Posts by Tim Worstal 1389 publicly visible posts • joined 12 Feb 2008
And I appear. Umm, yes, OK. Ata guess you need one atom to make such a clock? Which would make the lb of Sc I've got floating around in the souvenir box enough to make one for ever El Reg reader?
There's the Xmas offer sorted then.
BTW, Brexit was great, Yah?
"According to The Financial Tribune, an English language news publication focused on Iran that's operated by Tehran-based Donya-e-Eqtesad,"
I used to write for Donya. Fun gig - tho' my co-author has recently fled the country to avoid being hanged. Free market economics became not stylish. Despite my having written for them they're an accurate source.
"Maybe someone is confusing mass of ore with mass of refined Lithium? Dunno, but you're right, the numbers don't match."
No, this is OK. 1.6 million in 2025, up from 130k tonnes last year (numbers in the USGS 2023 report refer to 2022). Could be an over-estimate of course, but lithium is seeing soaring demand.
"Berkshire Hathaway Energy Renewables project, based in Imperial County, California. The project aims to extract lithium from geothermal brine"
Yep, it'll almost certainly work too. The membrane tech to make it so is good to go. Which also means that the Rhenish geothermal stations become a source, Cornish Lithium (and, umm Wearside?) also become available. The guys claiming to be able to extract direct from the Red Sea might be a bit hopeful. It's definitely extractable, but at what cost?
" There's another use of Lithium. A naturally occurring isotope of the element – 6Li – is a key ingredient in the fusion fuel of practical thermonuclear weapons. We mention that because Iran is so very keen on developing its own nuclear weapons."
And a rather fun bit there. Chile has those vast salt flats that are just stuffed with lithium. But there are only two licences to extract. The problem being that the nuclear ministry has to authorise any new licences. Which it never has. The reason being that when Pinochet rewrote the mining laws someone thought about Li-6 and inserted that requirement into the law. The effect of which has been that Pinochet's son in law (at, I think, SQM) has never had to face the disturbing possibility of competition in Chile. Because his licence was grandfathered in.....
Yer Wha? SuperBock is under fire? Eeek!
'Tis, of the beers here, mine of choice. As with the spread of lagers more generally it's really much like Dos Equis, etc, etc, around the world in origin. Germans and Czechs - some argument over that - worked out mid-19th how to make lagers and bottle/barrel them. Able to chill and brew year 'round (that Oktoberfest thing is about not being able to brew on the plains in summer, so let's celebrate the first deliveries of the autumn from the hillsides down to the plains).
So, once this was worked out likely lads spread out around the world with the new tech. And all sorts of countries have a lager beer brand, which is really just which likely lad arrived and prospered.
There may be more complexity to this story but that's good enough as an opener.
Portugal has two which have survived in large scale, Sagres (named after the city) and SuperBock. As I say, I prefer the second, But a little advertising story. One SuperBock ad a few years back was an opinion poll they'd done in Sagres. Something like, from memory, "The majority of people surveyed in Sagres said they preferred SuperBock". Which is very fun and most amusing.
I never did follow the story for long enough to find out how many times they'd had to conduct the survey to get that result.
The AI version costs tens of millions to train to a point. It becomes less valuable day by day after that.
OK, high capital cost, v low marginal cost of production. We know about things like that. What we end up with is some, few, of such expensive to produce Big Things producing output for many, many customers. You hope that the difference between marginal price collected and marginal cost to produce pays back that capital cost. Or, average production cost is well above marginal until some massive scale (average will, asymptotically, approach marginal as overheads are paid for as scale increases), you need to sell a lot to make back the investment.
This isn't that much different from the economics of search engines. Or even social media. You get very little from each user of Google or Twitter (G gets maybe $10 a year per user?) and you've costs in the billions of providing the product at any level to anyone.
To make it work you've got to sell the shit out of millions of pieces of output that is. And, as with social media, advertising might well be the way to pay for it. Because that way you can give it away "free" and thereby gain the scale of pennies and pennies that pay back the billions.
1) India's cars don't contain rare earths. Cars that do are those with catalytic convertors and or rare earth magnets in the stereo etc. Which isn't the sort of car they're talking about scrapping.
2) That's not what he said either. Rather, that India can now recycle rare earths and it's also going to have a car scrappage scheme. Two different things, not the one being part of the other.
3) India does have a rare earths industry, a small one. They have the mixed concentrate (from varied mineral sands operations) and a small separation plant.
4) The biggie. You can't have a circular economy when you're building a civilisation the first time around. Because you're not scrapping enough to provide the scrap to build the civilisation. Sure, you should feed old cars into steel furnaces. But you want steel for 1 million (to invent a number) out and you've only got 100,000 cars a year to put in (to invent another number). You've still got to go get the steel for another 900,000 cars a year from somewhere.
