I would have mentioned this earlier, but the excellent Illusion Sciences blog went over its Google Pages hosting quota, so I couldn't snag myself a copy of the SWF file to host on my own site.
Now the blog's back, so here it is:
Brilliant, huh?
More info at Illusion Sciences.
UPDATE: Here's a nifty HTML5 version of the same illusion, as per caerphoto's comment below.
Ben Goldacre's latest Bad Science piece returns to one of his, and my, pet peeves.
It's the "scientists have discovered the formula for" story. You know the sort - they've discovered the formula for the perfect sexy walk, or the ideal biscuit-dunk, or whatever.
These stories are invariably provided by PR companies and self-promoters, and used as gap-filler by understaffed newspapers and TV stations the world over. And they are invariably bullshit of the very highest order. The "formulae" seldom even make internal sense, and when they do make internal sense you can count on them being quite unconnected to anything in the real world.
And, and this is the part that really matters, these stupid stories help to create a public perception of "scientists" as white-coated "boffins" with no real comprehension of the world, who have nothing important to do all day, and nothing comprehensible to say.
You know you're looking at a really broken formula when you see one side of the equation all being multiplied or divided by one variable. The formula for "the perfect joke", for instance, has one side divided by the number of puns. This means all jokes that do not include any puns at all are either infinitely funny, or funny to an undefined extent, depending on which way you look at it.
Many of the "formulae" don't even get that far, though. They're just a misshapen assemblage of algebraic characters, such as you'd expect a seven-year-old to draw if they were pretending to be a mathematician.
This latest example, an alleged formula for determining the "naughtiness rating" of a woman's garment, is entirely representative. It makes no sense in the first place - as long as your nipples are covered, it's apparently impossible to be naughty at all - and the example of it used in the article is broken, with an obvious but un-noticed multiplication by zero making the book-promoting "Cambridge mathematician" responsible look, appropriately, an utter tit.
So far, so usual.
But then Ben points us to Apathy Sketchpad, where one Andrew Taylor has dedicated himself to the detection and analysis of every Stupid Formula Story that's ever made it to the news.
Sometimes, I regret to say, Mr Taylor finds himself driven to profanity.
An outfit called "Balance Health Products" has recalled its slimming "dietary supplement", on account of how the "Starcaps" supplement turned out to "contain an undeclared drug ingredient - Bumetanide - a diuretic available by prescription only."
Once again, I stand awestruck by what I now choose to call the Dietary Supplement Contamination Phenomenon.
The DSCP, for reasons nobody in the world will ever figure out, causes all dietary supplements that've been "contaminated" with a drug to be contaminated with a drug that does what the supplement is meant to do.
Well, after a fashion, anyway. "Black Pearl" arthritis pills that're "contaminated" with steroidal anti-inflammatories work really well - they're just rather dangerous. The same applies to "herbal" diet pills that've mysteriously acquired a stimulant contamination.
But the "Starcap" pills have only got a diuretic in them. So they'll give you an easily measurable weight loss when you start taking them, on account of causing you to pee out a larger-than-normal proportion of your body's water weight. But they'll make no difference at all in the long run. And, in the meantime, you may be somewhat dehydrated.
Still and all, though, this does qualify as another "supplement" that's mysteriously acquired a contamination of a drug that does what the supplement is meant to do.
It's amazing - slimming pills never get contaminated with Viagra, arthritis pills never get contaminated with digitalis.
It's a mystery for the ages.
This is my picture of the Moon/Venus/Jupiter "smiley face" conjunction that just happened. There are many like it, but this one is mine.
(The 500-pixel-high version doesn't look like much. The full-size one is better. There was a little cloud, though.)
I wasn't planning to take this picture, but I went for a walk to Echo Point and found a person there hopelessly trying to photograph the conjunction with a full-auto compact camera. Which actually did have a "Starry Sky" mode (among many others, including one called "Food"...) - but said mode was, of course, useless.
So I promised to take a proper picture and e-mail it to her. And when I got back to Echo Point with my DSLR and tripod there was a family there trying to take the same picture, and failing for the same reason. So I harvested an e-mail address from them too.
A few times, I've gone to Echo Point and it's been cloudy or foggy or otherwise not the ideal time to take a picture of the Three Sisters. On those occasions, I've offered to send disappointed photographers a picture I took on another date, because I've got some excellent ones.
