Posts by Justthefacts

EU Chips Act heading for failure, time for Chips Act 2.0

Re: Wrong idea?

Why are EU and MAGA both so *obsessed* with what the automotive industry wants? It’s a lousy industry.

The average is more like 1500 chips; 3000 is for a super-luxury fully-loaded with options that almost nobody has. The average price of those chips is about 50 cents each, the bulk is sensors and motor controllers, the *top end* of those 1500 is a bunch of €1.50 microcontrollers. So the total price is *maybe* €1000 of chips….which are selling at 10% margin. We’re arguing about €100 of value. Not even €100 per year, it’s €100 over the 15-year lifetime of the vehicle.

You want to swing your entire industrial economy around €8 per *household* per year. *Toothbrushes* have a higher Average Revenue Per User. Look it up.

Nationwide power outages knock Spain, Portugal offline

Well, no, that is clearly not the underlying cause. That may have been the *trigger* certainly. But their real problem is having a cascading-fault mode at system level.

White House confirms 245% tariff on some Chinese imports not a typo

Re: What would you do?

This is just a weird Republican talking-point, entirely free of truthiness. Since you want to focus on low-cost drugs, ie generics (why? because it suits your argument, and ignores the higher value ones where the EU and US are dominant?)

The generic drugs market is dominated by India, *not China*. Do I have a problem with that? No, I do not. Why should I, their quality-control and logistics in the volumes we are talking about is good, and so is their efficiency and cost-effectiveness.

Why do you have a desperate need to avoid being “dependent” on “India” for packs of paracetamol wholesale at 16p/pack? There’s clearly no economic argument. Among other things, if these were made in Yorkshire, you’d be just as “dependent” on the factory in Yorkshire, and subject to the whims of unionised action, flooding of the Ribbledale Valley etc. What you are describing is just being dependent on less diversified supply-chain. There certainly have been supply-chain shortages of drugs, but that’s usually turned out that there’s only 1 or 2 factories globally for one process-step, which nobody really knew about, and which had a manufacturing glitch. And usually in the USA by the way.

Re: 14 reasons why Trump’s tariffs won’t bring manufacturing back

“import tariff was based at the cost of the product as the point it enters the country and not the price sold in the shops”

This is true, but it’s only half the story when it comes to price-setting. There’s a reason why typical retail markups are 2.2x, somewhat independent of price. To sell something that costs £1000 wholesale, you have to work a lot harder in getting customers, than if it cost £10 wholesale. You need more premium retail locations (higher rent), more effort shmoozing the customer (more and higher-trained sales staff), more marketing/advertising spend, and there’s just fewer customers who can or want to afford it. Very roughly, retail costs will scale with the wholesale price.

The end-customer price might not track perfectly with the tariff. But thinking that the tariff only affects the raw cost of goods, while the rest of the value-chain remains fixed, is just wrong. For a concrete example, USA uses about 7.4billion barrels of oil annually, worth $480bn = 1.8% of US GDP. And yet, doubling the price of a barrel of oil in 2022 caused an oil shock and 8-10% inflation.

Self-driving car maker Musk's DOGE rocks up at self-driving car watchdog, cuts staff

Re: Self-driving is a fallacy

Horses for courses. About 20% of the country is either too elderly to drive safely, or too disabled to do so (eg blind). Most of them live lives that are significantly constrained, to being nearly housebound. So to answer your question “how much is it worth” - maybe £5-10k per year? Whatever a taxi would cost you, if that were your only option.

EU lands 25% counter tariff punch on US, Trump pauses broad import levy hike – China excepted

Re: Trump blinked

So…..basically QE, “print dollars”? Yes, they can. This will tank the dollar exchange rate, so all US imports will become much more expensive.

There are half a dozen technical ways for the USA to cope with the fallout, but all of them end with the words “so all US imports will become much more expensive”.

Re: EU to charge citizens 25% extra tax on loads of stuff and pocket a stash of their cash.

“You’ve chopped off your arm, to show that you can win with one arm behind your back. So I’m going to chop my leg off, to show that I’m John Wick and can kill you with a fucking pencil”

You may well kill the other guy. But at the end of the day, you’ve still chopped off your own leg to prove a point. Legs take at least thirty years to regrow.

“decided to try and pick a fight with just the one bogyman for now: China.”

That, and he’s playing the classic bully game of “identify the victim and get everyone to pile in on them”

He’s hoping that everyone will agree that China is “taking advantage”, and raise tariffs on China too. Whereas the smart thing to do is the opposite - global free trade, while the US self-embargoes with tariffs.

