Apathy Sketchpad

I’ve just read about a thing called the Dice-O-Matic. The gist is that the operator of GamesByEmail.com requires a lot of random numbers between one and six inclusive to feed his collection of online dice-games. And inevitably, people have complained that the numbers he’s used are insufficiently random.

And maybe they were, once. Originally, GamesByEmail used the pseudo-random number generator built into whatever the games are written in. Once ‘seeded’ with a starting number, such an algorithm will spit out a string of numbers which will have all the same properties as random numbers, except that if you know the seed, they’re totally reproducible (although still essentially unpredictable, much like the digits of π*). They’re generally seeded from a high-resolution timer, so this should never be a problem. They also repeat if you run it for long enough, so you should re-seed periodically. In theory, this should be fine, but you have to be very careful not to accidentally bias the selection.

Unfortunately, it’s very difficult to tell if your numbers are random enough or not. For example, some episodes of the dreary logical fallacy roadshow that is Deal Or No Deal used an Excel spreadsheet to randomise the assignment of 22 sums of money to 22 boxes — for which there are probably more sequences than there are grains of sand in the world — and the seeding was bad enough that only twelve of them arose in over forty shows. You can experience this for yourself: whether by accident or design, the Concentration mini-game in Super Mario Brothers III only ever shows players eight out of a possible 58 billion permutations of cards. The producers of Deal or No Deal switched over to drawing lots by hand.

  
xkcd’s ‘Random’ comic, which illustrates the difference between actual randomness and unpredictability, which is far more useful.

So (I infer) GamesByEmail switched to using random.org for their random numbers. Random.org link to their own story of a quiz show failing to randomise, this time costing them $100,000 in prize money (not that it brought anyone any happiness), and solve the problem of generating random numbers by means of four cheap radio antennae in Dublin, tuned into nothing in particular. The waveform of the white noise between radio stations is recorded, and the least significant bit (the last digit in binary; 0 for even numbers and 1 for odd) is recorded. Then, the stream of numbers are chunked into pairs, so 01001101 would become 01 00 11 01. 00 and 11 would be discarded as insufficiently random, and the first digits of the remaining pairs would be kept, so 01001101 gives two zeros. They throw away about 97% of the radio data, keeping only the most unpredictable bits possible. Your TV does a similar thing in reverse, when it blocks out random data and replaces it with a blue screen, while foolishly allowing Deal Or No Deal through unimpeded. It’s as near to pure randomness as you’ll get without invoking quantum theory (which states that some events in the universe are totally random, and indeed you can buy modules for your computer to generate random numbers in this way).

Of course, people still complain about the numbers from random.org. Of course they do. Random numbers, by their very nature, don’t look random. People believe in winning streaks, lucky socks, and prayer for exactly this reason. If I recall correctly, ball 44 was well known for a time in the National Lottery because it came up more than the others in the first few weeks, even though actually there were several sets of balls in use. Partly this is because humans have evolved to be shit-hot at spotting patterns, because in the wild that can stop us being killed. Natural selection favours the caveman who won’t eat the same berries that Ug, Thag and Og ate right before they died. In fact, generally people will eschew the berries after just one person dies. That’s a good plan for surviving in the wild, but it does make us spot patterns where none exist. Try it. Have random.org roll 16 virtual dice for you. I did it, and the sequence started 1155. That doesn’t look random. It had a 123 in it too. And there was only one 4. People tend to think numbers are random if they’re uniform: if I shuffled the numbers 1–6 into a random order (say, 341625), people would rather believe that was the result of six dice rolls than 115561, the first six that random.org gave me — but really the odds of getting one of every number are less than 2%.

If you encourage people to spot patterns, they can be relied upon to do so, regardless of whether the patterns exist. B F Skinner demonstrated this in pigeons in 1947. Pigeons were put in cages and fed periodically, “with no reference whatsoever to the bird’s behaviour”. At least six out of eight of them became totally convinced that they could cause food to be delivered by repeating some arbitrary motion such as turning anticlockwise. This has been replicated with humans, perhaps most famously by Derren Brown in Trick Or Treat, proving that Channel Four cater for both ends of the intellectual spectrum. Five guests were put in a room full of toys and instructed to accumulate 100 points to win a prize. In fact the points counter was controlled by two fish swimming around at random in another room (i.e., a poisson distribution). At the end of the game, four of the five guests were totally convinced they’d figured out a sure-fire way to score points. The other guest was Doctor Who. This may or may not be significant.

