Fact and Fiction

Thoughts about a funny old world, and what is real, and what is not. Comments are welcome, but please keep them on topic.

Tuesday, November 29, 2005

Blair says nuclear choice needed

A BBC news item entitled Blair says nuclear choice needed reports that

Tony Blair says "controversial and difficult" decisions will have to be taken over the need for nuclear power to tackle the UK energy crisis.

The prime minister told the Liaison Committee, made up of the 31 MPs who chair Commons committees, any decision will be taken in the national interest.

That raises much hollow laughter in those who subscribe to the conspiracy theory of life.

The message is clear that the decision has already been taken to build new nuclear power stations. This much is obvious from the stark fact that most of the current generation of nuclear power stations is nearing the end of its life, and we are also committed to reducing CO2 emissions (which nuclear power stations can do), and we have not made sufficient effort to invest in renewable energy sources such as wind and solar energy (which also reduce CO2 emissions), aaand we have a nuclear weapons programme that we want to support.

On the nuclear weapons issue, I'll bet Trident isn't the last generation of such weapons, because as long as someone can throw a nuke at us we will retain the ability to throw one back at them. That's the logic, and I don't see another credible way out of the dilemma.

All of this points towards the absolute necessity for keeping a large nuclear industry going on for the forseeable future.

Frankly, I am amazed at the pretence and games that Blair and his acolytes play, when it is so transparently obvious what is really going on.

Nuclear power? Unfortunately, we've got no other choices left open to us!

Saturday, November 26, 2005

SETI won't work

The search for extra-terrestrial intelligence (SETI) is running a long term programme that searches through radio telescope data for signals that could have been produced by extra-terrestrial intelligent sources (little green men, if you want). You can participate in this search by going to the SETI at home site, and downloading the software there. A few years ago I began to run this SETI software as a screen saver on my PC, but I never found any extra-terrestrial intelligence, and nor has anyone else.

After staring at this screen saver for longer than most sane people would consider healthy, it became obvious that the search strategy that SETI uses is totally naive, and it is bound to fail in its goal. In a nutshell, what SETI looks for are transmissions that resemble carrier signals, which are the signals that are sent out by transmitters which you "tune into" when you twiddle the dial on your radio (or press the auto-search button). Actually, SETI is rather cleverer than that, because a transmitter is likely to be moving on a curved path (e.g. it is on a planet orbiting about a star) which will make the frequency of its carrier signal vary with time, so the search takes account of the variations in the carrier frequency that this causes.

Why is searching for carrier signals a naive strategy for SETI to use?

The answer is that an intelligent transmitter would not use such an inefficient way of transmitting signals. Also, they might not want anyone to eavesdrop on their signals, so they would use a more stealthy means of communication. The most obvious alternative possibility is something that we humans already do; it is called spread spectrum transmission, which was (surprisingly!) invented by the actress Hedi Lamarr. The trick is to not use one carrier frequency, but to use a random sequence of many carrier frequencies, thus spreading the transmitted signal over the frequency spectrum in a way that makes it very difficult to receive if you don't know the random sequence used. If you combine this with appropriate "whitening" of the transmitted signal, then the spread spectrum signal is indistinguishable from background noise, and it is actually impossible to receive if you don't know the random sequence and the whitening method used. Generally, a transmitter is at its most efficient in terms of both bandwidth and stealth when its transmissions look like white noise, and the means used to achieve this could be much more sophisticated than merely being simple variations of the frequency hopping approach described above.

If we are already using spread spectrum techniques ourselves, then it is likely that an extra-terrestrial civilisation would be doing something that is at least as clever, and probably far cleverer. This is why SETI won't work.

In this week's New Scientist there is an article Looking for alien intelligence in the computational universe in which Stephen Wolfram makes the same criticism of the SETI search strategy. He then proposes that we do a search of the universe of all possible algorithms (i.e. the computational universe) for ones that have behaviours that are a cyber-version of an extra-terrestrial civilisation. This is an application of A new kind of science. Wolfram's proposal is that a cyber-version of an extra-terrestrial civilisation is as good as the real thing. This is complete rubbish, and I am surprised that he offers this as a serious proposal. It is like saying that virtual reality is the same as real reality. They might seem to be the same but they are not actually the same.

I think that the computational universe is a very worthwhile place to harvest, because there will be algorithms out there that do very interesting and useful things, and which could be used as the basis for whole new technologies. This is why A new kind of science is a very good thing that more people should pay attention to. However, this has nothing whatsoever to do with SETI.

I will caveat that last remark. There is the possibility that there are algorithms out there in the computational universe that might be used as a sophisticated way of encoding/decoding transmissions, and which extra-terrestrial civilisations might already be using. All we need to do is to find these algorithms, and we can then eavesdrop on the extra-terrestrial conversation, assuming there are no other gotchas standing in the way, such as the very interesting scenario that is described in Piers Anthony's book Macroscope; read the book, especially if your IQ is 150+.