EU: These are scary times – let's backdoor encryption!

Re: Ursula said it

Also, for clarity for the Wikipedia warriors: sigh, yes, nominally the back-and-forth includes so-called trilogue, which is Commission, Council of European Union, Parliament. This re-discusses and agrees amendments, in a series of meetings, and its majority vote among the three.

However, the actual meeting is this: the Commissioner is there with a large retinue of flunkies. The Council of Europe…*is allowed to be* a minister of a national govt, but I don’t think has ever been that, even once. In fact, it’s a representative of the *Presidency* of the Council of Ministers. Remember that, *Presidency*. The *Presidency* office is really a floating civil service all of itself; it is headed by a rotating head-of-state (every six months); but the people who actually staff it are (of course) full-time employees who are appointed by, and receive their salary cheques from, the Commission. Yes, the Polish PM can tell them overall priorities, but he isn’t really their boss, and anyway if the Commission doesn’t like them, it just waits until the next six-month Presidency for all-change. So, it’s majority voting where the Commission controls two out of three votes.

But….none of that matters anyway. Because the output of this big trilogue is….a set of minutes. The only requirement on Commission is that it produce an updated “file” (including proposed legislation). There’s no requirement that Commission modify the proposed legislation in the agreed direction. And it *does not*.

As I said, as somebody who has been part of the lobbying and decision process on significant parts of technical legislation, over many many years: neither I, nor anybody in the lobbying team has in fact ever *read* the trilogue minutes. Nor has Commission ever raised them with us. At all. They are simply ignored. The file gets re-submitted, with the mods *we* want, and goes round for a few goes, and gets mostly passed by Parliament, and if not, it comes back for another round of lobbying. The *trilogue* might as well go to the pub, for all the effect it has.

Re: Ursula said it

Right, so your claim is roughly: “These bits of legislation that appear to have passed into law despite being rejected by Parliament, are subtly modified on a technical level to meet Parliament’s concerns”

But you do admit, *as a europhile involved in the process*, that there have been *zero* examples of when Parliament successfully rejected legislation against the direct will of the Commission. Thank you. At last. The vast majority of europhiles have no contact or knowledge of how the business is actually done. They do not know this, and refuse to believe it when told. Thank you.

Onto your point: I used to work for Beelzebub on the other side of the fence as a, shall we say technical/management corporate guy regularly lobbying Commission. All I can say, having been involved in multiple re-drafts, is that at no point did either we or the Commissioners discuss the views of Parliament. We didn’t talk to MEPs, we didn’t read debate transcripts, we didn’t look at voting records or try to understand “changes needed for acceptance”. Nothing.

Nor did Commission represent to us that they were acting to address Parliament concerns. In their view, they were updating policy in line with whatever was internal or situational change had occurred. There was *certainly* no view that in any sense Commission had “got it wrong”’ or were being corrected. In all my dozens, probably hundreds of meetings, and hours prep-work with legal and management teams, re-drafts were simply another opportunity for us to re-angle and update our interests into the legislation.

And finally these “UK representatives” (by which I assume you mean UK govt). We talked to them. But you know, we’re in charge. We tell them what industry wants. We’re a major transnational for both UK, France and Germany. All the “national” representatives are basically reading from the same script we gave them, apart from where there are national site tensions internal to the company, and those definitely exist. I had louder “assertive exchange of views” with my opposite number in $Corporate Germany, than ever with the Commission.

I hope the view from Beelzebub clarifies matters on How Things Are Done.

Re: Ursula said it

Can you list *actual examples* of EU legislation, where the Parliament rejected the Commission proposal, and it didn’t just get passed back one or more times until they agreed?

Re: Ursula said it

So, your defence of EU legislation, is you think Germany doesn’t have to follow it? It’s funny, the French exceptionalists seem to think exactly the same thing.

And it’s exactly the same reasoning that got Trump elected by government workers too. Because obviously DOGE isn’t going to eliminate *my* position, only the corrupt Dems.

You are embarrassing yourself.

Americans set to pay more on all imports: Trump activates blanket tariffs

Re: Please explain

Honestly? I really wouldn’t “improve” that. A trade deficit is desirable for the USA. Importing stuff, means that the USA gets given goods from other countries, in return for US dollars *which the USA can print at its own discretion*. It’s free money. Tribute from the rest of the world, if you will, for privilege of dominating the world militarily and economically. A trade deficit means a strong dollar, which means you buy stuff from the rest of the world at prices in your own currency *far below* what they cost elsewhere. Equivalently, it means that your salaries are *twice* what they otherwise would be.