This is also true of rebar for concrete buildings, steel and aluminium for windmills, anything in fact. It's only when you're tearing down civilisation 1.0 (or 2.0) that you've an excess of scrap over demand for new rather than a deficit. So, Civ 1.0 requires more raw material than local scrappage can produce. At which point, why not import scrap from people doing the Civ 3.0 bit instead of using iron ore to make virgin? As, umm, India does right now.
The advanced and rich countries are, largely enough, exporters of scrap metal. The currently poor countries are, largely enough, importers of the same stuff. Which is as it should be. Because a poor place doesn't have enough old stuff being scrapped to produce the Civ 1.0 while tearing down Civ 2.0 to create Civ 3.0 produces an excess of it.
5) Finally. Anyone worried about running out of rare earths is insane. There's a shortage of separation plants, that's true enough. But then you need one of those to process scrap anyway. There's absolutely no shortage of rare earth concentrates out there at all. It's dollar a kilo stuff (as opposed to seaprated, wihch might be $40 a kg stuff) wihch isn't the priec of something in any form of absolute short supply.
There's something called the "Gifts to the United States" account. Think you're not paying enough in taxes? Send them a check. Raises about $4 million a year.
They also send thank you letters back. My test for someone demanding higher taxes. Can we see your thank you letter?
I once got the similar numbers out of the UK Treasury. How many people pay more tax than they need to? Or, more accurately, how many send in a cheque for more? In 2005 it was 5 people - and four of them were dead leaving bequests.
The number of people who believe so strongly in higher taxation that they pay, themselves and voluntarily, more taxation is remarkably small.
"Singapore to treat infosec as equivalent public good to fresh running water"
It's not a public good. Sure, it's good for the public, it's a good that is provided to the public but it's not a "public good". Which is something that is non-rivalrous and non-excludable. They can indeed turn your water off, it's excludable. And - barring that recycling - the water you've just drunk is not available to the next person, it's rivalrous.
The absence of pandemic cholera that results from a decent water and sewage system is indeed a public good. But the water itself isn't.
Yes, this is more than mere pendantry. Sure, infosec can be a state provided good to the public. But it's not going to be a "public good".
As you say, regulations must be passed into domestic law. In the UK breach of this one carries a penalty of up to 6 months in jail and or a fine of up to £5,000. As with the jams, jellies, marmalades stuff and so on.
And you're specifically missing the point I did make. The problem is not that there are standards. The problem is that the standards become law.
Industry standards are industry standards, criminal law* is criminal law, the two are not the same and shouldn't be.
Which is actually the base complaint about the EU regulatory system. It's mixing those two things which should not be mixed.
*Yes, it's criminal, not civil, because you can't go to jail for civil things.
No, EU law applies across the EU. It's just that the UK - with a small assist from Ireland - is the only common law jurisdiction. Therefore the stupidity of this sort of detailed legislative regulation is more obvious to us than to those in the generally Roman system on the continent.
Apart from anything else we generally believe that anything important enough to be actual law must be obeyed. A charmingly civilised idea but not one that stands up to centuries of fonctionnaires.
He's not Tim Worstall, no, but I am.
Bendy bananas.
It is actually true that the law stated that Class I bananas for direct human consumption must be free of excessive curvature. Breaching this was a criminal offence punishable by up to 6 months in jail and or a £5,000 fine.
That's not a myth at all.
The EU's defence of this was as follows. There exist industry standards as to what is Class I,Class II and so on. There are industry standards on all sorts of things, obviously enough. I even wrote one myself, the standardised scandium contract for the Minor Metals Traders Association.
Some part of the UN collects all those from the food industry into the Codex Alimentarus. This is entirely sensible. People trying to get into the industry can go look the standards up. Great.
The EU then went that one step further. They said that the industry standards must become, in detail, law. This is very stupid indeed.
Using the common law approach would be sensible. "If you say you're Class I 'nananananas and you're not then that's lying in trade which is an offence" and why not?
Saying that, "in the law, here's the entire and total definition currently used by industry and it's a criminal offence to do anything different" is stupid. Say people want to start eating not Cavendish but some or other of the hundreds of types of 'nanas? Which have more curvature? Say someone even just thinks about importing some to see? It's now necessary to go change the law in 28 countries plus a number of devolved administrations. Yes, the law must both be in Welsh and passed by the Senned.
