Saturday, May 31, 2014

Another picture with cat

A mysterious picture which arrived today .. (referencing this earlier post).

The original was taken, with Clare, down by the river Axe at Wookey Hole.

The hell of changing the router in a connected house

Today my BT Home Hub Wifi died and I entered the hell which is the installation of a new router.

I happened to have a spare BT Home Hub which I duly installed via the CD, carefully arranging not to have BT-Yahoo as my home page and not to be parasitized by the Yahoo toolbar.

My first mistake was to change the WiFi security password. The laptop immediately dropped the wireless connection and refused henceforth to recognise the router - and could not be persuaded otherwise. Trying a re-installation with the CD, it wouldn't accept my preferred password. So back to factory defaults and at least the laptop can now access the Internet.

What's easy? Mobile phones and tablets are no problem. They find the new SSID and the password is accepted no problem. Other devices are more problematic.

It seemed impossible to configure the D-Link cameras to the new router wirelessly. I have had to connect via Ethernet (in one case by dusting off my old Netgear Powerline devices).

The printer was a pain when I forgot that entering hex using UPPER CASE characters is never going to work. Eventually I did it again - properly - tapping away on the front of the printer's tiny screen.

The Sky Wireless Connector, used for iPlayer and similar, is reconfigured through the Sky Box itself and after (a) refusing to throw up the necessary screen, and (b) denying it had succeeded, it eventually worked.

It's amazing how many devices around the house are wirelessly linked to the Internet, and ludicrous that it took me two hours of messing around with diverse management interfaces to get most of these devices rehomed.

What's needed is a standard all-device management interface which is universally implemented, so that a home management system can do the whole thing once and for all. The amount of human input actually needed is zero - it should all be plug and play.

Tuesday, May 27, 2014

Paintings with Cats

Genius from Russian artist Svetlana Petrova.

Grant Wood's American Gothic

Leonardo da Vinci's Mona Lisa

Titian's Venus of Urbino
I'm still meandering through Jack McDevitt's 'Alex Benedict' series. A well-written blending of far-future science-fiction and detective-story. They're rather slow, but well-written and engaging; I don't know if I'll have the energy to get to number 6 (having just finished 3 - 'Seeker').

What else is on my Kindle app right now? Mindbridge and Slan (and 'The Devil's Eye').

My current non-virtual book is Francis Fukuyama's "The Origins of Political Order: From Prehuman Times to the French Revolution". Although I'm in the early sections I really can't praise this book highly enough. Non-ideological writing - ie without an agenda, scintillating prose style and super-intelligent analysis. If only journalistic pundits started from where this author leaves off we might be having grown up debates about politics and strategy and maybe getting things somewhat right.

Sunday, May 25, 2014

Northumberland in late May

We've been away for a few days in Northumberland, staying at Alnmouth, north of Newcastle. Here are some pictures.

Nigel and Clare at Warkworth Castle

Seals on the Farne islands

Clare at Alnmouth beach

Another view of Alnmouth beach

At the Kielder Observatory
The Kielder Observatory visit was by degrees intriguing and infuriating. The audience - c. 40 members of the public (all of whom had paid £15 per head and had undertaken a long drive to get there) - appeared in the large to be uneducated in things astronomical. Our host was the observatory's Director, Gary Fildes who
"... did not have a University education, indeed he left school at 17. Before we go 'tut-tut' we remember that the First Astronomer Royal, John Flamsteed, did not do either. Nevertheless, he made rather a good job of running the Royal Greenwich Observatory from 1675 to 1719, when Edmond Halley took over.

Gary, at present a Fellow of The Royal Astronomical Society, started his working life as a builder and after serving his time became a site manager, before founding his own business, Orion Bricklaying Services, which is still trading.

Gary's hobby of Astronomy and his astonishment at the scientifically dark skies of Kielder led him to take the giant step of founding the Kielder Observatory in 2008, after a number of years of 'site testing' and fund raising. "

(excerpted from here, the occasion of Gary's honorary MSc from Durham University).
Gary started the evening with a half-hour talk on astronomy. In my opinion this was not a success as Gary doesn't appear to have a deep understanding of the science of astronomy (physics, cosmology). It was more the chat of someone who has read a bunch of popularisations. I was particularly unimpressed by the suggestion that the universe is teeming with intelligent life but we don't see the aliens because we're in quarantine due to our bad ways. I think it's fair to say that that plays better at Glastonbury than in circles scientific.