Your trade deficit is the main driver behind the dollar being a global reserve currency: you are printing USD (which costs your citizens nothing), and exporting the dollar bills. Those dollar bills are what the rest of the world uses as its transaction currency. No exported dollar bills = no reserve currency. Every year, you print $3trillion. What do you think your inflation would be, if that all that newly printed paper stayed inside US borders?

This is a problem you want to solve?

Congratulations, China is about to solve your “problem” for you. You will lose global influence and dominance, almost overnight. The dollar will crash to half its current level, if that. Your input prices will double.

Re: Econ 101

Import duties only encourage local production *under some very limited assumptions*. Mostly they *discourage local consumption*, substitution is the norm.

This should be obvious. Let’s say you like both oranges and kumquats. Oranges are grown locally, kumquats overseas. If you add tariffs, the price of kumquats will go up (local production must be more expensive, otherwise it would already be done locally). So you mostly stop eating kumquats, and only eat oranges.

Reality is more complex. In practice, tariffs have increased *because local politicians hated being dependent on imports*. A section of the population already “won’t eat that foreign muck”, and with the addition of tariffs it becomes a Big Patriotic Deal to not eat kumquats. It becomes clear that if you are seen eating a kumquat, you will lose your job. The tariffs become irrelevant, as nobody can eat them without getting the Orange Purity League on their tail.

But the net outcomes are: not many kumquats get eaten in the tariff country; people who prefer kumquats don’t get their favourite fruit; the exporter will drop their price by a small fraction of the tariff (10% of the 10%, maybe 1%) to stimulate global demand; any remaining secret-kumquat-eaters see a large price increase; the beneficiaries are kumquat-eaters in *third-party importer countries* who see their kumquat price go down by 1%

Re: Well, he was right about

The list of people who the right-wingers call hard-left is never-ending.

It’s doctors, nurses, lawyers, judges, accountants & management consultants (insufficient qualified non-hard-left forensic accountants to audit government!), airline pilots, the military generally, all civil servants.

Re: how tariffs work 101

Any exporter that drops their prices, is still only going to do it by a negligible fraction. Worked example:

I sell something for £100, on which I make 10% gross margin = £10 profit. Now the target country wants to charge 10% tariff. If the exporter absorbs that, they make *zero* profit, so they won’t do that.

However, it’s also true that if they stopped selling to that country, they’d lose X% of total sales, which they need to make up elsewhere. And the only way to do that is by dropping their price to buyers in other countries [WAIT! I don’t think that’s what Trump had in mind! Yeah, that’s right, Trump is an idiot, if the USA is worried about subsidising other countries, that’s exactly what he is about to do, the net beneficiary of any trade war is *buyers in third party countries*]. At 10% gross margin, the equilibrium selling price drops by 1-2%. 90% of the tariff impact is on the import side.

Now, I know what the MAGA / EU tariff crowd think (same people, different flag). “We’re a huge market of 400 million people, we’re your most important market, if we raise the tariffs then we’ve got you over a barrel, you’ll have to pay them”. No. The reason why that’s wrong, is that *if* that is the market dynamic, then the importer *already* has the exporter over a barrel. They’ve *already* got squeezed prices to the maximum possible, the market dynamic is already priced in. That’s how prices work, they sit at the point where the seller is prepared to walk away if the buyer pushes harder, because there’s insufficient profit. If the market dynamic is as described, then the gross margin won’t be 10%, it will be 2%. So if the exporter absorbed, they would be *losing 8% on every tonne they ship*, which they aren’t going to do.

And for those saying “Ah, but TSMC are selling at 40% gross margin”. Yes, they are. Because they can. It’s telling you that if the buyer, walked away from the table, TSMC can get only slightly less than 40% margin from their remaining market.

Re: I feel liberated already...

They’ll also cause *exports* to drop a lot too.

Partly from retaliatory tariffs. Partly from non-tariff barriers: consumers in countries you’ve threatened to annex, don’t tend to want to buy your stuff.

And partly because chaos is bad for business in so many complex ways. eg we’ve all been through Covid supply chain issues, I *know* that those cargo ships are going to be stacking up outside Chicago, while their owners figure out if the trade is still profitable. In which case, are they coming *back* on time? And if they’re not, if I order 6000 tonnes of steel from US manufacturer, am I going to see it this side of Christmas?

It’s not at all obvious that the overall USA trade balance will improve.

Re: Trump is easy to model

“If your insurance is happy to pay $1500 for an ambulance ride then that is what they will charge.”

