42 ms·

>I have worked in a plant where I had half a critical mass in this hand, barehanded and dressed in street clothes, and half in this hand, wearing a lab coat, an

>I have worked in a plant where I had half a critical mass in this hand, barehanded and dressed in street clothes, and half in this hand, wearing a lab coat, and I’d put this half in a pocket on this side and this half in a pocket on this side and walk down the hall. If those two ever got together, there’d be a blue flash. They never got together because I was between them. There's "our design docs are photos of whiteboards" normalization of deviance and there's whatever the hell this is. Absolutely bonkers that there wasn't a demon core-like incident.

littlestymaar 4y ago
> They never got together because I was between them. I don't know how confident I would be about this, given that our body is mostly made out of a very good neutron moderator (water)…
- MagicMoonlight 4y ago
  Alpha radiation can’t penetrate the skin
  - littlestymaar 4y ago
    This has nothing to do with alpha rays (which are helium nuclei), but with neutrons, and fast neutrons are much less prone to generate a fission than thermal neutrons, so having a moderator in between your fissile material parts reduces the amount of fissile material you need to achieve criticality. That's how nuclear reactor work, by the way. And this is also what happened to John Bistline on the 4th of June 1945 at Los Alamos. Edit: in fact the first ever recorded criticality accident ever recorded involved the water in the body of a scientist (Otto Frisch), though the water was acting as a neutron reflector here, and not as a moderator.
avidiax 4y ago
People walk around with a loaded gun in their pants. They just (usually) carefully manage not to pull the trigger. Would it be better to walk down the hall twice? Slightly, sure. Turns out it wouldn't have made a difference that we know of.
- tomxor 4y ago
  What if two people walk down the hall doing the same thing in different directions? ... Oh I forgot my slide rule ... !#FLASH#! ... what was that!? Oh shit. Half joking. Is the material and spatial separation of a lab coat enough to prevent a reaction between two lumps of plutonium of sufficient combined critical mass? It might be since it's so bad at penetrating anything solid, but I don't know.
  - eximius 4y ago
    Considering the original demon cores were held apart by a screw driver, the answer is technically yes. It's still a horrifyingly unsafe thing to do. https://en.wikipedia.org/wiki/Demon_core#/media/File:Tickling_the_Dragons_Tail.jpg https://en.wikipedia.org/wiki/Demon_core#/media/File:Ticklin...
    - kennethrc 4y ago
      The Hollywood version of events: https://www.youtube.com/watch?v=AQ0P7R9CfCY https://www.youtube.com/watch?v=AQ0P7R9CfCY
    - earlyam 4y ago
      Further reading: https://www.science.org/content/article/near-disaster-federal-nuclear-weapons-laboratory-takes-hidden-toll-america-s-arsenal https://www.science.org/content/article/near-disaster-federa...
    - closewith 4y ago
      Technically, it was the neutron reflecting sphere was was held apart by a screwdriver.
    - bombcar 4y ago
      Remember the demon core thing happened more than once, too.
  - raverbashing 4y ago
    Good thinking. One of the things modern "computing practices" taught us is the myriad of ways things can go wrong, they were probably a bit more naive at that time It's Murphy's law squared
akira2501 4y ago
The article writes about it like it was an element of 1950s iconoclasm. Unfortunately, it never really went away. https://www.science.org/content/article/near-disaster-federal-nuclear-weapons-laboratory-takes-hidden-toll-america-s-arsenal https://www.science.org/content/article/near-disaster-federa...
- anigbrowl 4y ago
  Good grief. That's the most infurisating thing I've read in a long time, and I already knew a lot of tha background/historical incidents.
- pyuser583 4y ago
  Cherenkov radiation is pretty common in nuclear engineering, and occurs naturally in a number of contexts. It’s not some forebear of doom. Otherwise, a great article!
  - wincy 4y ago
    I got to see Cherenkov radiation once on a high school class trip with only water between me and the blue glow! Being a teenager I asked “what would happen if I jumped into the reactor?” The researcher (this was a medical and research reactor) laughs and says “well, if you just jump into the water and splashed around we’d have hell to pay with filling out an incident report to <something important sounding, I forget>, but physically you’d be fine since the water shields you from the radiation. If you did a swan dive down toward the glow your skin would all melt off before you got to the reactor” maybe he was just saying this for effect but it’s always stuck in my head that water is a really, really good insulator against radiation.
    - krisoft 4y ago
      > water is a really, really good insulator against radiation. Absolutely. There is this[1] horror story about a diver who was doing maintenance in a transfer pool next to two spent fuel pools. After he was done with the work he found an unidentified object at the bottom of the pool and was instructed to put it in his tool basket. The object was a broken piece of irradiated reactor component. They realised that something is very wrong when the diver was lifting the basket out of the water and the area radiation monitors started to alarm, so they lowered the basket back into the water. On the 10th slide of the document there is a diagram showing the water's shielding as a function of distance. At 1.4 meters there is no dose they can measure while they measure extremely high doses at centimeter distances. 1: https://isoe-network.net/publications/pub-proceedings/symposia/north-american-tc-symposia/fort-lauderdale-usa-january-2011/slides-14/session-iv-1/1756-ritter2011-ppt/file.html https://isoe-network.net/publications/pub-proceedings/sympos...
KRAKRISMOTT 4y ago
What's wrong with design docs being on a whiteboard? Not every company has time to painstakingly draw pretty UML diagrams in Figma when there are features that need to be built!
- brazzy 4y ago
  A whiteboard is a great tool to quickly iterate towards a shared understanding of the design at that point in time. But if you take a photo of it and think you're done with documenting your design, you A) lose most of the context of the decisions that were made to arrive at that design. And B) have ensured that your documentation will get outdated and become useless pretty soon. You've basically decided not to have design documentation but filled yourself into believing that you do. If you are serious about documenting software design to help future developers understand the system, you really do need those diagrams in a digital, easily updateable format and with textual explanations of the underlying decisions. And if doing that seems more work than it's worth, do it anyway, just with less details.
- alexjplant 4y ago
  You joke, but I once worked on a team where nobody was able to point me to any coherent high-level documentation of what the system actually _did_... they all just walked around referring to it by some proprietary name as though you were some type of misinformed dunderhead for not understanding them. It took two hours of a manager talking and filling up a wall-sized whiteboard for me to discover that it was just a hideously over-engineered amalgamation of a CDN and document repository. If they'd had a diagram, README, or had even been able to give me the one-sentence overview then I suspect that the project's head count and productivity would have looked much different. This is all to say that if you 1) invent novel terminology for things that already exist, 2) don't document anything, and 3) run your team like a high school clique then you never have to worry about finding yourself out of a job. Thankfully my current gig has loads of video training and an interactive dictionary for proprietary terms in an attempt to combat this.
  - AtlasBarfed 4y ago
    And all the silicon valley layoffs, likely filled to the brim with people that conscienciously documented their replaceability for their employer, will look at that "high school clique" and remark "that's how you deal with sociopathic management". Yeah, eventually management will nuke the whole division, but at least it provides those with jobs a bit more runway. How much of the bemoaning of such adversarial tactics by engineers themselves is simply annoyance at what it causes them at their jobs, or the fact they have internalized pro-management propaganda?
didntreadarticl 4y ago
I have worked in a plant where I had half a critical mass in this hand, barehanded and dressed in street clothes, and half in this hand, wearing a lab coat So they were only wearing a lab coat on one half of their body? I'm confused.
- didntreadarticl 4y ago
  Or maybe they had three hands, they're only mentioning two of them, and the lab coat is intact
- CommitSyn 4y ago
  One lab coat, one pair of jeans. Half core in lab coat pocket, half core in jeans pocket, to prevent the lab coat pockets from coming into contact if they were in opposite pockets of the lab coat.
  - quinndexter 4y ago
    Safety first!