The telescopes are able to image the planets as discs, but they're quite small (we've all been spoilt, of course, by flyby imaging and the Hubble Space Telescope). Mercury was just a point of light; Mars was a featureless disc; Jupiter's bands were visible as well as three of the Galilean moons, while it was possible to see Saturn's rings.

It did, however, get me thinking about the first people (Galileo et al) who looked up at bright planetary pinpricks, then down to their telescope eye-pieces where suddenly they found astonishing shape and structure. The heavens would never be the same again.

This wow factor was undermined by the amateurish, half-baked and confused presentations but it's hard to see how things could improve. The sight is remote, the hours profoundly antisocial, the temperature cold and the number of public outreach events high. Apart from Gary's small team of dedicated, but amateur, volunteers it's hard to see how they could attract a more educated set of presenters. Perhaps Gary should get someone to write a better script for himself.

Here are some more pix.

Friday, May 16, 2014

Cat up to his old tricks

Dead and drowned
Adrian threw the corpse out of the kitchen last night; it ended up in the mop bucket. Clare tipped it out this afternoon. A young life tragically cut short.

Enjoy the sunshine! Vitamin-D pills don't hack it

From The Times today.

There is very poor evidence that vitamin D supplements reduce the harm from having a low blood level of vitamin D (“Sunbathing, sunscreen and vitamin D”, May 10). This raises the question whether there are other benefits from sun exposure that aren’t detected from the vitamin D blood test. Dermatologists from Edinburgh showed that one reason why vitamin D supplements did not reduce blood pressure but sunshine did was that the skin exposed to UV also produced nitric oxide, a substance that is well known to reduce blood pressure. There could of course be other unknown benefits of sunshine.

Cancer Research UK is too cautious in recommending a few minutes a day without sun protection — white skin needs from 30 to 45 minutes before maximum vitamin D production is achieved and then turned off. It also forgets about DNA repair enzymes (the chief reason why even the worst known human carcinogen, ie, tobacco smoke, takes 40-50 years to cause cancer). Most of us, despite living all our life exposed to a multitude of carcinogens, don’t get cancer.

For more than 30 years radiotherapists have known that 90 per cent of DNA damage caused by a single dose of radiation is repaired within two hours. This is the basis of how they safely fractionate treatment to cure cancer and not cause more harm than good. I do not believe CRUK has the evidence to say it is not the same for UV and as a result I think it has probably been doing more harm than good for the past 20 years.

Tim Oliver

Professor Emeritus in Medical Oncology, QMUL

I was resting on a camping-bed this morning in the sunshine for 45 minutes ..

Thursday, May 15, 2014

Guessing a PIN

A few posts ago I was musing about the security of a PIN-protected smartphone. We all know that if you choose a four digit PIN (0000 - 9999) then on average it would take 5,000 guesses to break-in. But how many attempts would give a thief, say, a 5% chance of breaking-in (i.e. a 95% chance of you being safe)?

This matters because you want to set a thief a lengthy guessing-task, which still has a low probability of success.

So we want to know the probability of bad-guy success (guessing the PIN) in X attempts from N possibilities (where N = 10,000 for a four-digit PIN) .. as X varies. The distribution we want is one where we keep taking samples (i.e. guessing four digits) without replacement until we sample the one success, then the thief stops and prepares to ransack your account.

My first intuition was that the curve would be cumulative-binomial. The first few trials you have no information so the probability of an early guess working is rather low. About half way through the process, you have rejected about half the possible values so now your chances of getting the PIN are rising, while at the end you've rejected so many wrong choices that your chances of the next guess being right is almost certain. But I was wrong (about the binomial).

Cumulative Binomial Distribution Function

After some googling I came upon this. It turns out that the probability of taking exactly X attempts to correctly guess the PIN is exactly the same, 1/N, for any sequence-length of guesses. So in the case of the four-digit PIN, your chances of guessing right the first time are 1/10,000 .. as are your chances of guessing right on the second, third and indeed ten-thousandth attempt.

