I had a similar experience at [ a large company ] I used to work for. Had a French Press sitting out on my desk, was quite a bit cheaper than getting coffee in the cafeteria. Came in one morning, it was gone and had been replaced with a nasty note from security. Coffee makers were a banned item (due to fire hazard), so that meant that the French Press (as well as pour overs) were also banned, as they are used to make coffee. Rumor has it that they would also confiscate single-serve coffee pouches too.
"Secrecy is contagious: If something is secret, and something else touches it, it too becomes secret. Secrecy becomes a disease. Everything around the secret issue becomes secret, so the trial became a secret, so I became a secret."
(Best read in context because it's actually a quote within a quote, but that gets cumbersome to replicate in an HN comment.)
I'd also encourage people to read David Langford's book, The Leaky Establishment, which is definitely not in any way based on his experiences working at the Atomic Weapons Establishment in Aldermaston, aka the UK's atom bomb factory, in the 1980s; it says so right in the dedication at the front:
It tells the story of how our hero smuggles a discarded filing cabinet home with him one day, only to discover that someone has popped a plutonium warhead into the bottom drawer while he wasn't looking; he spends the rest of the book trying to smuggle it back in again.
I am not sure if it's secrecy or safety. Tr sorry about the orange aside, it seems the concern is less about whether an outsider would see a sphere and know that it is cricitcal to the design, and more that a scientist might leave an important model or an actual plutonium put out where it shouldn't be, and now only another scientist can tell that it's out of place. Seems reasonable to say "anything that looks like a plutonium pit should stay locked up or an alarm should be raised."
The application to spheres and the possibly apocryphal story of the orange is bordering on the absurd, and is amusing, for sure. But the prospect of having the Soviets pull ahead in the nuclear race kind of put a damper on people's sense of humour in those days.
But that said, these policies are about eliminating the grey zones enabled by imperfect and inconsistent application of human judgement, and are alive and well today in practically every organisation that even vaguely deals with sensitive material: The clean desk policy. Because sensitive material would be written on paper, any paper is considered sensitive by default. It's obviously entirely impractical for security staff to go through all the paper on your desk and validate that they only contain "safe" content, and frankly, it's a fairly high cognitive load to constantly evaluate that for yourself for every sheet of paper. So the policy is that everything is removed at the end of the day, and it's easy for everybody to understand.
The same again for physical security. Everybody wears their badge visibly all the time. Yes, it's obvious that you work here, and the guard knows you and you've said good morning to each other every day for four years, but by making the policy apply to everybody all the time, you remove the grey zones where a bad actor exploits a security guard's lapse in judgement: did you really leave your badge in the pocket of your other coat, or were you just fired in the other building? 99.99% of the time you really did forget it, but it's the other 0.01% we have security for in the first place.
Ah, so much absurd-calling in the comments here; but then, have you seen the certainly totally innocent butterfly drawings made by one bright young man Baden-Powell? My, oh my, now those butterflies have some weird resemblance to maps of our fortifications; but we wouldn't ban butterfly pictures in our compound, just because they resemble our secrets, that would be absurd, wouldn't it?
For me, knowing this, suddenly the article about spheres makes me much more uncertain about who's really right here.
(That said, the comment by derekp7 (https://news.ycombinator.com/item?id=13033511) makes a really good point of highlighting a particular aspect of legalese creep in this context.)
The danger of course is that if all spheres are forbidden, their very lack can leak the fact that they are significant to an adversary. Especially when they are something commonplace like a sphere.
A parallel problem appeared in stealth ships (at least early ones -- somehow the Zumwalt is supposedly stealthy too). Apparently although they didn't show up on radar, neither did the "chop" of the waves where the ship should be. Source: Ben Rich's book on the Skunk Works.
How commonplace are spheres exactly? If I take a look around the office, the closest thing that isn't a ball bearing in some chair or whatever that I would find is an apple.
The article mentioned an orange being deemed classified once it entered the facility, as well as ashtrays (I can guess that a closed ashtray would prevent spills and keep air currents from moving ashes out of the ashtray, though perhaps it was simply an esthetic thing). It wasn't remarkable to see a globe (world maps) in those days either -- not common, but not a surprise when you did see one). Looking around the kitchen I'm in I just see fruit and loose tennis and lacrosse balls.
It's a no win situation for the security staff: someone could make a couple of marks on an orange, or just use it as an aide-memoire.
You have two spherical cows. You forget to put them in your safe and within an hour, one is taken by security and "safely detonated" and the other is accidentally placed into a nuclear weapon, turning the weapon into a moosive failure.
I wonder whether anyone ever suggested that the complete absence of any vaguely-spherical object visible on the campus would itself communicate the importance of the spherical shape. If there's a spy who's able to see what's left out on desks, the fact that people are being threatened for anything visible of a certain shape is probably enough to let the spy deduce what's going on.
I think there's an interesting parallel between this and debugging. For example, change all your test data numbers into pairs except for one. Now all impair sums will include that number.
Though I have to say I'm a bit confused about the sphere example here. A sphere is a sphere is a sphere right? What's the classified bit that they're trying to hid? The exact dimensions?
Banning spheres wasn't intended to hide the fact that spheres were involved. It was intended to make it easier for the, presumably less knowledgable, security enforcement officers to not have to know "is this a secrete sphere or a normal sphere", by simply banning all spheres.
AFAIK, Nuclear weapons basically revolve around creating a critical mass of radioactive material. The issue is that a critical mass is actively trying to push itself apart (because it's obviously exploding). So the idea is that you get your mass, wrap it in a sphere of explosives that will detonate and keep the critical mass contained for slightly longer, which will result in a larger explosion.
The implosion design is more about offering a way to turn a non-critical mass into a critical mass on demand, since you don't want the thing going off the instant you finish manufacturing it.
The "problem" (scare quotes because, really, something being less efficient than it could be at wiping out a city in a single go is hard to frame as a problem) of the bomb blowing itself apart before most of the fissile material has a chance to undergo fission is believed to be solved by other methods.
Yes. The essential mechanism of a fission bomb is the formation of a critical mass within ~ a millisecond's time. Form the critical mass too slowly, and you get massive heat and radiation, but no explosion. Thus the standard cylindrical design with half a critical mass at each end, sitting in front of charges. Detonate both charges at once, slam the two halves together, boom. Crude but effective.
Imagine this problem, you have a bunch of scientists, trained PhD's, suddenly working at a top secret high security job. These are not professional military or security folks, their previous job was most likely in a university doing research and teaching 20 year old students.
How do you get these scientists to understand and take on the appropriate kind of thinking and way of acting that the security risks of this job requires? How can you be sure they get it and are paying attention to detail?
I realize to some this might feel like it's beyond common sense and going too far. And maybe it is. But it kind of reminds me of the Van Halen contracts where they banned brown M&M's. Of course it is not about the brown M&M's. https://en.wikipedia.org/wiki/Van_Halen#Contract_riders