In the U.K., the average cost of an ambulance call-out to the NHS is £417. Seems like “socialised medicine” is more efficient and cheaper.

https://www.kingsfund.org.uk/insight-and-analysis/data-and-charts/key-facts-figures-nhs

You’re right about US ambulance costs: $1300 + $15 per mile is the going rate.

https://www.talktomira.com/post/how-much-does-an-ambulance-ride-cost-without-insurance

I’m all for dropping tariff barriers completely in all developed nations.

It’s worth remembering the argument for less developed countries keeping them, to prevent the gorilla-in-the-room problem. If you are, say, a small island nation whose main economic output is a particular type of traditional handwoven hand basket. And you find that you cannot compete even *internally* with TEMU hand baskets at $0.30 each because of volume production. Then what do you do? If you drop tariffs then you will quickly find your island becomes a bogus re-seller of TEMU baskets, to gullible tourists, and what sort of life is that.

There’s also the cultural imperialism problem: if eg Peru dropped tariffs against US cheese (I don’t know if this is a real example or not), and US plastic cheese were cheap enough, Peruvians might end up eating that instead of Queso Fresco. Not because they wanted to, but because they couldn’t afford not to. The free market reply is that “yes, if freely chosen, that means their lives would be better”.But back in the real world, it’s worse to live in a carbon copy of a US slum, than in a poor but differentiated part of the world.

But in most of the rest of the world, tariffs are - economically a mistake; lower and less relevant than you think; more about national pride than economics. For example, the reason why the EU has tariffs at all, is not because they really need them. It’s an emotional tie to bind them, an excuse to exist at all, and a stick to bash their own constituents into compliance “there, see what would happen to you if you were outside”. Current typical tariffs worldwide are around 5%. Come on, take a step back, how much do you think *5%* really affects which country you produce in? I just checked, whole British leg of lamb costs £18.50/kg in Waitrose, £9.52/kg NZ leg of lamb in Sainsburys. It’s a factor of two. The tariff on NZ lamb is 12%. The tariff argument for on-shoring is just bullshit. If we doubled tariffs on NZ lamb tomorrow, people who eat the cheaper lamb which has *been on a freezer ship for two months* would still do so, and the people who are prepared to pay more, would still do so.

Re: I feel liberated already...

Exactly those points, plus so many more! Classical economics is the gift that just goes on giving!

Just for starters, pricing is hugely history-dependent. You get a completely different outcome whether you change prices directly from $100 to $110. Or change them from $100 to $120 “because tariffs” and then run a sale at $110. Or from $100 to $80 to “drum up interest”, and then up to $110 once you’ve got a buzz. Or start at $100, run a “closing down sale” at $30, and then try to sell at $110 (I suggest you don’t try that).

In fact, any small business-owner will tell you that if a business-line is just completely failing, dying on its arse, the smart thing to do is just double your prices. So many reasons, I won’t run a seminar, but “scamming the customer” is not one of them.

Re: I feel liberated already...

I think it’s a shame that Tesla will die. But Tesla’s work and worth is done. It’s easy to forget how things were before Tesla, the useful society impacts are:

#1 An electric vehicle does not have to be a Green muesli-eating hair-shirt. Electric motors give insane torque and power at very low complexity and weight, almost independent of the cost of the car, in a way that ICE simply does not. An EV can be made actively desirable as a drivers car. Of course, the Tesla itself happens to be a shit box Yank Tank that can’t go round corners and is glued together. But, *Ludicrous Mode* changed the world.

Until then, the image of electric vehicles were for milk floats, or poor people who couldn’t afford fuel.

#2 The existence and sales of Tesla convinced all the other car companies that they either built EVs or ceased to exist in ten years. Other companies had been sandbagging since forever, because EVs are a commercial disaster *for them*. Tesla made sandbagging no longer an option.

#3 Tesla introduced the idea that, although making cars itself is a crap 6% margin business, its a gateway to a much better business: self-driving taxis. That’s where you can start to charge 100k+ per year for a 30k box on wheels. That’s why you want to be in this business at all. Stapling panels together is shit-for-brains Motown stuff.

The fact that Tesla seems to have failed in execution of self-driving, and is now at most third-closest to that final goal, is not what they will be remembered for.

But yeah. They’re dead now; past saving.

Re: I feel liberated already...

And that’s not even the main problem!

Foxconn China assembly is a deeply seasonal business. There’s mass assembly work in the run-up to Christmas (minus X months latency, I think X=5). Outside that period, the required labour force is less than half, and is left fallow. A workforce of 10-15 million people every year go on a grand trek back to their rural homes several thousand miles away, train-hanging with knapsacks. During the on-season, it’s hundreds to each dormitory, shift-sharing sleeping space, singles only with wives/husbands 5000 miles away.