(Note: this means that all sequence lengths are equally likely before the thief has started his guessing game. Obviously the conditional probabilities change as the thief is going along - if the first 9,999 guesses are wrong, the final guess will be correct with probability one. This is explained in the link.)

The actual distribution is rectangular - what is called a uniform distribution. If the PIN can take N different values, then the x-axis registers the length of possible guess-sequences going from 1 to N, and the Y-axis shows the probability of guessing correctly in that sequence-length. The distribution is a horizontal line of height (i.e. y-intercept) 1/N. For absolute clarity, we are considering the space of all possible guessing-sequences from length 1 (first guess gets it) to N (last 'guess' gets it). The thief will actually execute only one sequence but doesn't know in advance the length of the one which will work. We assume he doesn't give up. Since the sequences are mutually-exclusive (you get it in one, or you get it in two, or you ...) we can add the various probabilities in what follows.

Example: uniform distribution for N=6 possible choices

If your PIN is four digits then N=10,000 and the chances of correctly guessing your PIN in guess-sequences ranging in length from 1 to 500 attempts is 500/10,000 = 1/20 = 5%. If an attempt takes 6 seconds, this is 50 minutes. Note that searching the whole space of 10,000 choices takes twenty times as long = 1,000 minutes = nearly 17 hours.

If your PIN is six digits then N=1,000,000 and the chances of correctly guessing your PIN in up to 50,000 attempts is 50,000/1,000,000 = 1/20 = 5%. If an attempt takes 6 seconds, this is 5,000 minutes = around three and a half days.

The security services advise nothing shorter than six digits for secure material.

Ten theses on systems and agency

1. The brain has different cognitive subsystems addressing systems thinking and agency.

2. Systems thinking - systemics - is mathematical, impersonal and logical.

3. Agency thinking personalises, attributing motivation and intention.

4. Systemics uses the conversational tools of reason, evidence, theories, deduction and consequences.

5. Agency uses the conversational tools of systemics, but applies them in the context of intentional agents and often subordinates them to emotionally-driven arguments (which can be logically-incorrect but surprisingly effective).

6. Systemics is suitable for dealing with the natural world (maths, physics, chemistry) and inanimate aspects of the human world (architecture, finance, the military, the law).

7. Agency addresses the natural world through ideology (religion) .. and the human world through persuasion and condemnation (politics).

8. Masculine thinking tends to the systemic, with Asperger's being a boundary case. This leads to increasing degrees of social incompetence -  society is agency.

9. Males who incline more to agency-thinking (actors, writers, politicians) often exasperate the more systemic by their illogic in the services of emotionally-affiliative positions. (Matt Ridley of The Times is an occasional offender).

10. All of this is separated from empathy, which is an affective, not cognitive function. Plenty of high-systemic individuals are enormously pained by discord, cruelty or pain. Sociopaths are something entirely different.
I was thinking about this while reflecting on some of Neal Stephenson's jokes in Anathem, jokes on the more intellectual of his characters (avouts) who show a woeful lack of insight into the human condition despite their impressive grasp of mathematics; these are also jokes at the expense of his likely readership.

I put my own hands up here.

Wednesday, May 14, 2014

Tablet and smartphone security

When my new Nexus 10 tablet arrived my first thought was to reproduce my Galaxy S3 experience. I downloaded the same set of apps and put a lengthy security code on it. But that doesn't really work. The security soon had to come off as the tablet became a shared and "to hand" family facility. It's got the newspaper, the crossword for Clare and it's good for books and the iPlayer.

Where's the sweet spot between usability and security? What I've done is uninstall Dropbox and Blogger, and not install any financial apps (Bank etc). It wouldn't be great to lose it but at least the impact would be bounded and more manageable.

The smartphone is altogether better protected but only I have to navigate its security.


I have a postal vote for the European elections coming up this Thursday. The ballot paper had something like eight parties: Out of Europe; English Independence;  British National Party; Conservatives;  Lib-Dems; Labour;  Green. I might have missed one.

In the spirit of Gregory Clarke's "The Son Also Rises" I did a bit of surname analysis. The small right-wing parties had bluff,  artisan names and were overwhelmingly male. The Greens were the most aristo and treble-barrelled. The Tories had the most women (4/6) and the soggiest tag-line ("for change in Europe"). The BNP had the compelling slogan "for a better Britain". Labour had the only non-Anglo but the other names were pure Hampstead - and there was no slogan.