My favourite is this one, which I took at 2:37 in the morning, by moonlight, with a thirty-second exposure.
Everyone I've made this offer to so far has declined, though. And fair enough, I suppose; the Three Sisters I photographed last year is not the same Three Sisters you would have photographed had you been able to see it through the bleeding fog. But the Smiley Conjunction I photographed is, within a fairly small time window, the same one that the people I sent it to saw.
I just now belatedly noticed that Derek Lowe has written another of his occasional, but always entertaining, posts in his Things I Won't Work With category. (See also.)
This one's about triazadienyl fluoride.
You know how sometimes when you react one dangerous substance, like chlorine, with another dangerous substance, like sodium, you get something perfectly harmless, like table salt?
Well, that's not what happens when you react fluorine with hydrazoic acid.
When I read The Barna Group's "Most Americans Take Well-Known Bible Stories at Face Value" (which, yes, was a year ago, but it's not as if there've been a lot of great breakthroughs in the field since then), I was not entirely surprised to read that "Americans ... remain confident that some of the most amazing stories in the Bible can be taken at face value."
Given that, as I've previously mentioned, the USA appears to be a country in which 21% of the atheists believe in God, it's not surprising that - to pick one example from the Barna survey - 64% of Americans (or at least of the Americans that the rather preachy Barna Group surveyed...) believe that Moses literally parted the Red Sea.
This, however, is definitely one of those situations where it would have paid for both the people doing the survey and those writing stories about it - presuming they didn't all just have an axe to grind - to sit down for a probably-unavailable minute and have a little think about exactly what their findings meant.
Since it would appear that they didn't, let's do it ourselves, shall we?
Look at that 21%-of-atheists-say-they're-theists thing, for example. This turns out to be, so far as I can see, an actual, fair, genuine result. 21% of people who clearly said they were atheists also clearly said they believed in a "God or universal spirit".
That finding is from a Penthouse Pew Forum survey, which I consider rather more reliable than a Barna one.
Pew, you see, make their methodology and detailed results freely available. There's a PDF, here, that shows you the actual survey questions, next to the results.
On page 27 of that document, there's what looks to me like a very fair way to quickly find someone's religious affiliation or lack thereof, which includes a re-questioning for people who've been given the final options "atheist, agnostic, something else, or nothing in particular" and chosen the last option, to make sure they actually want to be "nothing in particular", and not atheist or agnostic.
You can never make a survey question perfect; in this case there's the problem of people who, like me, hold the considered opinion that gods do not exist (atheism), but accept our own fallibility and thus admit that we might be wrong (agnosticism), however improbable that may be. We therefore tick the "atheist" box, but if later on we're asked whether we think there's the slightest possibility that gods may exist, we'll say yes, like an agnostic.
But the Pew survey is about as good as a quick multiple-choice test is ever going to be.
In the rest of the main "topline" document they roll all of the Unaffiliateds together into one line, which presumably explains why seventy per cent of that category actually report belief in a God or universal spirit (page 44), and 36% of them (page 45) say they're absolutely certain that said entity exists, neither of which beliefs are at all compatible with atheism or agnosticism.
You can get a nice detailed separate table that breaks down all of the religions (PDF), though. That table shows you that 515 people, 10.2 per cent of the 5048 "Unaffiliated" respondents, said they were atheists.
Taken all together, this indicates that the Pew Forum aren't getting their "21% of atheists believe in god" result by subterfuge.
Pew's main "topline" document doesn't break Unaffiliated out into Atheist, Agnostic, Secular Unaffiliated and Religious Unaffiliated in its tables, presumably to make them clearer. But it does not appear that they're pulling a statistical fast one by, for instance, rolling all of the responses from Unaffiliateds together and then just declaring 10.2% of those responses to have been from atheists, even if none of the atheists actually reported belief in a deity.
No, it really does seem that about 108 people that Pew surveyed clearly declared themselves to be atheists, and then clearly professed belief in a god of some sort.
That doesn't, of course, make a blind bit of sense. Atheism is not a religion, just as baldness is not a hairstyle and no car in the driveway is not a kind of car in the driveway. But it's not the survey-givers' job to educate people about terminology. If you want to say you're an atheist who believes in a god, they'll write your answer down like everyone else's, even if that answer indicates that you're ignorant, nuts or a prankster. Fair enough.