This is classic migratory casual labour; you can read the human story in any Dustbowl era book, everything old is new again. This is the *daily life* of the manufacturing worker that Trump is now importing to the USA.

The insanity, and illiteracy, is just awe-inspiring.

Re: I feel liberated already...

Elasticity means: if I increase the price by 1%, how much does the demand volume decrease by. An elasticity of 2 would mean demand decreases by 2%.

Naive classical economics textbook will tell you it should be roughly equal to the gross margin in the industry: if your input cost is £1 and you sell for £1.40, equilibrium elasticity should be about 1.4.

As anyone in business will tell you, naive classical economics is total bullshit. All its predictions are completely wrong, if you tried running your business like that you will be broke in a few years. Anyone *even using the word* elasticity is telling you they’ve read some old economics textbooks but never managed a business. In short, they’re a so-called free-market right-winger who has no experience of the free market.

Re: I feel liberated already...

Tariffs may cause imports to drop….but rarely cause local production to increase in the many cases where price is not the determining factor.

That’s only really true for raw materials or generics like steel etc, as happened in 19th century. In the modern world, the vast majority of trade is not like that. Mostly, consumption switches to less desirable things of a different type.

French consumers would not drink more Californian wine if it were cheaper. UK consumers would not buy Chevy cars if they were cheaper. US goods are designed for US consumer tastes, that’s why they don’t sell elsewhere.

Re: I feel liberated already...

Indeed, I’ve seen in several places, thats the only way to make the “maths” give the answers Trump has come up with.

It gives an insight into the way a transactional narcissists views the world. So, Cambodia, 97%, means that for every $1bn of goods the US buys from Cambodia, they only sell $30M. So Cambodia’s “worth” to the USA is only $30M, Trump thinks they’ve “stolen” $1bn minus $30M. Since Cambodia is worthless as an export market, they can “be tariffed” (yes, I know that’s not how it works) with no downside.

We learn another useful thing: as a response to the tariffs, appeasement is pointless! If UK dropped its tariffs on cars to 0%, our imports of US cars would not actually rise (because UK consumers won’t actually buy USA style of car, price is not the issue). Governments do not control balance of trade, buyers do.

Photoshop FOSS alternative GIMP wakes up from 7-year coma with version 3.0

How about GIM or GIP? GNU Image Manipulation / GNU Image Processor

Adding the P for Program is a forced acronym. How many other programs can you think of with Program in the name? It’s only there for cheap laughs.

It’s a widespread problem in the industry. I’ve worked on a major codebase where the key data structure is called SLUT. For Symbol Lookup Table, obviously. Various people pushed back strong and hard against renaming it, despite that: it didn’t meet coding guidelines; search-and-replace is trivial; other variables were routinely re-named ad hoc during refactoring for “readability”.

It’s bullshit, and we shouldn’t allow this crap. Gas-lighting that the naming is nothing to do with the standard usage of the words, is part of it being very not-okay.

Privacy warriors whip out GDPR after ChatGPT wrongly accuses dad of child murder

I do take your point that it’s even easier, and more defensible, to do this at training-time. Just classify and remove all “Persons Name” in the training data. Or more sophisticated anonymisation, this isn’t 2010 any more, we have robust procedures for data anonymisation. The issue of removing “Isaac Newton” from the dataset, is trivially solved by adding a whitelist of “famous dead people, as defined by having an Encyclopedia Britannica article”. I just don’t see this is as a Hard Problem. There isn’t really a good reason for the LLM *itself* to know or encode people’s names or info. Just use a software agent to look it up on Google like a normal person, it’s not 2022 any more.

By the way, it’s important to realise this is more an issue with perception and feeling lied to, rather than personal data *actually* being stored. The relevant version of ChatGPT has 200 billion parameters = 200 GB. This just doesn’t *possibly* have room to store actual info about any significant proportion of the population, with dozens of bytes per person (including allegedly the names, ages of children, and town of residence). Otherwise we’ve accidentally discovered God’s compression algorithm. And Llama 7b can do it in 7GB, so it’s 28x better compression…….

This one, at least, should be a trivial fix. Foundational models can simply be equipped with safety rails to prevent them emitting sentences about any named individual. Then we don’t have to worry whether that information is right or wrong.

LLMs are simply the wrong tool to search information from the internet. That is what a *search engine* is for, clue is in the name. Identify an authoritative location for the requested info, and provide a link, without attempting to pre-process the data.