SETI should focus on reality and not create a virtual simulacrum of reality. Remember that science is about connecting with experimental results; computational simulations are very interesting in the same way that pure maths is interesting, but without an anchor in reality they are not actually science. This distinction is the same as the difference between natural philosophy (now discredited) and natural science.

SETI should also extend their search strategy to look beyond communication using mere carrier waves, otherwise they are going to waste an awful lot of computer time hunting through the radio telescope data.

Monday, November 21, 2005

Super-computer trounced by human brain

IBM's Blue Gene super-computer has performed a record 280.6 trillion operations per second on the industry standard LINPACK benchmark. That is more than 10^14 operations per second.

The human brain contains around 10^11 brain cells, so Blue Gene could in principle deploy about 10^3 operations per second on simulating each brain cell in a simulation of the whole brain.

However, each brain cell has a lot of biological computing machinery associated with it (e.g. its large number of synapses, the dendritic tree that connects the synapses to the cell body, the cell membrane, etc), and the time scale for brain cell dynamics is around a millisecond, so the 10^3 operations per brain cell per second that Blue Gene gives you isn't anywhere near enough compute power for a real-time simulation of the brain.

If each brain cell (including all of its biological computing machinery) needs around 10^9 operations per second (1 per millisecond, times a conservative 10^6 to account for all the large number of synapses and the dendritic tree) , then the total number of operations per second that are needed for the whole brain is around 10^20 (10^9 operations per brain cell, times 10^11 brain cells).

Thus, conservatively 10^20 operations per second are needed for a real-time simulation of the human brain, but Blue Gene can supply only 10^14 operations per second. There is rather a large shortfall.

Clearly, we are nowhere near being able to simulate all of the neurons in the human brain in real time. Even if Blue Gene was fast enough, it would still lose out in terms of processing power per cubic millimetre. Blue Gene occupies a very large room (think "supermarket", or have a look at the photo here), whereas the human brain occupies a cranium.

And don't even think about comparing the relative energy requirements of Blue Gene and the human brain.

Blue Gene achieves its speed by connecting together a large number of conventional microcomputers, each of which uses a variant of a more-or-less standard computer architecture (the so-called von Neumann architecture). In the other hand, the human brain does not use anything remotely like this approach to the problem of doing computations. In effect, it uses special purpose biological hardware for each and every processing node (i.e. each brain cell plus its associated machinery), and this biological hardware operates as a fine-grained parallel computer.

To compete with a human brain, a remote descendent of Blue Gene will have to operate in a similarly fine-grained parallel way, and will thus have to be built using a much less clunky technology than tens of thousands of interconnected microcomputers.

I believe that the solution to this problem will emerge from nanotechnology, and it will use something analogous to artificial DNA to orchestrate the building of fine-grained parallel computers.

Wednesday, November 16, 2005

Endless projects

When I was very young, and I was just becoming aware of a thing called "science", I thought that I could do something really worthwhile if I dedicated my professional life to doing research into fusion power. Even back then (around the time Neil Armstrong was making his "one small step") the prediction was that it would take several decades to get to the point where we could build a commercial nuclear fusion power station. That suited me because it meant that I would be involved in the main phase of the work, and I would retire at about the time these power stations came on line, and they would then keep me nice and warm in my retirement. The idea of infinitely extensible project time scales did not occur to me when I was very young.

Now I read the article Dream machine in the latest issue of New Scientist that the projected year for operation of the first commercial power plant is around 2050, and that is an optimistic projection. 2050 is almost certainly after I am dead. I am so glad that I didn't follow up my idealistic childhood dream of doing fusion research.

I feel much the same way about where I could have ended up after doing my PhD on QCD. I would probably have eventually been seduced by the beauty of string theory, which had a resurgence at about the time that I finished my PhD, and I would have ended up doing research on a long road to ... well, it might be somewhere, but it also might be nowhere (see Peter Woit's blog for lots on this). Ha! I couldn't resist a gratuitous swipe at string theory!

I like the sort of research that can be done successfully using simple equipment (e.g. a standard PC running Mathematica), that achieves useful results with the investment of only days or weeks of my time. The information processing research that I currently do admirably satisfies these criteria.

Endless projects? Not for impatient me.

Blacklight Power

Amazingly, Blacklight Power will not go away. This organisation claims to have something to do with science, and that they can extract energy from hydrogen atoms as follows (quotation taken from here) :

... energy is released as the electrons of hydrogen atoms are induced by a catalyst to transition to lower-energy levels (i.e. drop to lower base orbits around each atom's nucleus) corresponding to fractional quantum numbers ...

Oh wow! I can see how a pseudo-scientist with a poor grasp of quantum mechanics might dream up this idea by noodling around with the algebra of the QM of a hydrogen atom. Fractional quantum numbers? Why not? If we do that then we can get energy levels lower than the previously accepted ground state energy. Fantastic! We can access a new source of energy. Hey, man, pass the bong.