What have we just done therefore? We set in stone the regulatory system as it was when written. We've entirely - because changing the law in 28 countries just to try an experiment and see if there's demand isn't going to work - closed off any adaptations, changes, moderations in those regulations.
The common law approach - if you say you meet Codex Alimentarus standards then you should - means that experimentation, growth, etc are possible. Because they can be changed just the once. Worth recalling that at some point, as a clone, the Cavendish is going to go the way of the Gros Michel and we'll all be eating some other cultivar.
The transposition of industry standards, in detail, into law entirely removes any adaptability. And we live in a world of changing tastes, changing technologies, therefore adaptability is the one thing that we must have.
Bendy bananas is actually a perfect example of the problem with EU regulation. It really is true that bendy bananas were a criminal offence. Then there's the claims from the EU that oh no they weren't. Followed by their complete incomprehension of the actual problem. We shouldn't have detailed, written, rules and laws at this level of detail. We need to regulate at a higher level of abstraction because we must have more adaptability in those details than using the legal system allows.
As I say, the perfect example of EU regulation. And why we must leave of course. Having the anal retentives writing the detailed rules which govern an economy of 500 million people just doesn't work over time. Therefore we shouldn't do that. The EU does and always will - Ceterum Censeo Unionem Europeam Esse Delendam.
"And in case you're wondering a "keiretsu" – which Byrne refers to repeatedly in the press release is a Japanese term meaning a group of companies that informally work together. It is often, though not always, equivalent to what Western nations would called a "cartel.""
More like a group of companies with shareholdings in each other and thus informally coordinated. They are often - usually - in wildly different industries, so it's not a cartel. Say, "Mistubishi Shipbuilding" and "Mitsubishi Cars" and "Mitsubishi Electronics " (I am only certain of the existence of one of those) as making up a Keiretsu. As opposed to a cartel, which would be more like "Mitsubishi Cars" and "Honda Cars" and "Nissan Cars" working together.
"Indium is going to run out in 20 years if we carry on using it the way we do," said professor David Cole-Hamilton, vice President of European nonprofit EUChem"
Yes, that's complete bollocks.
We don't have any recorded reserves of indium at all. From the USGS:
https://prd-wret.s3-us-west-2.amazonaws.com/assets/palladium/production/s3fs-public/atoms/files/mcs-2019-indiu.pdf
And we haven't had any reserves either. And yet we still get indium each and every year.
It's almost as if Wikipedia is correct here:
"Thus, the availability of indium is fundamentally determined by the rate at which these ores are extracted, and not their absolute amount."
*Sigh*
Very slight correction. Titanium's pretty easy. Well, OK, but to explain.
To get to titania, titanium dioxide, is pretty simple. That's why it costs $400 or so a tonne. Take ilmenite (maybe $50 a tonne, thereabouts) and dissolve in sulphuric. Or, alternatively, add coal, chlorine, and light. Nice exothermic reaction - thus self sustaining - and TiCl is a gas at these temperatures, you can distill it out of the other stuff. For the varied gases solidify again at different temperatures, have a long tube that gradually cools and each one will pop out of the little holes along that temperature gradient. Convert back to TiO2 once separated. Used to be one of ICI's big things, plant still exists - or did 5 years ago - up on Teeside somewhere. Hunter Chemicals now maybe?
OK, a bit more complex but not really. But the point is it's pretty cheap. That's why TiO2 is used for simple and cheap stuff like the white in white paint.
Converting the TiO2 to Ti metal, no, that's expensive. Last I looked, years back, nice Ti metal was $14 to $20 a kg.
Units of Ti are easy and cheap to get. Getting them into the form you actually want can be expensive....
Slight technical point. The REs don't in fact preferentially concentrate in the fly ash nor clinker. Pity but there it is. The germanium concentrates beautifully in the fly ash which is why we get 50% of the world's supply from that source.
Ash will have, around and about, whatever concentration was in the original coal minus the carbon. Which isn't quite enough to produce a viable concentration to extract. A number of US schemes testing this out at present and absent some environmental case - say, cleaning up ash ponds - lifting some of the cost it doesn't work.
"I really wish El Reg would let Tim Worstall re-explain his rainbows and unicorns once again in a guest article. Smart guy, but one wonders in this case."
I just think the EU is a terrible system which no one should belong to, let alone us. Entirely an arguable idea but that's where it all stems from. Centralised control of the lives of 500 million people just doesn't seem to work for me.
Chacon a son gout, obviously.
All that's been said about REs so far seems fair enough. It is true that finding a deposit large enough to make it worth exploiting is tough.
But this is to do with the separation process.
