You may only read this site if you've purchased Our Kampf from Amazon or Powell's or me
• • •
"Mike and Jon, Jon and Mike—I've known them both for years, and, clearly, one of them is very funny. As for the other: truly one of the great hangers-on of our time."—Steve Bodow, head writer, The Daily Show

"Who can really judge what's funny? If humor is a subjective medium, then can there be something that is really and truly hilarious? Me. This book."—Daniel Handler, author, Adverbs, and personal representative of Lemony Snicket

"The good news: I thought Our Kampf was consistently hilarious. The bad news: I’m the guy who wrote Monkeybone."—Sam Hamm, screenwriter, Batman, Batman Returns, and Homecoming

March 30, 2010

Rays From Space

By: Aaron Datesman

I didn’t pay much attention to the recent Toyota recalls, but am sort of amused to read that the culprit might be rays from space:

Federal regulators were prompted to look into the possible role that cosmic rays played in Toyota's product recall fiasco after an anonymous tipster suggested the design of Toyota's microprocessors, software and memory chips could make them more vulnerable to interference from radiation compared with other automakers. This is because Toyota has led the auto industry in its widespread inclusion of electronic controls in the manufacture of their various car models.

As electronic devices are made to perform more and more functions on smaller circuit chips, the systems become more sensitive and vulnerable to corruption, and thus more prone to interference from radiation, said Ewart Blackmore, a senior researcher at TRIUMF, a cyclotron facility in Vancouver, Canada, that works with companies to test and analyze the effects of radiation on products.

I like this because it sounds as if the chairman of Toyota was at a loss to explain why the cars built by his company were killing people, so he asked his twelve-year-old son what he thought might be the matter. As excuses go, “It’s RAYS FROM SPACE” sounds much more plausible than “The cars are possessed by demons”. However, I have some acquaintance with this field, and I find the explanation to be plausible. This is especially true since the current solar minimum (which is also currently serving to mask the effects of global warming somewhat) has increased the cosmic ray flux significantly over the last several years.

I’m interested in this because it points out how dangerous it would be to thoroughly computerize the control systems for (say) a nuclear reactor, an airplane, or a missile warning system. Or, I should say, how dangerous it is that we often choose to employ computers in those situations, because of a thoroughly false belief that computers are more “reliable” than people. In airplanes they call this computerization “fly by wire”. Missile warning failures are another, even more terrifying, example.

— Aaron Datesman

Posted at March 30, 2010 11:29 PM

In another of those military-industrial complex connections, the US always lauded itself on its high-tech fighter aircraft with their complex, computer-based avionics, whilst ridiculing the Soviet fighters, whose avionics often still used vacuum tubes. But Soviet fighters would very likely have been less susceptible to electromagnetic pulse (EMP) effects from nuclear weapons detonations than would US fighters, since their electronic circuitry was more 'clunky', and the induced currents would have been less harmful. Fortunately, we never got to find out whether those MiGs really would have beaten the F-16s under those conditions.

Anymore, though, explode a nuke in the high atmosphere over any major western country and watch every computer-controlled system within range fry from the EMP. Casualties from the blast and radiation would be nil, but the infrastructure, which is completely reliant on extremely small, unprotected integrated circuits, would be utterly crippled.

Shielding? What's that?

Posted by: NomadUK at March 31, 2010 05:42 AM

I don't buy the cosmic ray hypothesis for the fact that I doubt Toyota's arrangement of ICs is so unique that it ends up having some disproportionate effect vs other electronic appliances. Also, not all of these cosmic ray events would necessarily only affect accelerator function. You would expect the cars to cease to function in a variety of ways if CR was striking chips at random. This pattern does not seem to be evident.

I'm more inclined to view this as a combined software/hardware problem. I can imagine some software bug that made it thru development testing without appearing, but when combined with an intermittent sensor failure/misreporting the error appears.

Posted by: Willy at March 31, 2010 08:05 AM

FWIW, in the absence of any compelling evidence explaining why random cosmic rays would affect only the accellerator function, I completely agree with Willy. I'm sure comp.risks will cover it thoroughly.