littlestymaar 4y ago

> They never got together because I was between them. I don't know how confident I would be about this, given that our body is mostly made out of a very good neutron moderator (water)…

MagicMoonlight 4y ago
Alpha radiation can’t penetrate the skin
- littlestymaar 4y ago
  This has nothing to do with alpha rays (which are helium nuclei), but with neutrons, and fast neutrons are much less prone to generate a fission than thermal neutrons, so having a moderator in between your fissile material parts reduces the amount of fissile material you need to achieve criticality. That's how nuclear reactor work, by the way. And this is also what happened to John Bistline on the 4th of June 1945 at Los Alamos. Edit: in fact the first ever recorded criticality accident ever recorded involved the water in the body of a scientist (Otto Frisch), though the water was acting as a neutron reflector here, and not as a moderator.

avidiax 4y ago

People walk around with a loaded gun in their pants. They just (usually) carefully manage not to pull the trigger. Would it be better to walk down the hall twice? Slightly, sure. Turns out it wouldn't have made a difference that we know of.

tomxor 4y ago
What if two people walk down the hall doing the same thing in different directions? ... Oh I forgot my slide rule ... !#FLASH#! ... what was that!? Oh shit. Half joking. Is the material and spatial separation of a lab coat enough to prevent a reaction between two lumps of plutonium of sufficient combined critical mass? It might be since it's so bad at penetrating anything solid, but I don't know.
- eximius 4y ago
  Considering the original demon cores were held apart by a screw driver, the answer is technically yes. It's still a horrifyingly unsafe thing to do. https://en.wikipedia.org/wiki/Demon_core#/media/File:Tickling_the_Dragons_Tail.jpg https://en.wikipedia.org/wiki/Demon_core#/media/File:Ticklin...
  - kennethrc 4y ago
    The Hollywood version of events: https://www.youtube.com/watch?v=AQ0P7R9CfCY https://www.youtube.com/watch?v=AQ0P7R9CfCY
  - earlyam 4y ago
    Further reading: https://www.science.org/content/article/near-disaster-federal-nuclear-weapons-laboratory-takes-hidden-toll-america-s-arsenal https://www.science.org/content/article/near-disaster-federa...
  - closewith 4y ago
    Technically, it was the neutron reflecting sphere was was held apart by a screwdriver.
  - bombcar 4y ago
    Remember the demon core thing happened more than once, too.
- raverbashing 4y ago
  Good thinking. One of the things modern "computing practices" taught us is the myriad of ways things can go wrong, they were probably a bit more naive at that time It's Murphy's law squared

akira2501 4y ago

The article writes about it like it was an element of 1950s iconoclasm. Unfortunately, it never really went away. https://www.science.org/content/article/near-disaster-federal-nuclear-weapons-laboratory-takes-hidden-toll-america-s-arsenal https://www.science.org/content/article/near-disaster-federa...

