“Smart” Guns: Potemkin Safety

space invadersThis dumb idea keeps regenerating itself like respawning enemies in a zombie game, or, given the age and technology behind this dumb old idea, like the bad guys in Space Invaders, the ancient arcade video game (if you recognize the screen on the left, “the hill” is something you’re officially “over”).

Fortunately, not everyone is as weary of battling this issue as we are, and comes Herschel Smith with what it would take to convince him, or any of us, that these things work:

[L]et’s talk yet again about smart gun technology.  I am a registered professional engineer, and I spend all day analyzing things and performing calculations.  Let’s not speak in broad generalities and murky platitudes (such as “good enough”).  That doesn’t work with me.  By education, training and experience, I reject such things out of hand.  Perform a fault tree analysis of smart guns.  Use highly respected guidance like the NRC fault tree handbook.

Armatix iP1: bulky, underpowered, and unreliable. And they say it's the wave of the future -- if their coin-op politicians command it so.

Armatix iP1: bulky, underpowered, and unreliable. And they say it’s the wave of the future — if their coin-op politicians command it so.

He’s got a good point there. If you run an Ishikawa diagram of potential faults in a Glock 17, there are not a hell of a lot of branches on your fault tree. There are more on the venerable 1911 (and the 1911’s general reliability illustrates how dogged engineering can sometimes overcome baroque design). Now imagine the fault tree diagram for an Armatix iP1. Don’t forget the various modes of battery failures, radio frequency interference, need to use a weapon weak-hand or by a third party, etc. (The diagrams may suggest why the failed iP1 never seemed to exceed about 90% reliability, failing at a rate of about one round per magazine, and that may suggest why Armatix’s honcho, Ernst Mauch of HK’s you-suck-and-we-hate-you days, tried to get governments to order people to buy the piece of dung. But we digress).

Assess the reliability of one of my semi-automatic handguns as the first state point, and then add smart gun technology to it, and assess it again.  Compare the state points.  Then do that again with a revolver.  Be honest.  Assign a failure probability of greater than zero (0) to the smart technology, because you know that each additional electronic and mechanical component has a failure probability of greater than zero.

Get a PE to seal the work to demonstrate thorough and independent review.  If you can prove that so-called “smart guns” are as reliable as my guns, I’ll pour ketchup on my hard hat, eat it, and post video for everyone to see.  If you lose, you buy me the gun of my choice.  No one will take the challenge because you will lose that challenge.  I’ll win.

Yep. What he is asking the Smart Gun proponents to do is resolve an asymptote to zero, which is mathematically impossible, and probably, in this non-mathematical but real-world-physical case, functionally impossible. If you want to know why adding “Safety Technology” to firearms has never banished mishaps, a good book is Charles Perrow’s Normal Accidents.

Now, Perrow wrote the book as an anti-nuclear jeremiad, which may turn off some readers, especially those aware that a nuclear-power reactor control room is historically a safer environment than a Senator’s Oldsmobile, but he notes a very interesting thing: when you get the low-hanging fruit all plucked, that is, say, when the Air Force addressed items in the 1950s flying culture that had them pranging 1000 planes a year, you get a safety system that’s so optimized that adding anything more to it produces new, unintended and unanticipated points of failure.

We see this in aviation safety. American Airlines was concerned about loss-of-control accidents and so encouraged its pilots to seek “upset training” in aerobatic competition airplanes. One such pilot then tried the control inputs that worked in an Extra 300 (stressed to ± 12G in all directions, IIRC), in an Airbus whose tailfin was stressed to ± 1.5G. The result was a disaster, one caused by trying to increase the airline’s already very-high levels of safety!

Likewise, attempting to add safety features to firearms has led to fatalities and injuries. A classic example is the Glock “New York Trigger,” unquestionably a factor in several recent incidents of dreadful cop marksmanship, including incidents where bystanders were shot in addition to and even instead of armed criminals. The NY and NY2 triggers can be shot accurately by experts, but they greatly increase the dispersion of shots fired by average cops, and mandating them is tantamount to ordering your cops to shoot a few random citizens over the next decade or so.

But it looks like safety, to a superficial view (journalism, anyone?), and therefore it’s likely to spread. The “Smart Gun” is another example of this Potemkin safety. If it is discussed in your legislator (or, God forbid, your local police consider something like it), real experts need to come forward to counter the antis’ and interested manufacturers’ paid pushers.

One thought on ““Smart” Guns: Potemkin Safety

  1. Arturo

    ars technica just had an article on the recent smart gun conference. It pains me that tech publications are so far in bed with gun control, it is very difficult to find a tech site that isnt. Their coverage of the faults with smart guns could be described as poor. The police commander that was part of the conference even said he doesn’t trust the things and wouldn’t for a long time of proven results. Did he get majority of the article? Heck no, the politician that extolled their virtues did. The only benefit I could see for added electronic triggers is for trigger work on difficult shots and maybe weight. I just I should be glad that even printed his words….

Comments are closed.