Naturally I voted for the party which in my view was most likely to deliver a better Britain.

I'm reading "Anathem" by Neal Stephenson. It's a big book situated in a variant universe where intellectuals are locked away in "concents" as a secularised variant of monks. There are strong similarities with our own history,  and history of ideas (Immanuel Kant makes a disguised appearance). There's something rather secret and not quite right in orbit. Reviewers have criticised a slow first 100 pages and an excess of invented vocabulary. Neither is actually a problem: Stephenson's writing is designed to be immersive and the strange new words are easily guessable (when the author isn't explicitly providing definitions) and are needed to lock-in the essential otherness of this reality.

Oh, and we bought some new sleeping bags for our summer travels.

Friday, May 09, 2014

"A Troublesome Inheritance" - Nicholas Wade

Despite all you hear to the contrary, human evolution has continued up to the present: “recent, copious and regional” in Nicholas Wade’s words. This would appear to be of great scientific interest – what are the genetic differences between different races and what are they coding for? Unfortunately we have a ‘blank slate’ social-sciences establishment which denies even the existence of races and seems unwilling to admit to genetic input into such obviously inherited attributes as intelligence and personality. (‘Obvious’ means that everybody knows it really and that careful analysis, including twin studies, bears it out).

Is the author going to be sacrificed on the altar of ‘racism’?  Wade is naturally keen to defend the scientific study of human racial differences from the inevitable charges of racism and eugenics. This is a difficult discussion and one the author approaches historically, showing how American ideas of Caucasian superiority and eugenics were eagerly picked up by the German National Socialists. But in the urge to be on the side of decency one has to be careful to reason accurately:
“By analogy with animal breeding, people could no doubt be bred, if it were ethically acceptable, so as to enhance specific desired traits. But it is impossible to know what traits would benefit society as a whole. The eugenics program, however reasonable it might seem, was basically incoherent.” (p. 27).
Well, it seems that intelligence, a generally pro-social personality and good health are pretty good candidates for traits which would ‘benefit society as a whole’ and later the author will argue that western medieval societies effectively bred for those traits over the last thousand years to our advantage. We do our own private experiments in positive eugenics whenever we seek out the best possible marriage partner.

Such sloppy argumentation is, sadly, not uncommon in this book.

Having got his defences out of the way, Wade now gives us some science -a comparison between chimpanzee societies and our own. Chimps are highly aggressive and promiscuous; humans not so much. In fact the key differentiator is our marked ability to cooperate. Wade has some plausible ecological suggestions as to how these differences might have emerged and can back up the behavioural stuff with genes coding for hormones such as oxytocin (increased trust within an in-group) and mono-amine oxidase (associated with aggression).

The gene which codes for the latter, MAO-A, comes in different alleles – the “two-promoter group” in particular is linked with criminal violence. In a large study, (p. 55), Jean Ship and colleagues found that African-American men had a 5% chance of carrying the “two-promoter” allele (these were predominantly the delinquents). In Caucasians the proportion was 0.1%. Clearly we are at an early stage in this research but the correlations are certainly thought-provoking.

The next few chapters reprise the story of the out-of-Africa human expansion and how this is captured in genetic sequencing. This will be familiar to anyone who has checked out 23andme or similar companies. The arguments for the objective existence of races are kind of obvious to anyone without an agenda, and are apparent in analysis of allele frequencies. Nevertheless, it remains true that in 2014 we know next to nothing about what most of these variant alleles actually do. Most cognitive, psychological and behavioural traits are under the control of hundreds or thousands of alleles, each of small individual effect, which accounts for the ‘bell-shaped curves’ we see in population intelligence and personality attributes. This is a central problem for a book which is trying to create a compelling connection between race genetic differences and the distinctiveness we see today in human civilizations such as in the West (America, Europe), the East (China, Korea, Japan) and sub-Saharan Africa.