Random.org solved this problem by running constant statistical tests on their numbers. The numbers are expected to pass these tests most of the time — but not too often, or else that would be suspicious. GamesByEmail.com felt they needed something a bit more accessible to the kind of person who plays dice-games on the internet, so they built the brilliantly terrifying “Dice-O-Matic Mark II”. It is, in their words, “a 7 foot tall, 104 pound, dice-eating monster, capable of generating 1.3 million rolls a day”. It is literally a massive machine full of dice, which scoops them up, flashes them past a camera which notes down what numbers they show, and then flings them onto a ramp, whence they bounce back into the “pure seething violence” of the hopper full of dice ready to go round again. It runs about 90 minutes a day, and you can tell when it’s running from two rooms away. (It also uses some image processing which I found interesting because that’s what I do. If you want to read about it, visit GamesByEmail’s page.)

Ironically, I suppose, it’s technically less random than the random.org numbers were, but it’s a great PR move. After all, nobody can say it’s not a realistic simulation of dice: it is dice. But it neatly demonstrates the problem faced by people like lottery organisers: their job is to provide people with something people are practically designed not to be able to see. This may be why GamesByEmail add:

There is no doubt that I will still receive complaints about the rolls, but now I can honestly say I have done all that I can possibly do: the rolls you get are exactly as random as those you would get throwing by hand. As I promised earlier, if you donate to the site and are unhappy about the rolls, let me know and I will pull a die out of the machine, melt it flat and mail it to you, as an object lesson to the other dice.

According to Chortle,

Peter Kay is to release a follow-up to his bestselling memoirs The Sound Of Laughter this autumn.

What?

His existing book, which I have not read, is 368 pages long in paperback and was released on the 2nd of October 2006. The new one will be out in time for Christmas. That means that the book can detail, at most, 1179 days (assuming the books take an equal amount of time between writing and release). On average, there will be about 3.23 days per page (or, 31% of a page per day). That’s only slightly less than my work diary and I can never fill that. This book is going to read like a Twitter feed, especially when you bear in mind how much of that time must have been spent writing the damn book.

I’m sorry, but has enough genuinely interesting stuff happened to him in the last two years to fill that much book? I submit that it hasn’t. In fact, I confidently predict that Peter Kay’s second book is going to be basically all the same material as his first book, but with a couple of words changed here and there to make it sound like a whole new work.

Scientific journals have genuinely got the best business model in the history of anything ever. Here is how it works, in a nutshell:

• Other people, scientists, write their content for no fee.
• The journal then gets other scientists to review it. These scientists generally don’t get paid either.
• The authors edit the paper and send it back. Eventually, all the scientists reach a version of the paper they can all agree on (or the paper gets withdrawn). Then the authors pay the journal to publish the paper.
• The journal then charges anyone who ever wants to read the paper extortionate fees. $50 for a PDF file is not uncommon.
• The journal retains the copyright on the words they didn’t write describing experiments they didn’t do, and claim fees for reading it at least until the copyright expires and usually long after that.
• None of the scientists or their employers ever get paid.

This, to me, seems like an insane system. It survives because universities don’t care if they pay extortionate fees for such things and because it’s established. And probably when it was established it made sense — after all, who else but journals could publish things? But now it’s just academia needlessly funnelling money into a mostly pointless publishing racket. I really don’t see what it achieves.

I’ve said before that a better system would just be for universities to publish papers and let anyone who wants to comment on them comment on them. I hadn’t worked out the details, obviously, so it was one lunchtime rant in the pub opposite the lab and I didn’t think much more about it until the next time I hit a paywall demanding I give some publishers $75 to read a paper written by my own supervisor. But someone else has worked them out. It’s not perhaps an ideal system, but it looks pretty good to me and it’s compatible with the existing system.

