Saturday, February 6, 2016

Four bad messages from a Mythbusters episode

I had high hopes for the Mythbusters when they started, but these hopes were quickly squashed. Now in its last season, the Mythbusters have become a perfect representation of the 'people who "love" science, as long as they don't have to learn any.'

The recent episode "Driven to destruction" had four clear, though not explicit, messages; all of them were anti-science messages. Here they are:

I - Don't bother checking existing knowledge or consulting field experts

Adam wants to lift a car using only the suction of a vacuum cleaner to attach the car to the crane. He builds some suction cups and then places them on the car, without any consideration of the distribution of mass (and therefore of the lifting force necessary) in the car.

If Adam had consulted a mechanical engineer (or anyone with enough of an interest in mechanical structures to read a couple of books), he'd have learned that to lift an heterogeneous object using multiple attachment points to distribute the load, one needs to consider the distribution of mass and not just the total mass.

But here, like in most if not all episodes, the Mythbusters spurn extant knowledge and actual expertise and decide to pretend that science is 'make stuff up as you go.'

II - Calculations are boring, but show pretty charts (and formulas)

Adam's rig doesn't lift the car, it just creates attachment. Computing the attachment force is a simple matter: the total force is the maximum sustainable pressure of the system (vacuum cleaner motor fighting the atmospheric pressure on the output side, seals fighting it on the contact boundaries) times the surface of the attachment. This would be simple enough to measure and calculate (and then multiply by an engineering safety factor to account for faults).

Instead we get a chart about "linear relationship," which is true enough for the purposes of lifting the car, but doesn't even show what the calculation is. Also, because of the lack of expertise in how distributed lifting works, the calculations are actually quite dependent on where the attachment points go and therefore not linear at all.  (The point of saying "linear relationship" is to teach the audience yet another identity phrase.)

(There were no formulas in the show itself, but there are several, apparently randomly selected, during the opening credits.)

Note also that in the early part of the show pressure was measured in pounds per square inch, while in the last version of the experiment pressure was measured in millimeters of mercury. No effort was put into explaining how these relate to each other. Because the purpose of the gauge (and of the "measurement" for that matter) is to look and sound scientific without actually making any type of calculation.

"Math is hard," said Barbie the people who "love" science (as long as they don't have to learn any).

III - Experiments don't need controls or replication

As usual, Mythbusters experiments are made without a control condition and run only once. The lack of a control is less important in this episode, as they were really not testing any theories (unless one considers the quality of vacuum cleaner seals a theory), but the lack of replication is problematic.

Adam does make a lot of attempts to lift the car, eventually getting his rig to work. Once. Since there are all sorts of situation variables that aren't fixed, including the speed at which the crane operator lifts the load, that "experiment" needs replication.

Note that here we're not talking about the independent replication that is now debated in science (when team A publishes a result and team B checks that result by replicating the experiment). Independent replication has been the bane of the social sciences, for example. What we're talking about here is to make the car go up more than once.

IV - Change whatever elements of an experiment you want, no problem

When measuring the "force" (in fact the pressure) of the vacuum cleaner in the shop, and for the first few tests, Adam uses a home vacuum cleaner (looks like a Dyson), but later the experiments with the car use a shop vacuum cleaner, which in my experience creates a lot more pressure ("suction" is pressure). Jamie changes the explosive from a plastic explosive to ANFO (ammonia nitrate - fuel oil, a much slower explosive).

For the small-scale experiments to have any relevance to the large-scale experiments, all the elements other than scale should be unchanged. There could be a case for a different explosive if Jamie were trying to scale up the detonation speed, though that's hard to do correctly, but it would have gone in the other direction, using an explosive with higher detonation speed.

(The explosives are rigged by demolition experts, who could probably have taught Jamie how to do the detonation correctly, since it's their expertise; but that wouldn't work with the psychological premises of the show: that the Mythbusters are experts and experts don't ask for help -- both totally wrong.)

None of these things matter

To the audience, that is. Because their audience is full of people who "love" science, as long as they don't have to learn any. And they want explosions, words that they can use to impress equally ignorant friends (like stoichiometry), and the warm glow of looking down upon other people who don't profess "love" for science (but might actually know some).

And for those who believe that the Mythbusters might have some value as a motivator, consider the case of Planet Fitness: a gym where you pretend to work-out and people tell you how great you are doing, therefore preventing you from actually working out at a real gym.

The Mythbusters are the Planet Fitness of science education.