Yes, separating the 15 from each other (S and Y are easy with chemical means) requires a girt big plant. Last one built cost a $ billion at least. There's really no great way around this, it just has to be large to cover all the steps. And such a large plant wants a homogenous feed. Thus, natural size of the processing plant is girt big, desiring constant feed, deposit must be girt big.
At which point the conceptual error. Or, OK, the possible error and agreed, this is a little bugbear of mine own. Change the processing method. There are alternatives that are possible. Might even be economic too. No one's really done all that much looking in recent decades - since the, say, 1960s really.
I can think of one method that does work - whether it's economic I don't - and which could be set up to take hetergeneous feedstock. So, we could do the extraction in small lots from all those waste piles and off-feeds from other processes and so on. Which neatly solves our problem of a lack of large scale deposits worth pushing through a traditional plant.
All I need is the odd £20 million to go do the research....oh, all right, I'll test it in a lab first for £50,000 if you insist.
"I get the feeling Tim vanished because..." El Ed desired to spend his editorial budget - as is his absolute right - in a different manner. No complaints from me, that's how the industry works.
Fun discussion of a recent paper:
"Children Hurt Women’s Earnings, but Not Men’s (Even in Scandinavia)
Motherhood is the biggest cause of the gender pay gap. It might take fathers to change that."
https://www.nytimes.com/2018/02/05/upshot/even-in-family-friendly-scandinavia-mothers-are-paid-less.html
Actually, children seem to raise men's pay. In a sexually dimorphic species the arrival of children produces asymmetric responses across genders. Remarkable that really.
"A major study of Americans has punched another hole in the official British government medical advice that there's no "safe level" of drinking.
The cohort study of around 100,000 individuals found that infrequent drinkers and teetotallers had an increased risk of mortality."
That official advice reached its conclusion by the interesting tactic of not recording the outcome for teetotallers at all. Thus only showing the increasing risk from increasing consumption and neatly excising the higher risk of no consumption.
No, really, that is what they did. The bastards.
"Yes"
"It would be really interesting to see Worstall's opinion now that the 'leave' vote has been hijacked by zealots who really want to walk away completely. I'd be genuinely interested if he still thinks the UK will be better off crashing out with no deal, than it would just staying in the EU."
I'm one of the zealots. Leave, get out entirely, allow them to sail off to perdition as they so obviously desire. I have no desire for us, nor anyone else, to stay in an organisation that can do something as stupid as the eruo.
There's also been a study on this. Patrick Minford. Revert to WTO rules, our exports to them face those tariffs. We decide to go for unilateral free trade. We don't charge ourselves tariffs upon the things that we desire to import. Essentially, what we did in 1846, Corn Laws and all that.
Net effect is to grow the UK economy by 3%. Yes, grow.
But I really am a zealot. I think the very existence of the EU is a bad idea, that everyone should leave it. And I believe that near whatever the economic arguments about it. Proper, full on, zealotry I'm afraid.
All most kind but I fear I'm a touch unfashionable under the current editorial dispensation.
If I were to write here on this I'd add in these three facts:
1) We only need two numbers to explain all of the observed gender earnings gap. Mothers make less than non-mothers among women (about 9% for the first child, lesser extra amounts for each subsequent). On average, of course, with all these numbers. Fathers make more than non-fathers among men, about 8% or so. Yes, always controlling for all other factors like age, education and so on.
Sexually dimorphic species - one that started out as hunter gatherers with sexual division of labour - has division of labour in child rearing. Really?
2) The stat being used is of all men and women, part and full time, within each company. Back a decade Harriet Harman and the Fawcett Society started bandying about the pay gap of this unadjusted form, part and full timers together. The Statistics Authority, in the form of Sir Michael Scholar, wrote an open letter insisting they stop. To blend in this manner was extremely misleading and more likely to confuse than inform. One can and should use part time to part time, full time to full time. So, we then get a law insisting reporting is done in the misleading manner, do we?
3) Consider what must be true if this is about discrimination. Women are cheap compared to their skills, talents and output. It is therefore possible to make a fortune by discriminating in favour of hiring that cheaper female talent. Dame Steve did exactly this in the 60s and did make that fortune. No one is doing this today. We must therefore conclude that it isn't about discrimination, even that women are not being underpaid for their output.
Then, the bit in the next comment about women who might have children but don't getting lower pay. Actually (and this is just the way the stat was collected) never married childless women in their 40s enjoy a - small to be sure, 1 or 2% - pay premium over the average male. Lesbians do too, presumably something to do with the lower incidence of children. Interestingly, gay men have a pay gap against them in reference to hetero men. Quite possibly that influence of being or not a father affecting the averages.