Dubbed GPeerReview (I don’t know what the ‘G’ stands for but the author’s name is Gashler so that’s likely), the idea is that you post your paper on your academic website, email people you think would be interested, and those and any other readers can review it. They sign their review, along with a hash of the paper, with public key cryptography so you know who wrote it and what about. That way, you get an idea of how much support a paper has and, crucially, what kinds of people support it. The author of the paper puts up the most credible supportive reviews they can find. In theory, if it becomes accepted then there’ll no longer be any need for conventional publication. It’s a very clever system. (See also, the more established ResearchBlogging.org – which Gashler says could be complementary to GPeerReview but covers rather different ground. I think I agree with him on that — it’d be great to see things like that running it tandem.)

I’d love to see something like this made to work across academia. I suspect, though, that what kills it will be that real people don’t understand nerd stuff like public key cryptography. Everyone else in my research unit gets all annoyed if I try to use LaTeX or Bibtex at them. (Well, the dentists do — the other physicists love it. I witnessed a long argument about a week ago over the relative merits of Microsoft Project versus the open source alternative, which boiled down in the end to ‘well the free software probably is better but if we collaborate with anyone else they’ll demand we use Project’ which to me seems like a really crappy way of doing things — I’d rather piss people off by doing the right thing than pander to idiots and help keep Microsoft’s monopoly on proprietary, buggy software healthy.) They act as if Word and EndNote are somehow better. In my experience, Word doesn’t work properly and EndNote formats citations basically at random. LaTeX is a pig to get set up but at least once you’ve done it it stays set up. To be honest, I think that’s another thing that needs sorting: we need a specialist scientific markup language. Maybe a form of HTML (or other XML), with a standard equation format and a few extra specialised tags, perhaps including COinS for citations, which the reader software could be configured to render as a conventional reference, or as a hyperlink, or as whatever they like. A CSS-like ‘default’ style for a particular paper would be fair enough, but the current system that forcibly changes the format depending on which journal happens to have published the paper is rather silly. I don’t want a stack of PDF files all formatted differently. I want a folder full of pictures and ASCII-encoded markup that I can process and output how I like. Get into the twenty-first century. That’s how we do things here, because it’s a better way of doing it.

And there’s no reason that all of the above couldn’t be implemented really very easily, and I’d love to see peer review evolve into something more open and transparent than the existing system, which still relies on the trustworthiness of journal editors and the word of a few unidentified reviewers per paper. But we need nice, simple user interfaces on every part of it or else Joe Scientist isn’t going to actually bother to do it. We need a nice WYSIWYG program to edit the papers, then a nice Wordpress-style package to maintain your site, and a nice package to let you write reviews without much effort. Make it simple, and people might adopt it. Which is frustrating, because by rights you’d think a good scientist would be exactly the kind of person who would leap at the chance to adopt an open, collaborative, technological and free solution to a problem. Those are the qualities that science runs on. And I can’t see what we’d lose by switching to such a system, other than a load of jobs at journal publishers — and I’m sure the big journals would find a way to adapt. Perhaps they’d act as aggregators or run interesting comment pieces more often or something. (I should link to this very interesting discussion, where Gashler explains what journals do that is useful and that GPeerReview doesn’t do. I’m not convinced it’s all really a job for journals per se, but someone will have to keep doing all that. Personally, I think universities should do most of it.)

Bah. I just get frustrated when people cling to what they know instead of adopting obviously better alternatives, like Linux or metric or atheism or not torturing people. I guess that’s just a failing I have. But I’d love to know how any of the above could be shoehorned into the modern scientific community.

• 0.01s: I’ve knocked that pint glass.
• 0.1s: Oh, shit, it’s falling. Maybe I can catch it if I thrust my arm in this direction…
• 1s: No, that’s just caused it to shatter all over the pots and pans instead of the floor. That’s much worse.
• 10s: Actually, if it had landed on the comparatively soft lino floor, it might not have broken. Shit. Right. Anyway. That’s enough standing around staring at broken glass. Time to clean up.
• 1m 40s: Although usually untidy, I appear to be a neat freak when shards of glass are involved.
• 16m 40s: Oh, God, it’s everywhere. I’m going to be eating bits of glass for ages. I wonder how I am supposed to get bits of broken glass off a non-stick pan. I hope this isn’t the kind of glass that has the same refractive index as water.
• 2h 46m 40s: Tell room-mate we’ve lost another pint glass. Why can’t they make pint glasses out of the same stuff as car windscreens? They break at the slightest provocation; why can’t they be shatterproof?
• 1d 3h 46m 40s: Hasn’t happened yet. If still alive at this point, I will assume all is well forever.
• 1w 4d 13h 46m 40s: I will have forgotten the whole thing.

