52 things Tom Whitwell learned in 2018

Here is one of them:

35% of Rwanda’s national blood supply outside the capital city is now delivered by drone. [Techmoran]

Here is another:

Advertisers place a single brown pixel on a bright background in a mobile ad. It looks like dust, so users try to wipe it off. That registers as a click, and the user is taken to the homepage. [Lauren Johnson]


Those weirdly expensive books on Amazon could be part of a money laundering scheme. [Brian Krebs]


Expensive placebos work better than cheap placebos. [Derek Lowe]

And if you ever doubted it:

There is a small but thriving startup scene in Mogadishu, Somalia. [Abdi Latif Dahir]

Here is the whole list, definitely recommended.  Via Anecdotal.


Expensive placebos work better: one has to ask whether the placebo effect works across health care (expensive hospitals get better results, expensive physicians get better results, etc.). How about expensive lawyers? I once knew a lawyer who would charge new clients relatively high fees because they not only expected it but were happy to pay high fees because the high fees were an indication that he was the best lawyer and they only wanted the best. What about high priced bankers, high priced economists, etc? Is this a market failure?

"Expensive placebos work better than cheap placebos. "

You get what you pay for, although this data point suggests it is (in some cases, at least) as much a pull factor as it is a push factor.

I don't buy 26, 32, 34, or 48.

Is that after reading the links provided?

I do wonder about 26: "Men who’ve experienced earthquakes are willing to take more risks and gamble more. Women show no such effect."

I experienced several fairly severe earthquakes while living in southern California.

But there weren't as severe as the one that hit Japan. And it's possible that the Japanese might change their attitudes towards risk differently from how Americans do.

So one wonders what would happen if someone tried to replicate the results using say the 1994 Northridge quake. Were Angelenos who lived in the more shaken-up areas more likely to travel to Las Vegas post-quake (as opposed to traveling to Arizona or the Bay Area)?

"US nuclear testing between the 1940s and 1970s may have killed ...": I liked the "may". The article doesn't tell us how the economist linked radiation levels to deaths. There are those who think the zero-threshold linear model probably substantially overestimates the risks of low level exposures. So the sceptical reader might assume that the economist will have adopted it since it will give the most satisfyingly high predicted death tolls.

(I know of no doubt about the risks of high level exposures though maybe I'm out-of-date on this; perhaps Chernobyl will have lessons here. )

Yeah, the economist's article is bunk.

Meyers claims that all of the bomb testing " exposed millions of Americans to harmful radioactive material" which is only a statement made by someone using the obsolete LNT hypothesis even if through secondary sources.

If you look at surveys of health physicists, 85% say LNT should not be used and those experts in opposition grew over time. After the accident at Fukushima, UNESCO finally discarded the LNT in 2012.

Meyer's second error (although the first is sufficient) is when he writes: "[Simon and Bouville] estimate that fallout from domestic nuclear testing caused 49,000 thyroid cancer deaths." But Simon and Bouville state the testing, based on the incorrect LNT model, caused 49,000 cases of thyroid cancer, not deaths. Thyroid cancer has a 99% 5 year survival rate and essentially 100% for children.

In 1990, thyroid cancer appeared four years after the Chernobyl accident in about 6,000 children, all who drank highly contaminated milk and a few died among those where the cancer was detected late.
(As of 2007, about 50 to 60 people have died prematurely due to Chernobyl.)

I know nothing about this stuff. Are cancer and other radiation related problems in Japan far less than most expected?....or far less than common assumptions would have predicted?

The American Health Physics Society expects no deaths from radiation due to the Fukushima accident. The NY Times, which had poor and hyperbolic reporting, quoted "nuclear experts" including one who quit college as a music major to join the anti-nuclear movement. The Times also quoted an anti-nuclear physicist, Hippel, who was in his 70s at the time, that there would be hundreds of permature deaths, but he used the LNT hypothesis.

A Berkeley physicist, Richard Mueller, went against the LNT model and against the Health Physics Society and decided anyone with exposure found at Denver or higher was at risk, despite no evidence for higher cancer rates there where radiation is 3 to 4 times the exposure of most background radiation in the U.S., and concluded 100 people would both get and die of cancer. But half of those cancers can already be cured so he should have written 50 deaths by his cut-off. Yet Mueller really should have concluded "almost no deaths" because those cancers wont appear until 2040 when the cure rate will almost certainly be way higher than today.

With the caveat that I've not kept up with this stuff, the deaths caused by radiation at Fukushima are probably about what I expected, somewhere between bugger all and damned few.

Shortly after Chernobyl I was briefly pressed into the service of Her Majesty's Government - unpaid, I may say - performing part of an emergency study of the reactor and associated operating practices. Jesus - if you want an example of why you should have nothing to do with socialism ..... Anyway, to repeat myself, I haven't kept up with the field. I already knew, however, that the standard tropes about the dangers of low-level radiation erred on the high side. I assume that's caused by a mixture of ignorance, dishonesty, and hysteria. Also, no doubt, by Soviet propaganda - the effects of which might well linger for many decades. (What's the half-life of Soviet propaganda, eh?)

In spite of those remarks I would myself recommend against building nukes on land exposed to regular earthquakes or vulnerable to tsunamis. So, for example, Germany OK, NZ not OK.

'As of 2007, about 50 to 60 people have died prematurely due to Chernobyl'

Including or excluding the people who actually died as part of the incident and its immediate aftermath?

That includes everyone. The WHO put out a report on Chernobyl in 2005 that concluded 4,000 people will eventually die prematurely, but that is far too high as they used the LNT hypothesis. It is possible that over 100 people will die prematurely from the accident at Chernobyl although life expectancy in Ukraine is only 72.

In which case your original number is too low, but at this point, what are a few more dozen premature deaths anyways? The number of people prematurely dead in the Ukraine since 2014 (including everyone aboard a shot down airliner) is considerably higher anyways - and just as ongoing as whatever not particularly large number of deaths one can attribute to Chernobyl.

Of course, the considerable amount of money spent entombing the reactor could have been used for other things, but why include something like that when talking about human cost?

Why is my number too low? That was the estimate given in 2006 or 2007. By now it is likely higher, but not by the hundreds.

It may also have killed Elvis Presley, and caused the NY Mets to suck.

"Vanillionaire" (51) is my favorite new word.

"Expensive placebos work better than cheap placebos."

Veblen goods?

#3 Well, duh. Or they could be an elaborate message drop system used by Chinese, NorKo, and Russian spies. The purchase of the book is verification of securing the assets, then the information is sent to the drop site found in code in the books.
Or it could be nothing.

#4 makes very little sense. First. it is just an unsubstantiated claim apparently made by the Zipline itself with no references to actual data. Second, we have no idea how much blood at all is shipped outside of Kigali, and if the total amount is statistically negligible, calculation of fractions delivered by various methods is pretty much meaningless.

From Lancashire Post:

"18,000 hidden pipe leaks in the North West over the past 12 months"

We already knew there were, "4,000 holes in Blackburn, Lancashire:

43 I understand completely. Factories are beautiful at night. Among the more dramatic images you can see in real life.

Comments for this post are closed