Such differences clearly interest the author and drive the second half of the book - which he deems ‘speculative’.  The author is particularly interested in how humanity made the transition from its default social model, kin-based tribalism, to states and empires. Citing Francis Fukuyama (“The Origins of Political Order”), Gregory Clarke (“Farewell to Alms”) and Daron Acemoglu & James A. Robinson (“Why Nations Fail”) we are taken on a multi-millennial tour of the great civilisations of the world. What was the evolutionary impact of these novel social environments on population psychology? His thesis is that of gene-culture coevolution – that people were selected for ‘tameness’ and prosociality as well, perhaps, for greater intelligence. There are good reasons for believing this is likely, but it has to be said that it may take a few more decades to get compelling genetic evidence.

The “IQ and Global Inequality” authors Richard Lynn and Tatu Vanhanen get short shrift (p.191). They have demonstrated high correlations (c. 80%) between measurements of national IQ and GDP.  Given the well-known strong genetic underpinnings of observed intelligence (in societies without widespread material deprivation) it’s surprising to see Wade backsliding into confusion here. By highlighting poor and outlying data he makes his doubts clear but it just reads like he has an agenda. Perhaps he thinks his book is controversial enough as it is.

The final chapters cover the astonishing intellectual success of the Ashkenazi Jews, citing the well-known work of Gregory Cochran, Jason Hardy and Henry Harpending (written up in “The Ten-Thousand Year Explosion”) and further speculate about the reasons behind the rise of the West in recent history. Interesting but does not break any really new ground.

This is a readable book but one which lacks a strong sense of direction. Occasional sloppy reasoning and ideological argumentation can irritate. Gregory Cochran has identified some errors of fact (on the “West Hunter” blog) which highlight the limitations of the author’s genetics sourcing but this is not where Wade’s real interests lie. He is clearly infuriated by the assumption of human genetic uniformity in economics, history and public policy and is consequentially highly motivated to give an account of the world as if human and racial differences actually mattered. In the current absence of hard genetic results his views remain plausible but speculative and will probably be ignored by politicians and policy-makers for another generation, more’s the pity.

Thursday, May 08, 2014

An interstellar colony?

David Waltham's book "Lucky Planet" suggests we may be the only intelligence in the visible universe. This, he suggests, is because of the difficulties in stabilising a planetary climate over the multi-billion years required to evolve intelligent life.

Of course, there will be plenty of planets in our galaxy which could (perhaps with a degree of terraforming) support sustainable human colonies now we exist - if we could only get there and put the work in. We have just 500 million years before the Earth becomes uninhabitable.

The Wikipedia article on interstellar travel suggests that sending conventional massive spacecraft with loads of people and tons of equipment between the stars is really a non-starter. What is the minimal payload it makes sense to send?

Assume we can image and research target exoplanets from the solar system. We want one with oceans, land and at least carbon dioxide in an otherwise benign atmosphere. Send a small package of photosynthesising eukaryotic cells and let them propagate in the ocean. Wait a billion years for intelligent life to evolve and get around to building radio telescopes. Send them engineering schematics for our own DNA and cell structure,  artificial wombs and AI nannies. We're there folks, with the maximum of bits and the absolute minimum of atoms.

Wait, you say, it's all too slow.

In which case we have to send more atoms first. It's always better to send bits at the speed of light and reconstitute complex objects at the other end, the issue is how to bootstrap the process. The minimal remote terminal has to decode designs sent using (e.g.) modulated laser light and fabricate stuff using a 3D printer type of thing. Sadly, it takes an advanced technological civilization to build and operate those kinds of things - far too much for our first nanoprobe landing on a moon under another star.

I suggest the only way we're going to get this interstellar colonisation thing to work is by copying the only effective bootstrap process we know - life itself. Starting from a nanoprobe seed we have to design cycles of incremental development using only local resources which ends up with a fully developed remote fabricator - in effect a matter transmitter. Then let terraforming and colonisation begin!

Note: designing a suitable techno-pseudoevolutionary process is left as an exercise for the reader.

Wednesday, May 07, 2014

A plausible solution to the Fermi paradox

Review of "Lucky Planet: Why Earth is Exceptional - and What that Means for Life in the Universe" by David Waltham.