This is the same thing as I did for the Telegraph, but for the Daily Mail. This one was harder because their search function is bad and their website unreliable. Also the dross between the formula stories was more depressing. But then, it did turn up the brilliant formula for the perfect horror film, so that’s something…

• The Perfect Day: O + NS + Cpm/T +He, except when it’s [W+(D-d)]xTQ MxNA
  As in the Telegraph, this was repeated year on year on year on year.
• The Perfect Christmas: PX = (8F * 4P +£23 * 8F + 3G + 2W + 2W:3C + 5T:1NR / 3D) / 3D
  As covered previously. Presumably even following this formula Christmas is still worse than June 20th.
• The Perfect Bacon Butty: N = C + (fb (cm) . fb (tc)) + fb (Ts) + fc . ta
  Apparently, “the experts at Leeds University tried 700 variations on the traditional bacon butty.” I did my undergraduate degree at Leeds and I can vouch that this is true, although I had no idea it was research. Dr Graham Clayton is to blame for this.

Out of interest, N = force in Newtons required to break the cooked bacon. C = Newtons required to break uncooked bacon, fb = function of the bacon type, cm = cooking method, tc = cooking time, Ts = serving temperature, fc = function of the condiment/filling effect and ta = time or duration of application of condiment/filling.

• The Perfect Present Wrap: {(d+2h+w)2 2(w+h)2 — whatever that means
  Thanks to Dr Sara Santos at the University of Manchester, “we now know why we put everything in boxes”.
• The Perfect Sitcom: formula not properly explained
  As in the Telegraph. Repeated, presumably on Dave.
• Staying Awake At Work: CDA + CT + KF TMT
  Bear in mind that KF stands for “knacker factors”, so this is Maths. This comes from “experts at fatigue management consultancy Awake”.
• The Perfect Cheese Sandwich: W = (1 + bd/6.5 - s + (m-2c)/2 + (v+p)/7t) * (100+l)/100
  “Geoff Nute and his team” of “sensory analysts at Bristol University” produced this equation, which says that without a tangy sauce, you need infinite cheese. This was in the optimistically named “science and tech” section.
• The Perfect Penalty: (((X + Y + S) / 2) x ((T + I + 2B) / 4)) + (V/2) -1
  As in the Telegraph.
• The Perfect Breasts: the nipple should be 45% of the way down. Apparently.
  “Patrick Mallucci spent many hours poring over photos of topless models in lads magazines and tabloid newspapers to formulate his theory.” Enough said.
• The Perfect Teeth: no formula
  This isn’t strictly a formula. It’s really a set of rules about what makes a nice smile, of use to cosmetic dentists. I’ve seen at least one of them discussed in the British Dental Journal, but that was to debunk it. Hard to say what the truth is. Better at least than “the perfect cheese sandwich”, but still…
• How To Beat The Post-Holiday Blues: ((j+c) x (r+t) - (h+o))/b
  Professor David Holmes of Manchester Metropolitan University “carried out the research on behalf of Churchill Travel Insurance”, who will presumably use it as a basis for premiums on Post-Holiday Blues Insurance. As ever, all the variables are listed on wholly arbitrary 1-5 kind of scales. According to the Mail, b represents “whether gaps between holidays too long”. Yeah. Gaps too long. Also, verbs for losers.
• Will You Get Seasonal Affective Disorder?: X a x ((24-b) x (c+d+e) + f x (g+h+i))
  Here, we learn why you should always use the multiplication sign instead of the lower case ‘x’. Also why you should remember to include the equals sign. This was devised by “consultant psychiatrist Dr David Wheatley” and “commissioned by Kira St John’s Wort, makers of a herbal “happiness” supplement, as part of a study on depression”. It has to be said, the list of instructions is sufficiently varied and complex to give the whole thing an air of credibility. But still…
• How Beer Goggles Work: no formula supplied
  “Bausch & Lomb PureVision, one of the world’s biggest eyecare firms” got “Professor Nathan Efron, Professor of Clinical Optometry at the University of Manchester” to do this. I don’t know why it’s so often universities I’m at where this stuff happens; before I started at Leeds, Dr Clarke, who took our electronics lectures, was asked by some supermarket or other to work out an equation for how to flip a pancake, and I guess they were expecting him to wander off and make something up, but no, he built a huge red trebuchet-looking thing to flip a strange cardboard pancake. I was there for four years, and some of my friends worked on this for a brief period. Partway through my course he retired, an act which made him much easier to locate — his workload went from insane to average. To my knowledge, the only thing this project has ever achieved is to break countless platinum-iridium tips for the tunnelling microscope. I suppose that means that they at least have more credibility than the “oh, it’ll be b times a plus 4d over qpr” crowd, but still…
• When Heyfever Is Worst: 6.02pm on May 29; no formula supplied
  Stay indoors at that time, is my advice. “Dr Adrian Morris, allergy specialist for Boots Health Club, … created the hayfever formula”.
• The Perfect Horror Film: (es+u+cs+t)² +s+ (tl+f)/2 + (a+dr+fs)/n + sin x - 1
  Who says modern films are too formulaic? This is science! Look! It has a fucking sine function in it: 