Now, let's look at the Barna Group's survey.
Oh, wait a minute, we can't. They'll be happy to sell us umpteen books about being a better Christian or their copyrighted Christian Leader Profile test, but I can find no trace on their site of even the opportunity to buy a copy of any of their actual survey questions and results.
So now we're in the woods. Who knows what questions Barna actually asked, and what answers people actually gave?
With the right survey, you can get people to say pretty much anything you want. You can even influence their beliefs. ("What effect would it have on your vote if you were to discover that Candidate Smith is a child molester?")
Like Sir Humphrey persuading Bernard that he both supports and opposes reintroducing conscription, the framing of the questions makes all the difference. Especially when you're asking people about things that they don't actually think about much, or even care about much, like whether David actually killed Goliath.
As anyone working in this field knows, you have to take considerable care, even if you're scrupulously honest, to make sure that the meaning of your questions, and the meaning of the respondents' answers, is clear.
Stop people coming out of a church, for instance, and ask them if they believe in the Immaculate Conception. Most of them - Catholic or Protestant - will probably say that they do. So you can tick down ninety-whatever-percent on your survey and then issue a press release saying that belief in that doctrine is very strong, hurrah.
What most of the people will have thought you were asking, though, is whether Jesus was born of a virgin. The Catholic Doctrine of the Immaculate Conception actually states that Mary was born free of original sin, on account of how Jesus could not be incubated in the wicked womb of a normal woman, and if she weighs more than a duck she's a witch.
Your average rank-and-file dozes-through-the-sermon churchgoer is somewhat unlikely to know this. Your non-churchgoing ticks-the-box-marked-Christian person on the street is very unlikely to know.
So if I were running a survey like the Barna one, then apart from making sure I released the questions and not just digests of the alleged answers, I'd also make very clear exactly what stories I was asking about, without just using common names that people often misinterpret.
(To be fair, the Barna survey probably generally did that; there's not a lot of room for error when you're asking about stories like Jonah and the whale or Daniel and the lions' den.)
But I'd also scatter in a few Bible-ish stories that were actually made up just for the survey. Jesus... blessing the fields... of the Moabites, say.
If respondents say they believe stories that not even Dan Brown ever mentioned - as, I bet, many of them would - then clearly people's statements of belief in the other stories should be taken with a large grain of salt.
The Barna survey press releases do, however, tell you something about George Barna, who is I think representative of a peculiar movement in American Christianity. This press release about the Bible-story survey manages to restrict its preachiness to a "Reflections on the Data" section, but the one I mentioned earlier contains a number of places where George expresses the strangely popular, and reliable-like-clockwork, belief that Christianity in the USA (and elsewhere!) is "under siege".
"...While the level of literal acceptance of these Bible stories is nothing short of astonishing given our cultural context...", for instance, and earlier on "Surprisingly, the most significant Bible story of all - 'the story of Jesus Christ rising from the dead, after being crucified and buried' - was also the most widely embraced."
Outside observers may find this slightly bizarre, since Christianity in the USA is obviously massively dominant...
...and Christians who don't believe (or at least say they believe) that Jesus was resurrected are pretty hard to find.
But there's also no shortage of talking heads eager to opine that the evil forces of secularism are constantly gaining ground in their unholy mission to re-name trees with lights on them, and so on.
To be fair, Barna goes on to say (in the third person...) that the real problem is that all of the nominal Bible-believing which his who-knew-what-it-asked survey discovered doesn't translate to much in the way of actual "Christian" acts. So I suppose that's the grain of rationality within the "Christianity under siege" belief; that lots of people say they're Christians, but you can't find a lot of true Christians among them.
But, again, this doesn't seem very surprising to anybody who accepts the not-too-hard-to-support point of view that Christianity is just another major religion, which the overwhelming majority of adherents use not to lead them into the light, but to justify whatever they wanted to do anyway. Yes, believing the Bible ought to lead to defined-by-Barna-as-"Christian" behaviour. But no remotely sensible reader of the New Testament could possibly conclude that Jesus would find it acceptable for you to drive a Lexus to church - and yet "prosperity theology" has sprung up to bridge the gap.