anigbrowl 4y ago
Good grief. That's the most infurisating thing I've read in a long time, and I already knew a lot of tha background/historical incidents.
pyuser583 4y ago
Cherenkov radiation is pretty common in nuclear engineering, and occurs naturally in a number of contexts. It’s not some forebear of doom. Otherwise, a great article!
- wincy 4y ago
  I got to see Cherenkov radiation once on a high school class trip with only water between me and the blue glow! Being a teenager I asked “what would happen if I jumped into the reactor?” The researcher (this was a medical and research reactor) laughs and says “well, if you just jump into the water and splashed around we’d have hell to pay with filling out an incident report to <something important sounding, I forget>, but physically you’d be fine since the water shields you from the radiation. If you did a swan dive down toward the glow your skin would all melt off before you got to the reactor” maybe he was just saying this for effect but it’s always stuck in my head that water is a really, really good insulator against radiation.
  - krisoft 4y ago
    > water is a really, really good insulator against radiation. Absolutely. There is this[1] horror story about a diver who was doing maintenance in a transfer pool next to two spent fuel pools. After he was done with the work he found an unidentified object at the bottom of the pool and was instructed to put it in his tool basket. The object was a broken piece of irradiated reactor component. They realised that something is very wrong when the diver was lifting the basket out of the water and the area radiation monitors started to alarm, so they lowered the basket back into the water. On the 10th slide of the document there is a diagram showing the water's shielding as a function of distance. At 1.4 meters there is no dose they can measure while they measure extremely high doses at centimeter distances. 1: https://isoe-network.net/publications/pub-proceedings/symposia/north-american-tc-symposia/fort-lauderdale-usa-january-2011/slides-14/session-iv-1/1756-ritter2011-ppt/file.html https://isoe-network.net/publications/pub-proceedings/sympos...

KRAKRISMOTT 4y ago

What's wrong with design docs being on a whiteboard? Not every company has time to painstakingly draw pretty UML diagrams in Figma when there are features that need to be built!

brazzy 4y ago
A whiteboard is a great tool to quickly iterate towards a shared understanding of the design at that point in time. But if you take a photo of it and think you're done with documenting your design, you A) lose most of the context of the decisions that were made to arrive at that design. And B) have ensured that your documentation will get outdated and become useless pretty soon. You've basically decided not to have design documentation but filled yourself into believing that you do. If you are serious about documenting software design to help future developers understand the system, you really do need those diagrams in a digital, easily updateable format and with textual explanations of the underlying decisions. And if doing that seems more work than it's worth, do it anyway, just with less details.
alexjplant 4y ago
You joke, but I once worked on a team where nobody was able to point me to any coherent high-level documentation of what the system actually _did_... they all just walked around referring to it by some proprietary name as though you were some type of misinformed dunderhead for not understanding them. It took two hours of a manager talking and filling up a wall-sized whiteboard for me to discover that it was just a hideously over-engineered amalgamation of a CDN and document repository. If they'd had a diagram, README, or had even been able to give me the one-sentence overview then I suspect that the project's head count and productivity would have looked much different. This is all to say that if you 1) invent novel terminology for things that already exist, 2) don't document anything, and 3) run your team like a high school clique then you never have to worry about finding yourself out of a job. Thankfully my current gig has loads of video training and an interactive dictionary for proprietary terms in an attempt to combat this.
- AtlasBarfed 4y ago
  And all the silicon valley layoffs, likely filled to the brim with people that conscienciously documented their replaceability for their employer, will look at that "high school clique" and remark "that's how you deal with sociopathic management". Yeah, eventually management will nuke the whole division, but at least it provides those with jobs a bit more runway. How much of the bemoaning of such adversarial tactics by engineers themselves is simply annoyance at what it causes them at their jobs, or the fact they have internalized pro-management propaganda?

didntreadarticl 4y ago

I have worked in a plant where I had half a critical mass in this hand, barehanded and dressed in street clothes, and half in this hand, wearing a lab coat So they were only wearing a lab coat on one half of their body? I'm confused.

didntreadarticl 4y ago
Or maybe they had three hands, they're only mentioning two of them, and the lab coat is intact
CommitSyn 4y ago
One lab coat, one pair of jeans. Half core in lab coat pocket, half core in jeans pocket, to prevent the lab coat pockets from coming into contact if they were in opposite pockets of the lab coat.
- quinndexter 4y ago
  Safety first!