  The experts have taken blood and guts (Sin x) and subtracted it by the stereotypes (1), to make Sin x - 1, saying Jack Nicholson’s character in The Shining turned into the total opposite of a protective father figure.

  See? That’s Science! Don’t say it isn’t! “Mathematician Anna Sigler, … a former graduate from King’s College, London” did this research. A former graduate, no less. Presumably her degree was revoked when they saw what she was doing with it. The Shining won, by the way. “The research was carried out for Sky Movies, which will be showing The Shining and other scary movies this weekend.” Coincidence.

• How To Wash Your Hair: formula not stated
  “Kerys Mullen, technical manager at Dove, said: “A lot of people ask us about the best way to wash their hair so we decided to work out the ideal formula.”"
• The Perfect Boiled Egg: formula not stated
  How to boil an egg, by several chefs. At the bottom, input from “Dr Charles Williams A physicist from Exeter University [sic]“, who “has worked out a formula for the perfect boiled egg based on the ‘heat-diffusion equation for spherical objects’”. Fair enough, but I for one will trust the chefs on that one.
• The Perfect Day To Change Your Life: M x O + Bh (H+R) x S; max. May 18
  This is the handiwork of Cliff Arnall, the same Cardiff University muppet responsible for the best/worst day formula the Mail and the Telegraph obligingly publish twice every year. If anything this is worse than that one: “Under the formula M stands for motivation and O for opportunity while Bh is bank holiday proximity. The H in the second half equals increasing hours of daylight, while R equals reflection time and S, simply success.” Yes. And..? Surely the aim is to maximise S? In which case, shouldn’t it be on the other side of the equals?
• The Secret of True Happiness, no less: P + (5xE) + (3xH)
  Thomson travel got “psychologist” Carol Rothwell and “sports scientist and ‘life coach’” Pete Cohen to “insist their equation is a useful guide to our levels of satisfaction with life”. Because just asking “are you happy” doesn’t work. Not enough maths, see.

Here is a list of “mathematical formulae” and “scientific equations” which detail every aspect of our day-to-day lives, all “calculated” or “devised” by “scientists”, “academics”, “economists” and “mathematicians” from various embarrassed universities. These are all taken from the Telegraph. Don’t imagine other newspapers are better…