Similarly, the idea of karma ought, you would think, to lead people to good behaviour. But instead, your average Hindu-in-the-street is quite likely to believe that karma means that miserable beggars, children raped by their parents, or any other unfortunates you care to name, are suffering righteous punishment for bad deeds in a past life. And, again, that the prosperous deserve their prosperity, for surely god(s) would not have given the rich so much money if it were not their just reward.
All of this makes sense, if you don't think there's One True Religion that should guide its followers to be obviously better people than those who've foolishly been raised in some other, fictional faith. But to people like Barna, who believe that their particular religious variant is that one special phone-line to God, the entirely ordinary behaviour of their fellow believers can only be explained by the evil actions of external forces, besieging the chosen of God and leading - nay, forcing - them away from the righteous path they'd otherwise obviously choose to follow.
Adding fake-Bible-story questions to the survey could have helped Barna out, because it would have given him a chance to claim that people of disappointing morality who believe that David fought Goliath, but also believe that Josiah, um, washed the Pharaoh's feet, clearly do not in fact know much about Christianity and could therefore not be expected to be particularly righteous.
Adding fake stories, though, could also have measured the credophilia - indiscriminate collection of beliefs - that lies at the core of a lot of religions.
If your religion says that faith by itself is a virtue, you shouldn't be surprised if you end up with a bunch of people who'll believe almost anything. And who'll think that holding those beliefs, without doing anything else in particular, is enough to get you into heaven.
The Metafilter thread about Michael Crichton's unexpected death is less respectful than most death-threads there.
Which is, of course, no big deal. Someone always says "show a little respect, you wouldn't act like this at the guy's funeral" when obituary-thread commenters not only omit the traditional moment-of-silence dot, but even say bad things about the deceased.
But Crichton's grieving family are probably not frantically refreshing Metafilter right now. And MeFi users are, overall, pretty enthusiastic about the advancement of human knowledge. And Michael Crichton did human knowledge a few significant injuries, especially with his later books.
A lot of commenters said they loved Crichton's books when they were kids. I bet I would have, too, but I think I just didn't read any. Maybe The Andromeda Strain, but I'm not sure.
I'm glad only the earlier, less anti-science ones would have been available then, though.
That's because I read State of Fear as an adult, and the only part of it that seemed obviously stupid at the time - never mind the implausible environmentalist Giant Conspiracy, stuff like that's normal in thrillers - was the magic gadget that caused enemies of the enviroterrorist baddies to be struck by lightning, in cheerful defiance of basic electrophysics.
(Enviroterrorists with the ability to suck lightning out of clouds wouldn't need to be enviroterrorists any more. They could just start building lightning power stations.)
The real scientific problem with State of Fear whistled right over my head, though. The book contains several Author Filibusters about climate change - or, more specifically, the allegedly poor quality of our knowledge about climate change - and all of those sounded plausible enough to me at the time.
Only because I looked into it later do I know that Crichton was talking absolute cock, and had to be either fully aware of this fact, or senile, or wilfully ignorant.
(See also: Poor old Antony Flew.)
Crichton didn't seem to be very good at dealing with criticism. He famously named a throwaway character in a later book after someone who'd given State of Fear a very bad review. The throwaway character was a baby-rapist. Classy.
Anyway, I like to think that if I'd read State of Fear when I was twelve, I'd also have looked up the facts afterwards. But I bet I wouldn't have.
The world is, self-evidently, well-stocked with people who don't do any more due diligence on what they read than I would have when I was a kid. People believe bestselling thrillers that make statements about the nature of the world, especially when those statements are the core of the whole story, as they are in all of Crichton's works. If you're writing about things that happen on Planet Zarnax or in the Cthulhu mythos or whatever then that's one thing, but if a book's whole anvil-subtle thesis is that the scientific consensus about climate change is wrong, you need to take your share of the responsibility for everyday voters believing that you're right.
Heck, even utter garbage like Left Behind has an enthusiastic audience of people who don't even bother checking its statements against the Bible. Much less check to see whether, to name just one of a very long list of outrageous wall-bangers, the United Nations really does have the power to take over the whole world on a whim.