Give it a bit more time and we'll have a really interesting piece of research that can be done. Same sex coupledom is clearly becoming more common, as is such same sex couples having children. We will therefore be able to study a population where gender (or even sex) is divorced from primary child carer entirely. Be great fun to see what the pay gap is then.
My bet is that primary child carers would face about the same gap as women do today. Meaning that the gap is about primary child carers nowt else. But then we all do hope for confirmation in the future of our own assertions today, don't we?
Take your point, yes, but that's not what is being alleged.
https://www.justice.gov/opa/pr/nine-iranians-charged-conducting-massive-cyber-theft-campaign-behalf-islamic-revolutionary
"The members of the conspiracy used stolen account credentials to obtain unauthorized access to victim professor accounts, which they used to steal research, and other academic data and documents, including, among other things, academic journals, theses, dissertations, and electronic books. The defendants targeted data across all fields of research and academic disciplines, including science and technology, engineering, social sciences, medical, and other professional fields."
I do read that as breaching the academic publishing paywalls more than anything else.
"If 30+TB of data cost $3.4B this implies a cost of about $1 per kB"
That's the DoJ lying through their teeth.
Their claim is that the US universities paid about $3.4 billion for the information. Most of which was actually subscriptions to academic journals, the papers from which the Iranians downloaded.
Loss to the universities, zero. They've still got all the information they started with, still got their subscriptions and haven't paid any more for them.
Loss to the publishers like Elsevier - not, clearly, the nominal cost of the downloaded papers. But rather the loss of revenue from those papers which would have been paid for in the absence of the data theft. Obviously, some sort of sum there but it's not $3.4 billion. Tending towards $0 would be my best guess.
Wealth and income is one of those difficult problems to which there is no correct answer. Take our footballer, makes a mint for a decade, then lives off it. If we tax both wealth and income then that pattern of lifetime earnings pays more tax than earning the same total amount over a lifetime but at a lower annual rate.
Is that a fair way to do it? Dunno myself but it is a problem, one with perhaps no right answer.
My own prejudice (and it is prejudice, this is not an economic point, it's a gut feel) is that people on less than median earnings should not be paying income tax. Pigou taxes (carbon, congestion etc) yes, sin taxes yes, VAT even, but not income taxes. Those on above median should be paying for society. There's a limit on how much you can get out of them meaning that government will be smaller than it is now. But, you know, there's prejudice for you.
Re Trump's tax bill. Some bollocks in it, the cut on pass through entities is bollocksy in a bad, bad way. They already pay much less than other forms of business organisation. The personal tax cuts being time limited is a result of the silly way they measure. It's the effect upon the deficit in a decade's time which is used as the benchmark. Thus tax cuts (like some of Bush's) have a roughly decade sunset clause. The clause to benefit Trump - weeeell....that's much more like making sure than an extant break (depreciation on buildings and yes, buildings do depreciate, even if land doesn't) stays on the code rather than inventing something new.
Good stuff too. Doubling the standard allowance takes many more on less than median earnings out of the Federal tax net (which is largely, not exclusively, taxes on incomes, sales and property tax are State or even more local) which accords with my above prejudice. Limiting the mortgage deduction is good. Reducing the state tax allowance is good (for you can deduct your state and local taxes from your taxable income, thus saving your marginal Federal tax rate on what you've paid more locally. This means that high tax states are being subsidised by lower tax ones at the Federal level, shouldn't be happening at all. People want a high tax state in NY? Great, you pay for it, why should Kentucky?).
All of those above are unremittingly good.
There are good arguments either way about the corporate tax cut. The US rate is waaaay too high (even if we're going to commit the mistake of thinking that companies pay tax, rather than the true mixture of shareholders and workers in some portion). You have to add the dividend tax rate (15%) to the corporate income tax rate of 35% (because dividends can only be paid from tax paid profits, that offshore cash etc cannot be used) and standard economics says returns to capital should be taxed lower, not higher, than labour income.
But what's really interesting here is that no economist - not even Paul Krugman who really doesn't like the idea at all - is saying that it won't boost growth. It will. Absolutely. The argument instead is over "How Much?" Some trivial amount we don't care about while the plutocrats make out like bandits (copyright P Krugman) or a substantial amount that will benefit all (Trump and friends and a few others but not many).
If the state and mortgage deductions had been abolished overall I would have supported this as being excellent, whatever else was going on. Limiting them is a benefit but, you know, Meh for the overall.
"On the Soviet side, the communist state was so paranoid that earlier that month it had shot down a Korean passenger jet that had accidentally wandered into Soviet airspace,"
I spent a decade as the business partner of the nephew of the bloke who ordered that. Mixed bag, Soviet military types.....
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