• The Perfect Sitcom: quality = (rd+v)f÷a+s
  Dr Helen Pilcher, a neuroscientist (according to the Telegraph) whored her name to this ‘research’ which was commissioned by UKTV Gold to promote their endless repeating of that clip of Del Boy falling through the bar: f in the equation means “the amount someone falls over”. This is about the level of humour the Telegraph seems to like, because…
• The Perfect Joke: x = (fl + n^o)/p
  In this case, n represents “the amount someone falls over”, and is raised to the power of “the “Ouch” factor”. It won’t surprise you to learn that this is the work of the same Helen Pilcher, although this time helped by comedian Timandra Harkness. It should be some measure of Harkness’ fame that I wouldn’t like to guess what gender the name Timandra indicates. To their very limited credit, the telegraph article does include rants from Jimmy Carr, Bernard Manning and Ruby Wax explaining that the formula was stupid (in their own obnoxious ways). Also Nicholas Parsons, but he’s not obnoxious. Why was this in the news? “The Comedy Research Project, a live stage show featuring Helen Pilcher and Timandra Harkness, will be performed at the Science Museum’s Dana Centre on June 15 and 22 [2006].” I bet that was a fucking blast.
• The Perfect Day: quality = O + NS + Cpm÷T + He
  This was pulled out of the arse of Cliff Arnall (not Lou Reed), a psychologist and former tutor at Cardiff University, because Wall’s Ice Cream asked nicely. The Telegraph notes it “does not take into account the gloomy forecasts for the British economy, fears caused by falling house prices, rising inflation and stagnating pay rises, England not playing in the Euro 2008 and a damper than normal start to the summer”. All the factors in the formula are utterly subjective and the whole thing is worse than most. The comments on the Telegraph pages are fun. This is especially perverse because in 2006 the perfect day was three full days later. (The Telegraph really do obligingly report this, from either end, every time they’re asked.)
• The Perfect Bra: formula not supplied
  This one is actually real (albeit slightly over the top) research! It could genuinely improve your life (moreso if you are a woman). I know; I was as surprised as you are.
• The Perfect Rugby Kick: KP = CSP - s + w + r + yⁿ + cr + sc + mt + xⁿ + ctw
  This is just a shopping list of things that affect a rugby kick. And “y to the power of n represents other factors”. My word. This drivel comes to us no thanks to “Andrew Cushing and Prof Paul Robinson at University College Worcester for the research company QinetiQ”.
• The Price Of Cleaning: price = time × £6.16/hour
  This is a note that the average wage has increased, listed in terms of how much people lose out on by not being paid to brush their teeth (30p, although it doesn’t say how much they save by not having to get private dental treatment if they don’t brush). Barclaycard convinced Prof Ian Walker, an economist at Warwick University to endorse it.
• The Perfect Marriage: formula not supplied
  “Prof James Murray of the University of Washington” says this formula has a 94% success rate in predicting if a couple will divorce, although really I’d want to know sensitivity and specificity, otherwise you could conduct a survey of evangelical Christians and the terminally ill, say they’ll all stay together, and declare yourself the winner. They later ran a second article about how it was nonsense.
• The Perfect Chip: formula not ready at time of press
  That’s right, because Dr Gama Khan won’t just sign off on whatever nonsense Tesco ask — that, or Tesco asked for a big long experimental phase they can publicise for months. Khan says “The competition is intense because everyone wants to go down in history and finally crack the secret of the perfect frozen oven chip. I am looking at a lot of chips. Some days I’m testing them continuously from 9.30am to 4pm. It actually can get quite sickening, particularly when I always smell of chip fat.” And it’s true. Everyone wants a slice of the elusive Nobel Prize in Fast Food.
• The Perfect Football Penalty: odds of scoring = (X + Y + S)×(T + I + 2B)÷8 + V÷2 - 1 [simplified]
  This was commissioned by Ladbrokes, and is credited to “by scientists at John Moores University in Liverpool”, which quickly becomes “Dr David Lewis, a mathematician”. I think this quote tells you all you need to know about the mathematical ability of everyone involved in this report (emphasis mine):

Dr Lewis and his team found the six variables that influence a successful penalty kick are: V = velocity of ball once struck, T = time between placing ball on spot and striking the ball, S = number of steps in run-up to strike, I = time that the ball is struck after goalkeeper initiates his dive, Y = vertical placement of ball from ground, X = horizontal placement of ball from centre and B = striking position of boot.