Given the power of popular books, it's simply irresponsible to put misinformation about matters vitally important to the whole world in books which you - and your bank account - know are read by orders of magnitude more voters than read the scientific papers that prove you wrong.
If you're the only loud voice that's talking rubbish, then it's not such a big deal. But when there's a genuine culture war going on about climate change, or evolution, or dear-god-now-they're-coming-after-neuroscience, the side you pick matters. State of Fear seems to have become a sort of easy-reading textbook for climate-change deniers. Look, for instance, at the whole-hearted support Crichton got from fellow bullshit artist James Inhofe.
People should be allowed to write, publish and read whatever the heck they like, no matter how contrary to fact it is.
But if you're in the business of lying to people about matters of grave global importance, I for one am not going to shed a tear when you die.
(On the subject of books that contain wise and wordy characters who entirely agree with the author, see also Robert Heinlein, whose books I loved as a kid. In my memory, Time Enough For Love is completely awesome. So I'm not going to make the mistake of trying to read it again now. Fortunately, kids' sci-fi that actually gets the science more or less right also exists.)
Do you want to tackle your alcohol addiction with safe and effective herbal medicine?
Look no further than Great Home Remedies' How To Prevent Alcohol And Drug Addiction page!
I particularly liked the part that says:
"Very effective remedy: Take 1 lovage root and 2 laurel leaves, add 250 ml of vodka and leave it in a dark place for 14 days. Let an alcoholic drink a whole glass. Usually even 1 time is enough to stop an alcohol addiction, but you may do it 2-3 times."
I suppose it must be the laurel that does the trick there, since lovage is one of the flavouring ingredients in Bénédictine liqueur. Which has not demonstrated a strong tendency to cure people who drink it of wanting to drink. Although if you polish off a whole bottle, you may not very much desire to drink Bénédictine any more.
(I call this situation a "I haven't drunk X since Y" story. Popular Xs include black Sambuca, ouzo, and any pre-mixed cocktail based on Baileys Irish Cream, especially if it's a generic copy thereof. The Y part of everybody's story is usually very similar.)
250 millilitres of 80-proof vodka will, of course, also give you about the same amount of alcohol as a six-pack of beer. In one belt. But any proper alcoholic should be able to handle that, with a water chaser.
(If you leave the vodka in an open container for the 14 days then a significant amount of the alcohol may have evaporated off, of course. But they don't tell you to do that.)
More seriously, one problem with this and various other herbal remedies is that the amount of active ingredients in a given plant can vary widely even within the one species. Different plants can have different concentrations depending on their strain and how and where they grew, and fresh bits of the plant can be very different from dried bits, too.
And, even more importantly, instructions that tell you to use "2 laurel leaves" do not specify which of the thousands of members of Lauraceae family they're talking about. Some members of Lauraceae are not known as laurels - cinnamon and avocado, for instance - which makes it a little easier. But there are several other "laurels" that aren't members of Lauraceae at all.
When you're playing a computer game, you know that when you pick a "nightshade mushroom", or whatever, you've definitely got the right thing, because there are only a dozen species of pickable plant in the whole game, so anything that looks like ginseng or mandrake root must be. In the real world, though, almost no plants have a common name that's not applied to many other quite different plants.
In this case, the "laurel leaves" they're talking about are probably "bay leaves" from the Bay Laurel. But then there's the entirely unrelated California Bay Laurel, whose leaves are poisonous. But maybe where you live, "laurel" means Camphor Laurel (mildly poisonous, but utterly different from Bay Laurel), or Cherry Laurel (berries edible, everything else poisonous).
You'll face the same problem with most other medicinal, and even simple food, plants. Something that looks like fennel, or like a parsnip, or (of course) like an edible mushroom, can kill you. And it even applies to the other ingredient in the anti-alcoholism six-pack cocktail; buy "lovage root" and you'll very probably get the usual kind of lovage, Levisticum officinale, but then there's the related "Alexanders", a.k.a. "Black Lovage", and Laserpitium latifolium, "Bastard Lovage", and even a poisonous lookalike sometimes called "Water Lovage". And that's not even all of the lovages!
Take-home message: Use herbal medicines if you like. Make herbal medicines if you like. But make sure you pin down the full Latin name of your ingredients before you eat them, and don't trust any source that doesn't give you the exact names.