It’s been fashionable and perhaps even comforting to believe in the essential unity, benevolence and even environmental-competence of life on Earth. The Gaia hypothesis makes us feel good, but hard-nosed evolutionary biologists and planetary scientists crunch the numbers and just can’t get it to work. Forget the galaxy of a billion friendly alien civilizations, perhaps there’s just one: ourselves. Perhaps we’re just very, very fortunate.   Here’s a much abbreviated summary of what David Waltham has to say in this lively and intelligent book.

Our very existence shows that the Earth has experienced life-friendly climatic conditions for billions of years. During this time the output of the sun has increased by 30% while early high levels of greenhouse gases such as methane, water vapour and carbon dioxide have been almost scrubbed from the atmosphere. These changes ought to have produced enormous and lethal climatic variation yet somehow, by some magic, the effects have largely cancelled out.

For some people, this shows that powerful negative feedback mechanisms are at work, stabilising the climate for life. Strange then, that such benign processes are so hard to pin down. The alternative view is that for most planets like the Earth, the climate did indeed transition to fire or ice, with the consequent destruction of any biosphere; the Earth is special and very, very lucky.

 Of course, the fact that we’re here at all to make such an observation indicates that for the Earth it could hardly have been otherwise. This is called the principle of Anthropic Selection - to be contrasted with the Principle of Mediocrity, that the Earth is not that special in the universe.

David Waltham systematically takes us through the unique features of the Earth.  Our star, the sun, is unusually large and bright – most long-lived stars are smaller and redder than ours. However, they are prone to stellar flares which are extremely harmful to the biosphere. The Earth has an astonishingly strong magnetic field which deflects the solar wind, which otherwise could split water vapour into hydrogen and oxygen allowing the former to escape into space – this is how a planet loses all its water.

Despite the early sun emitting only 70% of today’s output, the Earth remained suitable for life due to the immense greenhouse effect of the early atmosphere. As the sun heated up, the greenhouse effect reduced in tandem: carbon dioxide was washed out of the atmosphere by rain and locked up in sedimentary rocks, while methane was oxidised away as soon as early photosynthesis evolved.

The Earth did not experience a smooth, stabilised, homeostatic climate – there were episodes of great heat interspersed with at least four ‘snowball earth’ episodes where the entire planet became icebound. Thanks, however, to plate tectonics and volcanism, carbon dioxide was released back into the atmosphere to unfreeze the Earth and to allow early life to reboot.

Some people believe that this is an example of the Gaia principle – life stabilising its own environment. The author sees instead systems of climate dynamics that could so easily have sheared off into uncontrolled positive feedback or blundered into wild oscillations. In his opinion, this is exactly what happens to most planets like ours ‘out there’ - but as a consequence, they have no observers to later theorise about it.

Parenthetically, the author’s concerns about current anthropogenic global warming are consistent with his view of underlying instabilities. It’s not so much that increased carbon dioxide levels in the atmosphere directly warm the climate; it’s more that they catalyse changes in more potent greenhouse gases (water vapour, methane) and it’s not at all clear that there are negative feedback mechanisms which could dampen their effects. The climate models are very complex and who knows if they’re either comprehensive or correctly tracking all the mechanisms?

Another climate-changing influence is the Earth’s axial inclination (currently around 23 degrees) and its orbit around the sun. Under the impact of the other planets in the solar system, the shape and tilt of the Earth’s orbit is continually changing on long-period cycles (69,000 years for orbital tilt and tilt-direction, 400,000 years for orbital eccentricity with other influences clustering around 100,000 years). These affect solar heating and drive the ice ages. The Earth also precesses on its axis every 26,000 years. We’re very lucky that these numbers are rather different because if they converged we would experience orbital resonances, and the inclination of the Earth’s axis would become unstable and chaotic (of the order of a few million years). This would trash the climate, leading to the extinction of all complex forms of life. How did we come by that luck?

It’s somewhat well-known that our large moon ‘spin stabilises’ the inclination of the Earth’s axis. What is less well-known is that as the moon continues to spiral away, the precession rate will slowly decay and in 1.5 billion years time resonance will occur with the orbital periods discussed above. At that point, the Earth will have an unstable spin axis. This is of academic interest only, as for reasons concerned with the sun’s increasing output, the earth will become uninhabitable for multi-cellular life within the next 500 million years. But, if the moon’s radius had been just 10 km larger and the early Earth’s day just ten minutes longer, the Earth would have an unstable spin axis today. What are the chances?

