Hacker News Reader

Comment by Mithaldu | Hacker News Reader

The entire thing is worth a read, but here's a short summary:

John Bordne claims that missile launch crews in okinawa received nuclear missile launch orders via calm encoded voice comms, which seemed suspect, since they were given at defcon 2, not defcon 1. According to the claim they ended up contacting the sender of the codes and requested codes to either stand down or upgrade defcon, at which point the sender issued stand down with considerable stress in his voice. Bordne also claims that to his knowledge no real consequences for this ever happened.

He's asking for corroboration to help with his own old memories and to straighten out history.

View on HN

I'd also include that they received the coded message twice showing that it's unlikely it was misheard or misread on either end. That was my initial reaction/theory, that the code was close to the confirmed order but one side messed up some how. With a second confirmation it seems like the 'NOP' (dummy) code either accidentally matched the LAUNCH code or the correct code was given to (or procured by) the major who transmitted the mid-shift messages.

It's an interesting question of how do you ensure you never accidentally transmit the correct code without the sender knowing the correct code and being able to command a launch single-handedly. Even if the code is really long, generating random NOP codes doesn't give the level of safety I'd like when accidental collisions can end the world.

Maybe the book was sorted alphabetically, such that the code for LAUNCH MISSILES was listed next to the code for LUNCH TIME or something.

My understanding is that the correct launch authorization code would be locked in a safe until the command post had received a message authorizing them to launch. It's safes and authorization confirmations all the way down as I understand it.

I am sure the system was designed to resist intentional manipulation by intermediates, but what about plain old telephone-game mishaps? If LAUNCH and LUNCH TIME are represented by one-time-pad blobs that are right next to each other on a list, it doesn't matter how many two-person-locks or feet of steel prevent the operator from learning which blob is the LAUNCH blob because an off-by-one mistake is still easy to make as ever.

I was saying that the launch codes would only be available if the CP had received their own launch codes and there wouldn't be a list with the launch codes and lunch codes to make that error.

The question is whether the codes were sufficiently far apart that you couldn't stumble onto one by accident, or whether they were similar enough that you could generate LAUNCH MISSILES by accident with a typo, even if you never went near the safe and had no idea what your message actually meant.

>It's an interesting question of how do you ensure you never accidentally transmit the correct code without the sender knowing the correct code

Use a good hash function and give only the hash to the sender? Before sending, the sender hashes the codes she has and compares the hash with the "hot" hash.

My own thoughts separately: If this is all correctly remembered and truthful, it could mean that someone managed to sneak those codes into the command chain and the person passing them on wasn't even aware they were the launch codes. In that case i would suspect any consequences, if they happened, to be kept quiet on as well.

Or it's super mundane and just a coincidence/typo.

I'm curious what the actual protocol was for defining the codes. Could it have been an accidental collision? depending on the codes, what is the probability all three matched?

You have to wonder if there was some sort of covert op in play, and cooler minds prevailed (Capt bassets' rigorous adherence to protocol maybe was not planned for by the attacker. Refusal to launch at defcon2, which could have been overlooked by a less attentive sr official).

What sort of covert op? Who would want to start WWIII?

In the decade or so before the Cuban Missile Crisis there were some US military leaders who thought that a "preventative" attack on the Soviets was a good idea - by the time of the Cuban Missile Crisis the public estimates of Soviet capabilities were so high (overestimated by a factor of 100 in some cases) that people probably thought it was too late for such an attack.

http://www.japantimes.co.jp/life/2005/03/13/to-be-sorted/the...

it's probably crack-pot conspiracy theory, but you could definitely sense there were elements of our government who thought a pre-emptive strike to level russia/china in one swift move would be agreeable. Perhaps they thought they would not be capable of retaliating, or the retaliatory strikes would not be nearly as impactful?

I'm not suggesting that is a rational approach, but I've heard crazier things uttered in terms of foreign policy even in modern day (e.g.: nuke entire regions, turn them into glass).

It terrifies me to no end that people could ascend into power with such ideas, but we'd be fools to think it doesn't happen. Look at any big org (company/government, whatever).. the best and brightest are not always the ones being promoted up, -- it's often the sociopaths.

I believe the thinking among many at the time was that war with the Soviets would result in "acceptable" losses to the US i.e. 20 million Americans dead, give or take. This was probably not too far off given extremely limited Soviet capabilities at the time. (Europe would have fared far worse, but I don't know that they cared too much.) There's a scene in Dr. Strangelove where this is advocated, and I think it's reasonably true-to-life.

> Bordne also claims that to his knowledge no real consequences for this ever happened.

Well, the sender in question was allegedly court-martialed.

Did I miss that in the article? Or do you have another source?

You missed it.

Ah, found it. Thanks.

real consequences. If the story as told is correct the sender was initially not aware they'd sent missile launch codes.

Reply on news.ycombinator.com