If people are using LLMs as search engines, then they are fools. There are plenty of good use-cases for LLMs, this isn’t one of them. But yes, fools will use it that way, so probably safety rails are required.

Trump orders all government IT contracts consolidated under GSA

Re: ps f47

I strongly suspect the F-47 is going to be one of the most epic clusterfucks in history. Not just because Boeing, although that too.

But: what are the odds that the development of this plane, tech demos flying since 2020, just *happened* to be ready a couple months after Trump took power, so he could announce it? As a coincidence, close to zero I reckon. I’ve got nothing against the project, know nothing about it. But the timing makes me think this project has probably been an ongoing roll of problems for years, and none of the tech really works yet.

And Trump is going to force them to fly something, anything, before the next election, ready or not: “the F-47 produced during the engineering and manufacturing development (EMD) phase of the contract will fly during Trump’s presidency”

This thing is going to fall out of the sky.

Datacenters near Heathrow seemingly stay up as substation fire closes airport

Re: sure safety-critical airport functions

That’s not realistic. But two separated connections to the grid is.

Ubuntu 25.10 plans to swap GNU coreutils for Rust

Re: Speed comparisons

I’ll bet a printf formatting bug would not be welcome either…..or indeed the 10 new printf bugs identified just today alone, with 338 open bugs in uutils……

Re: So far, the Rust coreutils pass approximately 500 out of 600 GNU tests

“There's a big difference between failing a test for edge case settings on command line parameter parsing and having a use-after-free memory error”

Is there? Because from where I sit, the first is *orders of magnitude* worse than a simple use-after-free memory error.

The vast, vast *vast* majority of command-line stuff being executed, is sitting there in bash scripts. Mostly legacy, often decades old, and almost always undocumented. So now you’re telling me that across the world, bash scripts which no living human understands, are going to “sometimes” do something “slightly different” than what they did before? And they often have admin-level access, and operate on things like file permissions.

Slow hand-clap. Badly done, sir, badly done. *This* is exactly the sort of crap I was warning about. Rust people are *obsessed* with just one single type of bug, and completely miss the wood for the trees.

Time to ditch US tech for homegrown options, says Dutch parliament

Re: before they die more slowly.

Ayn Rand of course, died in poverty. To her own great surprise, she personally had no particular talents or skills that other free individuals wanted to pay her for. She lived off Social Security, and Medicare, which she claimed with the aid of a social worker. The state paid for her operation for the lung cancer, due to her lifetime of heavy smoking of her individual choice.

Basically, she was a Republican.

This one weird trick can make online publishing faster, safer, more attractive, and richer

Re: How much?

My tuppence: the problem is that one has to spend *a lot* of time on any one website to add up to even a tenner a year in value. The average Netflix user spends 3hrs on a day there….for £72 a year with ads or twice that without. You’d have to spend half an hour per day on ElReg to have the same value proposition, and that’s not likely.

I would think, more like an aggregated club of websites, subscribe to one, get them all, everybody gets a fraction, makes more sense. This has been the endgame for TV. The problem is this soon becomes Nokia walled garden / Spotify / Disney+. What happens is that the owner of the umbrella brand gets all the power. However, it is a real alternative to ad-based funding of the internet.

AI running out of juice despite Microsoft's hard squeezing

Re: They gave the AI scam their best shot, but...

It *does* pass the Turing test, rather easily. But, it turns out that metric of “how good are you at giving plausible but incorrect impression” is not a good one…..

Microsoft quantum breakthrough claims labeled 'unreliable' and 'essentially fraudulent'

Re: Only the beginning...

“the big two journals…..Science and Nature”

Perhaps I have just worked in the wrong fields, but I don’t think any field/sub-field has either of those journals as its “journal of record”. They’re more for splash-publishing to the wider science/public/journalism world. They are properly peer-reviewed and the papers are sensibly written, not Pop Sci, but it’s not where the “real work” is done. Every sub-field has its own premier journal, and the famous two just can’t be that.

I’ve also noticed that over the years, there is in fact a rather high % of fraudulent stuff there which has had to be retracted. Partly because OMG Breakthrough features heavily, and such stuff turns out to be fraud nearly as often as not, because incentives. Sign of the times.

Surprise! People don't want AI deciding who gets a kidney transplant and who dies or endures years of misery

Re: Who dares judge the value of another?

The alternative is just to flip a coin. Between two or three people for whom, honestly, the difference in predicted outcome is insignificant. The evidence base and accuracy for those outcome predictions, is pretty high error-bar.

I actually agree with you, we shouldn’t allow value judgements into a technical decision. Just, note there really isn’t any actually objectively-correct answer.