• The Perfect Sandcastle: 0.125S = OW
  This simply states the ideal ratio of sand to water. Personally, I would just use the pre-prepared wet sand b the beach, which must surely be about right because it does seem to work. “Prof Matthew Bennett, the head of environmental and geographic sciences, Dr Brian Astin, the head of the School of Conservation Sciences, and Rob Haslam, laboratory and technical services manager, then spent two days testing the samples for their suitability for sandcastle building. … Teletext Holidays, which commissioned the research, will be holding a sandcastle-building championship on July 24 [2004] in Great Yarmouth.” This work was replicated the following year by “an MIT team, led by Sarah Nowak and Arshad Kudrolli” who reached exactly the same conclusion (although they phrased it in a simpler way). This might be nearly useful to some engineers somewhere.
• How To Open Champagne: P = T÷4.5 + 1
  P and T are pressure and temperature. I think this is not made up, although not really that useful in real terms: essentially it says that if you cool the champagne it is less likely to explode on you. This comes from “Dr Steve Smith, a lecturer in wine studies at Coventry University”, who “was commissioned to develop the formula after a Marks & Spencer survey found that 50 per cent of women are too frightened to open a bottle of bubbly because they fear that the cork will fly out prematurely, hitting them or a precious ornament”.
• The Perfect Place To Shop: D=f(m,b,c)
  The function f is undefined. “Retail and consumer trends expert Tim Dennison has come up with a formula to help Yellow Pages calculate how diverse and lively high streets are.” It says little town streets are more diverse than city centre ones. Nobody is surprised.
• The Perfect Newspaper: no formula
  It’s the Telegraph. Shocking. I suspect this is bad self-congratulatory reporting of some tiny little statement the academics made, but then I work for Manchester University so I am biased (although I’m not certain which way).
• How To Pour Gravy: amount of gravy = (W - D÷S) ÷ D × 100
  According to “Dr Len Fisher, an independent food scientist at Bristol University… who was funded by the manufacturer Bisto”, this is important because “more than 150,000 gallons of gravy is left every week.” Hard to see what Bisto have to gain by this, except of course that they’re in a newspaper.
• The Perfect Book: formula not done at time of press
  …although it’s going to be Agatha Christie, says Dr Roland Kapferer.
• The Perfect Biscuit: formula deemed to complicated for Telegraph readers
  This was led by Professor Bronek Wedzicha of Leeds University and “half funded by the Department for the Environment, Food and Rural Affairs and half by United Biscuits.” The researchers insist that this is real research rather than a publicity stunt, and I see no reason not to believe that, especially since they spent £91,000 on it.
• The Perfect News Story: “never trust your own instincts” but rely on “tried and tested formulas, bland ingredients and using up old scraps and leftovers from the day before, particularly the choicest cuts from the Daily Mail - no matter how stale.”

Some time I might do this for other newspapers, although I’m not sure I could read the ones in the Daily Mail faster than their hacks can produce them, so perhaps I won’t.

A new report by the independent think tank Reform says that the number of maths graduates in Britain has fallen from 84,744 in 1989 to 60,093 in 2007. That’s a loss of 24,651 in only 18 years, or 43.4 microHertz. If this trend continues then we will have no mathematicians at all by the year 2,050.87952, and by 2,075 there will be -33,033 mathematics graduates. Since a mathematics graduate is expected to contribute an extra £3,080 per year to the economy, this will represent an annual cost of over one hundred megapounds per year. That level of spending would exhaust all the Earth’s money in only 30,000 years, meaning that the world’s economies will be at the mercy of the huge amount of negative mathematicians.

Something must be done.

/// <summary>
/// Aligns two images. Uses highly-local, non-affine alignment.
/// Also compensates for local brightness and colour changes.
/// </summary>
/// <param name=”Input”>The image to align</param>
/// <param name=”Reference”>The image to align it to</param>
/// <returns>The image aligned to the reference image</return>
public Image Align(Image Input, Image Reference)
{
    return Reference;
}

I’ve just added a new download. It’s a screensaver, which displays an image of the Mandelbrot Set, before slowly zooming in on whatever part of the fractal it thinks is probably interesting.

