Ready, Aim, ID Check: In Wrong Hands, Gun Won't Fire
camera_lumina at hotmail.com
Mon Jan 10 12:42:47 PST 2005
And we'll probably have many years of non-Smart-Gun type accidents...eg,
Drunk guy at party put gun to his head and blew his own brains out, assuming
it was a smart gun, or, trailer park momma gives gun to toddler assuming its
a "safe" smart gun.
>From: "Trei, Peter" <ptrei at rsasecurity.com>
>To: "John Kelsey" <kelsey.j at ix.netcom.com>, "R.A. Hettinga"
><rah at shipwright.com>, <cryptography at metzdowd.com>,
><cypherpunks at al-qaeda.net>
>Subject: RE: Ready, Aim, ID Check: In Wrong Hands, Gun Won't Fire
>Date: Mon, 10 Jan 2005 15:04:21 -0500
> > >Ready, Aim, ID Check: In Wrong Hands, Gun Won't Fire
> > > By ANNE EISENBERG
> > I just wonder what the false negative rates are. Seem like a
> > gun that has a 1% chance of refusing to fire when you *really
> > need it* might not be worth all that much. Similarly, one
> > that you can't get to work if you've got a band-aid on your
> > finger, or a cut on your hand, or whatever, loses a lot of
> > its value. On the other hand, a gun that can't be made to go
> > off by your toddler is a pretty huge win, assuming you're
> > willing to trust the technology, but a 90% accuracy level
> > sounds to me like 10% of the time, your three year old can,
> > in fact, cause the thing to go off. That's not worth much,
> > but maybe they'll get it better. And the "suspect struggles
> > with cop, gets gun, and shoots cop" problem would definitely
> > be helped by a guy that wouldn't go off for 90% of attackers.
> > --John
>A remarkable number of police deaths are 'own gun'
>incidents, so the police do have a strong motivation
>to use 'smart guns' if they are reliable.
>In New Jersey, there is some kind of legislation
>in place to restrict sales to 'smart guns', once
>they exist. Other types would be banned. (Actually,
>getting a carry permit in NJ is already almost
>impossible, unless you're politically connected.)
>This particular model seems to rely on pressure
>sensors on the grip. This bothers me - under the
>stress of a gunfight, you're likely to have a
>somewhat different pattern than during the
>Many 'smart guns' also have big problems with
>issues which arise in real life gun fights -
>shooting from awkward positions behind cover,
>one-handed vs two-handed, weak hand (righthander
>using left hand, and vice versa, which can happen
>if dictated by cover or injury), point vs
>sighted shooting, and passing a gun to a disarmed
>There are other systems which have been proposed;
>magnetic or RFID rings, fingerprint sensors, etc.
>The one thing that seems to be common to all of
>the 'smart gun' designs is that they are
>conceived by people with little experience in
>how guns are actually used.
>To look at a particularly ludicrous example, try
>For a gun to work, it is just as important that
>it fires when it should, as that it does not
>fire when it shouldn't. A safety system
>which delays firing by even half a second,
>or which introduces a significant false
>rejection rate (and 1% is way over the line),
>is a positive hazard.
>When the police switch to smart guns, and
>have used them successfully for some time
>(say, a year at least) without problems,
>I'll beleive them ready for prime time.
More information about the cypherpunks-legacy