If it’s done that way, ok. I might even agree that “deserving” judgement of the recipient should be ignored.

But the coldness of writing down the criteria, and then having those judged and reviewed by your medical peers, leads to a certain “political correctness” in the criteria. Nobody wants to put in writing that they wouldn’t give the kidney to the convicted murderer. Even if in fact, at 3am with nobpdy else to judge them, they would just pick somebody else.

Re: Medical work by people has always been better

>>“It essentially transforms the question of transplant recipient selection into: Do you feel lucky, punk?”

I have no specific experience of transplants, but yes my experience of *all* healthcare is that the primary determinant of your outcome is luck.

Do you have the heart-attack at 1am on a Friday night when the clubs are chucking out, or at quiet time just as the consultant is doing their round.

Is your second life-changing diagnosis orthogonal to your first, or is the first-line treatment excluded by the set of drugs you are already on.

When 111 took the history, did you understand the key question and give the correct coded answer as per textbook, to go down the correct treatment pathway. Or did you use the wrong codeword, and get put on the non-urgent pathway.

It’s all a dice-roll.

Re: But why should we trust humans with moral decisions?

You wouldn’t *tell* the AI to do anything. You feed the AI with thousands of similar ethical decisions, questionnaires given to a few thousand people. Based on that dataset, the AI autocompletes the most likely response, in the actual situation. In other words, it’s democracy, with all the human biases of society. The question is, do we believe in democracy of the common people, for this purpose? Or do we train it on the responses of medical professionals? Or of medical ethicists, or who?

The problem as always in real life, if you have to ask somebody’s opinion, you’re pretty much deciding the outcome yourself, by your choice of who to ask.

Re: The main cause of waste is time

The issue is that your decision process is *still* massively loaded with moral judgements, which some may agree with, some not. Your key moral assumption is in step #3 - that the only relevant factor is the probability of success of the operation itself. Bearing in mind that way more people need kidneys, than are available. So, just picking some of the other criteria listed in the study, are *you* are clear that:

You would prioritise a 75year old who also has liver failure (match/success proba = 80%), over an otherwise healthy 30yr old (with match 70%)?

You would prioritise an unmarried person (match 80%) over a single mother with three young kids (match 70%)

You would prioritise a convicted murderer (match 80%) over a pioneering doctor (match 70%).

ASML will open Beijing facility despite US sanctions on China

Re: And so the end run around merrica begins...

Maybe…. but I wonder if you are clear what you actually mean by US-controlled IP….or if it’s more of that “ARM / x86 ISA royalty” nonsense.

To prototype any new CPU design, you need FPGAs, lots of them. And for complex CPUs you need really high-end ones. You just can’t do this fully in simulation, it’s orders of magnitude too slow. There are only two viable FPGA vendors, and both of them are US-owned.

All the vital chip-design EDA software vendors, like Synopsys, Cadence, Mentor Graphics, Keysight, Altium are all US-based. Whether its synthesis, place-route, power analysis, clock trees, power domains, computational lithography, DFM / DFT / reliability. It’s all US software. And it’s all *annual licenses, seat-based*, none of it is “bought and owned”. The open-source stuff is just toys for university students.

There are certainly stages in the chip-design / manufacture that are *not* US-bottlenecked. But enough of them *are* that EU cannot make a chip without US agreement without at least 20+ years R&D.

Similarly with the various sets of expertise out in APAC: not just the silicon fabs, but test, advanced packaging etc.

What EU excel in, is the most complex silicon fab tools, a fantastic high-margin business, selling in units of only dozens of machines worldwide, but each one selling for tens or hundreds of millions. They lead the world in that. Not just ASML, but various Austrian, Swiss and a couple German companies.

Look, it’s an incredibly complex global supply chain. Every region has its specialism. This tariffs/protectionism stuff is literally suicidal for all concerned. The EU trying to make sovereign chips, with the other regions actively sanctioning, is laughable. But so would be if the USA tried. And so would be Taiwan, or China, or South Korea. It just *doesn’t work that way*.

Re: Huawei Euv?

ASML is a Dutch company. The EU Commission wrote and enforces these sanctions, not USA.

Scotland now home to Europe's biggest battery as windy storage site fires up

Re: Back of the envelope

“Many” more of these, indeed, to provide actual full backup for fully wind and solar (zero fossil fuel).

About 20,000 Dinorwigs in fact. That’s for the electricity usage we have today, it doesn’t include what we would need for the much grid demand when all vehicles go electric; and all home heating/cooking goes electric.

