions will be made with a view to how they affect income.

Economic activity is a flow, not a stock. There might well be a bubble, but you can't prove it like that.