So why does it pay to have a large moon? The author suggests that a moon almost large enough to eventually generate axial instability also stabilises the spin prior to that, and in doing so allows the planet to have relatively mild and infrequent ice ages - another case of fine-tuning for intelligent life.

The author concludes that the chances of all these things coming together to guarantee a four billion year life-benign climate are so remote that the Earth is possibly the only planet with intelligent life in the entire visible universe: we are quite alone. This solution to the Fermi Paradox might be considered depressing, but it should increase our caution – “We may just find out the hard way that planets with nasty climates are quite easy to produce.”

The reader may be left with another thought: although few planets may experience multi-billion year climate stability, this is hardly a pre-requisite for interstellar colonisation, and there’s a lot of unoccupied real estate out there.

Sunday, May 04, 2014

Reading Skiplex - first impressions

We were at Reading Skiplex yesterday for the first time. There are two ramps like the one below in a bright, spacious, light-industrial-type building. Stepping onto the stationary mat (resembles a thin-texture company carpet)  is a slightly strange experience: the static friction is quite high and your skis don't slide. As the belt is turned on there's a sharp jerk and the friction is suddenly far less. Occasionally the instructor hoses the surface down as you'll see in a video in a moment.

Alex and instructor Ollie

We all, in our various ways, found it difficult. Skiplex is extremely unforgiving of less-than-perfect technique and getting it wrong means the skis catch and you fall over, or you're swept off the mat area. The instructor is impressively fast with his portable kill switch.

Skiplex is progressive in its skill development: see here for the Skiplex logbook with its ten levels. The question we asked ourselves: how much are you just learning to ski on the Skiplex mat with all its strangeness and uniqueness, and how much of this transfers to snow as value-added? As you would expect, the staff are reassuring on this point - but we think it needs a little more exploration.

Here's a video of Adrian, an experienced skier, dealing with Skiplex in its faster-steeper mode. We did some more videos which can be found in this folder.

At the end of our session, one of the staff told us about a snowboarding instructor who had cranked the machine to its full speed and gradient on his first day of employment. Trying to do a turn, the snowboard edge had caught, and he'd been propelled into the air. Two somersaults later, he banged his head landing, and was knocked out. That instructor now concentrates on skiing.

We were also told that significant progress can be made in six lessons (their package) but it depends on getting stuck in and not being frightened of falling over and acquiring bruises. Never, it appears, was 'no pain, no gain' more apt.

Skiplex progress record

We were at the Reading Skiplex yesterday. For more on that click here. However, at the end of the assessment session you get a logbook where your progress is documented (this is normal at ski schools). Here's mine - but it's chiefly useful to review the overall syllabus. Click on any image to make it larger.

Friday, May 02, 2014

Bad writing ruins good ideas ("The Cooperative Gene" - Mark Ridley)

After the late heavy (asteroid) bombardment of the Earth 3.9 billion years ago, single-celled life seems to have evolved almost immediately. These were the prokaryotes (simple cells without nuclei). It then took a further 1.6 billion years of prokaryotic stagnation before complex cells evolved (eukaryotes) - the kind of cells which form all animals and plants. The evolution of multicellular life then took a further billion years. These are big gaps.

Here's a time line.

Mark Ridley believes that many of the delays in life's tangled and protracted history are due to the problems of overcoming DNA copying errors. This is what he has to say in the preface to his book "The Cooperative Gene: How Mendel's Demon Explains the Evolution of Complex Beings".
"The DNA in a human being is 6600 million letters long and codes for about thirty thousand genes. In contrast, the DNA of a bacterium is two or three million letters long and codes for two or three thousand genes. Copying mistakes will have become more numerous as the DNA grew longer, for much the same reason as they happen when we are copying written text.
"The evolution of complex life required mechanisms to deal with copying mistakes in the DNA. The first mechanisms improved the accuracy of the copying itself. The earliest life forms probably made about one copying mistake in 100 letters, but bacteria had reduced the rate to less than one mistake in 1,000,000,000 letters.