And in fact when the hotter summers mean that UK homes start to install air-conditioning to stop people dying. Remember, those heat pumps you’re telling people to install instead of gas boilers, all run in reverse as air-conditioners. This is great, a genuine Green Dividend……except how are you going to stop people actually *using* them?

Re: Back of the envelope

Standby plants still need to be built and maintained, and all those costs are borne by the increasingly few GWh they produce. This has four problems:

#1 It just plays into the hands of the environmentalists who want to claim fossil fuels are expensive. They aren’t, they are just being the patsy for transfer costs from Wind.

#2 You’re assuming that companies will be prepared to keep gas plants on standby, by being paid subsidy from govt to do so. The lesson from coal, is that they *are not prepared to do that*. Turned out there was *no* price that govt could pay, which companies were prepared to accept to keep coal as standby. They simply decided that running a shell was not a business they were interested in, and shut it down against the protest of govt

#3 At some point, maybe 20 years from now, those empty gas-fired shells running 30 days per year, are still going end-of-life. They will shut down and need to be replaced. New gas plants will need to be built, for the sole purpose of running for 30 days per year. Hands up if you want to bid for the contract of building and owning that? With zero guaranteed running time = zero revenue in any particular year. Yeah. Me neither. So how exactly does this “sustainable” future work? Not very sustainable is it. It only works by living on your parents trust fund.

#4 These gas plants, which will be running for 30 days backup per year. Where are they getting their gas from? In todays world, there is a constant pipeline from wells running 24/7, througn refineries running 24/7, 500,000 tonne supertankers which take 30 days to cross the world, and re-gasification of LNG at receiver ports.

In your intermittent world, the entire world fossil fuel infrastructure also needs to start up and shut down on a dime. Except, it’s got an end-to-end latency of *months*. Tell me, how do you predict the wind months ahead, to turn the whole industrial infrastructure off and on again? This does not work. Its toy model economics.

Eutelsat in talks with Euro leaders as they mull Starlink replacement in Ukraine

Re: Eutelsat has been memestock in the last week (+650% at one point)

And for the same reasons. There is no underlying ongoing business there. It’s easy to quadruple the price of something essentially worthless, because it doesn’t take many fools to multiply a near-zero market cap.

When you say +650%, a better perspective is “has spiked to the dizzying heights of mid-2023”. Half what it was in 2021. Down 80% on 2015. Pre dead-cat-bounce, it wasn’t quite in the 99% club, but very very close. This parrot is no more, it has gone to join the choir visible.

Re: Hmm

True or not, the big problem is fhese would be reasons why *Russia* should stop.

But “Russia” isn’t in charge, Putin is, and he doesn’t really care.

Re: Sanctions

BRICS. Brazil, Russia, India, China, (South Africa)

You are going to impose sanctions on Brazil, India and China, to punish Russia? You have gone stark staring mad.

Re: Hmm

No, it isn’t. You may hope, but this is what an attritional stalemate looks like.

UK must give more to ESA to get benefits of space industry boom, says Brian Cox

Re: Um....

That depends which space projects you mean. If you mean *science experiment* projects, then yes. The scientific community does not need two GAIA spacecraft, or two LISA interferometers. By banding together, we avoid pointless duplication; can afford bigger experiments; and mostly importantly of all on the science/analysis side participate in a nexus of where the best and brightest go. The Camelot effect. Also, worth bearing in mind that on science projects, most of the *value* is not under ESA funding or auspices at all, it’s the university Principal Investigators with (national) R&D funding. That’s the positive side.

On the other side, the engineering side, ESA is a catastrophe. Not only do you not “need scale” for telecoms and earth observation R&D, it is the exact reason why Europe has failed. When you have to “ask permission” to do developments, you prevent the actual doers from just JFDI. Nobody in either Airbus or Thales (the two European primes) has *any interest* any more in whether the systems they build are commercially viable. The only purpose of these programs (in their minds) is to get paid by ESA for doing an R&D. If you are in any meetings, the strategic purpose of one R&D program is simply positioning to be in the best place to win the next R&D program. Boondoggle programs like EDRS-C become entirely self-sustaining.

ESA’s programs performance on value for money (less than 10% of the money spent by govt goes on actual engineering), and ridiculous industrial structures (georeturn) is well-documented and I won’t rehearse it too much here. Other than to point out, that in a *normal* economy, projects get done by the companies with the *greatest* capability. But in the ESA-funded world, the main criteria are to put the workflow in the hands of those *least* competent to do it, in order to “develop the capability of the industry”. That’s not my sour-grapes belief, it’s literally written down in black and white in the contract Statement of Work.