Posted by: NomadUK at March 31, 2010 08:58 AM

Last paragraph: so true. A bit of education in secondary school would not go amiss, about what a Turing machine is and why they are intrinsically more chaotic than ordinary machines.

Posted by: Cloud at March 31, 2010 09:47 AM

There is no getting away from the fact that technological progress generally carries hidden risks, even apart from the known risks that greed and ambition lead people to ignore. Learned people have even said so:

"You do not know what the full implications of discoveries are, until you have made them. Epidemic poliomyelitis is an example of this. It was, in effect, created by hygienic measures designed to deal with other diseases. The measures designed to protect the world from polio may, in their turn, for all we know, lead to some other quite unexpected consequences which may be to man's disadvantage."

--John Rowan Wilson, Margin of Safety (published 1963), as quoted in The River by Edwin Hooper (which is a remarkable book in which the quote is extremely pertinent).

Posted by: N E at March 31, 2010 01:33 PM

Where I worked the joke was that "mechanical" failure was rare and the first to get done in the development process; "electrical" failures were common but usually blamed on something mechanical (how carefully it may have been "soldered" or "wired" or whatever) and would always come in second to mechanical in the development process; and then when "electronics" appeared and "sofware" engineers and programs to monitor and control "everything" (the whole device) took over - hardly anything ever worked reliably again, for long. (Programming changes and updates never end.)

I did think at one point maybe "neutrios" could pass through the device and trigger a digital readout to switch from a "1" to a "0" or something... Any chance of that?

Posted by: Grandpa Ken at April 1, 2010 11:45 AM

In one of his later books, Jeremy Rifkin talks about how the Stock Market crash of 1987 was in part due to the programming instructions that the computerized end of the Stock Market was driven by.

So although things may not have been so bad in 1987 to actually warrant such a crash, the fact is that once a huge group of people instructed their brokers to start selling off, then the computers jumped in and forced the sell off of large groups of stocks. And it took a while before they could program a "Stop" to the madness.

What is even scarier is that right now, the systems controlling this nation's nuclear defense are set up in the exact same way. So if a flock of geese should trigger a Def Com alert, I am not sure that a human being could intervene the way that occurred back in the carter and Reagan years.

Posted by: Elise Mattu at April 1, 2010 05:44 PM

I did think at one point maybe "neutrios" could pass through the device and trigger a digital readout to switch from a "1" to a "0" or something... Any chance of that?

It is, I suppose, theoretically possible, but the thing about neutrinos is that they don't generally interact with regular matter; they can travel through a lightyear of solid lead and not strike anything. So the chances of one triggering an event in a silicon chip are pretty low.

I think the description of the decline in reliability as we 'progressed' from mechanical to electrical to software/firmware devices is spot on and would be hilarious if we weren't all living on top of the massive house of cards we managed to construct out of software over the past 40 years.

Posted by: NomadUK at April 2, 2010 02:43 AM

Re: the last paragraph: Luddite! The computer chip has & will save & make better all our lives. Just ask Google. The future is The Cloud. Long live The Cloud!

Posted by: David H. at April 2, 2010 04:01 AM

Was watching a Discovery channel doc on the mysteries of lightning the other day, its recently posited that what sparks a bolt may in fact be cosmic rays. Apparently there doesn't seem to be nearly enough juice in thunder clouds to account for the voltages, and high speed cameras pick up x-rays and gamma rays strangely emanating from lighting bolts.

Its pretty humbling to think our little 'mote of dust' is violently reacting to explosions on the other end of the universe. Let alone a Prius.

(although I tend to agree with Willy, it seems doubtful their proprietary hardware could be effected in such a specific manner.)

Posted by: BenP at April 2, 2010 07:50 PM
A bit of education in secondary school would not go amiss, about what a Turing machine is and why they are intrinsically more chaotic than ordinary machines.

"Elementary chaos theory tells us that all robots will eventually turn against their masters and run amok in an orgy of blood and the kicking and the biting with the metal teeth and the hurting and shoving."

So yes, until our Prius arises and slaughters its puny fleshly masters, color me skeptical about the cosmic rays thing.

Posted by: mds at April 5, 2010 02:44 PM