"This huge improvement is due to the use of DNA for the master-copy — it is an impressively error-proof molecule — and a molecular machinery for proofreading and repairing mistakes. But the possibilities for improving the accuracy of copying seem to have been exhausted by the bacterial stage. The basic DNA copying machinery has remained much the same since then, and we make copying mistakes at a similar rate per letter of DNA as bacteria do.

"Our total error rate is much higher, however, because we use so many more letters of DNA code. Between bacteria and us the length of the DNA molecule has increased 1000-fold and the DNA has come to be copied 100 times per generation, against the once per generation of a bacterial cell.

"Our total error rate has gone up 100,000 times, and whereas a bacteria makes a mistake once in every 1000 offspring, we make over 100 mistakes in every offspring. It is something of a paradox how we can persist, while making so many copying mistakes in our DNA. The solution is uncertain, but is probably — sex. Sex can act to concentrate the copying errors in some of a parent's offspring, leaving other offspring relatively error-free. Sexual life forms could evolve to be more complex than clonal life forms."
Ridley's book is a tribute to how much can be explained by starting with one fundamental engineering-type puzzle - how can you faithfully copy DNA, an essential step in evolutionary replication? - and use it to account for many apparently-diverse phenomena:
  • the stretched timeline of evolution itself
  • the reason for sex
  • the strange process of meiosis ( 2 => 4 => 1 chromosome)
  • ageing (we can't accurately clone our own DNA over too many cellular generations)
Perhaps future genetic engineering technology can address this final problem.

The New York Times reviewer had the same problem that I guess all readers experienced with this book: it's really badly written. Professor Silver's thorough review attempted to be diplomatic:
"The title of Ridley's book suggests an alternative perspective on the gene's role in evolution from that presented by Dawkins in ''The Selfish Gene.'' But Ridley's ideas are not that different, and they've been expressed in more lively prose by others. Like the exodus of Israel from Pharaoh's Egypt, the story of evolution must be told anew to each generation. In both cases it is probably best to rely on earlier texts."
Here are some of my own words for Ridley's writing: stodgy, indigestible, use of inappropriate student humour, gratuitous (and adolescent) attacks on the Bible.

It is possible to plough through the dense and indigestible prose, and the persevering reader is rewarded with some truly paradigm-changing insights. Professor Silver is not quite right: Dawkins didn't already write this material - one just wishes he had.

Thursday, May 01, 2014

Gosh! We're not all clones!

Busy recently: Alex & Adrian were here for for two weeks over the Easter - we hung out at the Mendip Snowsport Centre and tried not to crash as we worked on our parallel turns. Then I suddenly realised that the holidays are fast approaching .. we're away in France and Italy in the summer and the logistics are frightening: route planning, campsite selection, parking for tourist visits to tightly-packed Italian towns. Why don't people routinely quote GPS coordinates (latitude, longitude) to all sites of interest and/or utility? Hint: Google Maps and right click on a location, select "What's here".

More here on the theme that human beings are physically and psychologically diverse, so we need to get over it. More specifically, this self-evidently also applies to different ethnic groups when we're not being too polite to notice.

Steve Sailer writes:
"Nicholas Wade, the New York Times’ chief genetics reporter, has published 1,052 articles in the newspaper of record since 1983. For most of this century, Wade has been methodically waging war in the Science section of the NYT against the liberal creationist myth that race isn’t real. He has now written a definitive book on the existence of biological differences among races, A Troublesome Inheritance: Genes, Race and Human History, which will be published in May."

Here's a brief extract from Wade's book.
"The view of economic development generally taken by economists is that people have little or nothing to do with it. Since all humans are identical units that respond the same way to incentives, at least in economic theory, then if one country is poor and another rich, the differences cannot have anything to do with the people but must lie in institutions or access to resources.
"On the basis of this theory, the West has spent some $2.3 trillion in aid over the last 50 years without managing to improve African living standards. Could something be not quite right with the theory?"
He can expect to be excoriated for the crime of noticing uncomfortable truths. It's good to support bravery by buying the book, and as he's smart, well-informed and a good writer, we may learn something to our advantage. Because of widespread liberal denialism, we're stuck in the rut of simply asserting that such differences truly exist and have significant real-world consequences. But what then are the public policy options? They seem almost too scary to think about - and how tragic is that?