• Code Factory Entry Here
• Direct Download Link Here (7.1kB, requires .NET Framework, which you may already have)

Technical notes:

• The settings screen assumes you understand what the Mandelbrot Set is. If you don’t, the point is that for each point on the screen, the colour represents the number of iterations required before the value exceeds a certain cutoff. You don’t need to know what that means, but the cutoff is one of the options. Since some points (those within the Mandelbrot Set) will never get there, so the maximum number of iterations option lets you choose when it should give up. A higher cutoff will produce smoother colours but take longer to process. A higher number of iterations will produce more detail but again take longer to process.
• The colour step controls how fast the colour changes. A low number will give less contrast but a less psychedelic image.
• The minimum spread to zoom, which must be between 0 and 1, controls how interesting something has to be before it will be zoomed in on. Some parts of the image take longer to process than others, so the maximum difficulty to zoom, which again must be between 0 and 1, controls how long a part of the image could take before it will be ignored. If no area of the screen matching these rules can be found, then it will revert to the entire Mandelbrot Set image and start zooming again.
• After a great many zooms, you start to hit the limits of double-precision floating point numbers, so changing the maximum zooms option puts a hard limit on how many times it zooms before going back to the start. I find the 40th zoom is where it starts going wrong, though to be fair by that point you’re looking at an area that would have been 1% the size of a hydrogen atom at the original zoom level.
• There’s no frame rate limiter in this. If you have a ludicrously fast PC at low resolution, it may run too fast. You could try increasing the first two numbers to silly levels. That ought to slow it right down.
• I have no idea what happens to this screensaver on multiple monitors.

While reading something more sensible on the internet, I was directed, by a Google ad, to a website called “Relativity Challenge”. The author describes himself thus:

Steven Bryant began studying Einstein’s theory of relativity in the mid 90’s with the intent of returning to graduate school to pursue a PhD in physics. During the process, he identified mathematical inconsistencies in each of Einstein’s derivations of the Special Relativity transformation equations. Correcting the problems has led him to produce the model of Complete and Incomplete Coordinate Systems, offering an alternative view of space and time. … Steven has a Bachelor of Science and a Masters degree.

Steven seems unaware that someone with a Bachelor’s degree is a bachelor — that’s why they have a bachelor’s degree. They do not have a bachelor. That would be crazy, but then, there’s a very good chance that Bryant is crazy.

His main thesis is that the derivation of ξ in Einstein’s 1905 relativity paper is wrong. It’s not wrong. I’ve checked it myself, and I also have a Master’s degree (and the knowledge required to correctly apostrophise it) so I must be right. I did it literally on the back of an envelope. It was an envelope from The Cooperative Bank, who will doubtless be pleased at my recycling it so. But wait, Bryant has proof that the derivation I just did was wrong. He invites us to substitute numbers into the various forms of the equation. If they are all equivalent, he reasons, then the answer should be the same every time which, he imagines that he demonstrates, it isn’t:

You can try it. All the answers in the results column are correct for the given values. What he’s done, either cunningly or stupidly, is to use values that don’t correspond to any possible physical universe. The derivation of ξ is done using other equations which state that x′ = x − vt and t = x′v ÷ (c − v), and his values of x, v and t don’t satisfy those equations. It’s like saying “Newton says that F = ma, but given that F = 5, m = 2 and a = David Duchovny, we can plainly see that Newton’s law is flawed.” You can’t make something true using only the word “given”.

So we’ve established that Bryant can’t do maths, punctuation or semantics. He also can’t do physics. As well as the imaginary maths error, he also says that the Michelson-Morley experiment was wrong. The experiment, done in 1887, was designed to measure the absolute speed of the Earth, and the result was that it didn’t have one. Bryant says that this analysis was done wrong, because he’s redefined frequency to include an extra, redundant length term. Then he’s incorrectly added this term into the equations and discovered that now the Earth is going about 30km/s, and therefore relativity is wrong.

He also doesn’t understand that Special Relativity only holds when things aren’t accelerating. He claims that an advantage of his idea is that “the twin paradox goes away”, but it also goes away under General Relativity. His proof appears to involve a lot of talking about cats and birds in cages on trucks. I honestly don’t understand it well enough to find specific errors. It would be like looking for continuity errors in Jabberwocky.

The upshot of all of this is that the universe is broken and E no longer equals mc². It now equals mw², where w is defined as “the speed of the phenomenon in question”. Presumably, therefore, we should put nuclear power stations on trucks and drive them around so they make more power. This also implies that nuclear power should produce a hundred million times less energy than we expect (and observe) it to. You would think that someone would have noticed that.

Bryant has not noticed that.