There is a teeny little problem with this fractional quantum number idea. The wavefunction has to be multi-valued or has to diverge at infinity to accommodate this sort of quantum number, and since the wavefunction is the projection of the state vector onto a set of orthogonal basis states, this means that this projection would have to be multi-valued (or have to diverge). What is that supposed to mean? It means you have got it wrong.

Unfortunately, many intelligent people who are not formally trained in science fall for this sort of hocus pocus because of the Parrot Effect. Even if you are only slightly gullible, then provided that you hear about something from either a single apparently respectable source, or from a large number of unreliable sources, then you are likely to believe it.

Blacklight Power? I hope you haven't bought shares in it.

Sunday, November 13, 2005

How to be invisible

Continuing with the "Aerial" album by Kate Bush, there is an absolutely fascinating song called "How to be invisible", in which she explains how it is that she manages to seemingly vanish when she is not wearing her public KB persona.

The chorus of "How to be invisible" is

Eye of Braille
Hem of anorak
Stem of wallflower
Hair of doormat

I assume that this is a description of KB when she is wearing her alter ego. But this is clearly a description of a gypsy/traveller person. Such people often wear anoraks (i.e. hem of anorak) and have thickly matted hair (i.e. hair of doormat). If such a person kept a low profile (i.e. stem of wallflower) then they would effectively be invisible (i.e. eye of Braille).

What a revelation! Now I understand. Thank you to KB for sharing this with us.


The new album "Aerial" by Kate Bush has a song called π (Greek letter p), which is about

Sweet and gentle and sensitive man
With an obsessive nature and deep fascination for numbers
And a complete infatuation with the calculation of π...


She recites π to a large number of decimal places. Actually, she doesn't recite the precise digits of π, and even omits a large block of digits, as has been noticed by many people. Now I don't believe for a minute that KB would make careless mistakes, because she has the reputation for being a bit of a perfectionist. That means that the "mistakes" in the digits are deliberate. How intriguing! This has just got to be a number puzzle that KB has set us to solve.

I don't have a solution to this puzzle. All I have done so far is to listen closely to what KB sings and how she sings it, and I have listed what I hear below.

I break up the digits into groups the same way that KB does (actually there is more structure in the way she sings the digits than I show below), I use bold font to denote digits that she particularly stresses, I put square brackets around digits that depart from the true π sequence, I use "BV" to show a contribution from the backing vocals, and I use "?" to show where I am not sure what I hear.

[BV chorus]
[BV mini chorus]
2307816406286208[22 digits omitted]8214
8[BV 0?]
[BV chorus]
8230[BV 0?]664709384460955058223

I have had lots of ideas on how to crack this one, but I'm not so obsessed with numbers that I am going to waste a sunny Sunday sitting at a computer working on the problem. This one gets solved in "slow time".

Has anyone got any ideas on solving this number puzzle? One idea that I reject out of hand is the suggestion that there isn't a number puzzle here!

Monday, November 07, 2005

Bad apples in science

In science you occasionally get bad apples, i.e. scientists who fabricate results. Fortunately, science is a self-correcting activity in which results have to be reproducible to be accepted, and theories have to be falsifiable in order to count as science. If nobody else can reproduce a result that is obtained by only one scientist then it is erroneous, no matter how talented that scientist is. A theory that makes no predictions about things that we can measure is not science, no matter how elegant or philosophically deep that theory is.

There are some famous recent examples of bad apples in science, such as J Hendrik Schön and Luk Van Parijs, both of whom lost their jobs and no doubt their careers as a result of their reckless dishonesty. The self-correcting nature of science worked for us by exposing these rogues, but the fall-out is that people trust scientists a bit less than they did beforehand. I wonder how many more as yet undiscovered cases of fraud there are out there.

What I don't yet understand is why would any scientist think that they could get away with fabricating results? They must know that other scientists will be attempting to reproduce their results. They must know that they won't receive any long-lasting recognition until other scientists actually do reproduce their results. No Nobel prizes have ever been rashly awarded because someone suddenly produced a surprisingly impressive result.

Is it possible that some scientists can so completely delude themselves that they really believe their faked results? I know self-delusion can easily happen, but usually a good night's sleep and a pause for reflection cures the problem. Maybe some people can't reset themselves in this way.

Is it possible that the demand for a continuous stream of publications forces some scientists into grey areas where they publish what they think their results would be if only they had the time to do the work? I am very familiar with this pressure to publish according to a schedule that is determined by "bean counters", and it isn't nice.

Tuesday, November 01, 2005

Trick or treat - no thanks

When I was a child "trick or treat" was an American tradition. Now I am an adult it has migrated to the UK as well. The problem is that the kids expect a treat (i.e. cash) in return for doing nothing whatsoever, not even fancy make-up and an imaginative costume.

Last year my local newspaper published a picture that you could post in your window to say "no thanks" to trick-or-treaters. I decided that with a few modifications the picture would not only say "no thanks", but that it would also enter more positively into the spirit of Halloween.

Here it is (and it works as well!):