The Great Boston Molasses Flood of 1919 (2018)
newenglandhistoricalsociety.com
newenglandhistoricalsociety.com
I find interesting how the fact that molasses is a non-Newtonian fluid helps explaining its behaviour: https://en.wikipedia.org/wiki/Non-Newtonian_fluid
Continuing the analogy, Some mechanical engineers were better at learning from their exploded boilers than others were.
A few (many?) of us in the industry have been quietly watching and learning from the explosions occurring across the street. About once a week you find some wreckage strewn across the way and you find familiar stories in the tangled mess - "Too much webscale", "didnt care about the business", "meme language crippled productivity", etc. Usually doesn't take a forensic mastermind to determine why a software product exploded - at least in strategic terms.
> Does this mean we need more stringent regulatory and certification regimes for IT and computer security?
No. We do not need any more arbitrary regulatory & certification processes in our lives. If you have a specific business application that requires additional scrutiny (i.e. nuclear reactor scram control system), then the appropriate domain-specific regulations & certifications should be applied. It makes absolutely no sense to impose these constraints upon the field at large.
As soon as you deal with customer PII, it does make sense to mandate a) insurance and b) standards.
Just how much PII got compromised because of failure to adhere to basic IT security standards? Target got hacked because they put IoT devices on the same network as the cash registers and CC readers. Web shops are hacked every day because of software unpatched for years. Hospitals and other critical infrastructure get hacked on a weekly basis because of even more unbelievable security issues (like, once again, running unpatched software and lacking network segmentation).
And don't get me started on the utterly disgusting shit you see in smartphone BSPs. The stuff that vendors do there is just mind boggling - if I were a secret service looking for an exploit, I'd start in the horribly patched-together kernels. Or in IoT devices that are outdated the very moment they leave the factory floor.
Yes, we definitely need a lot more mandatory quality control and standards.
> If you have a specific business application that requires additional scrutiny (i.e. nuclear reactor scram control system), then the appropriate domain-specific regulations & certifications should be applied.
Feel free to replace "nuclear reactor scram control system" with "PII" or any other less urgent thing that makes the particular business nervous. There are many problem domains where you literally cannot fuck this stuff up even if you tried.
The stakes for indie game developers are substantially different than those for a F500 insurance company.
We definitely need to get the fuck out of other people's business. If you dont want someone to have your PII, dont share it with them. Regulating everyone on the same axis is pure tyranny when the problem space is so large.
No they are not, if your indie game has network connectivity that has security bugs it can and will be exploited.
Engineering is just hard, full stop.
What does follow from that is "listening to engineers is hard", because it costs money, time and other resources.
Why am I not surprised they tried to blame somebody else but themselves
https://news.ycombinator.com/item?id=13043559
https://news.ycombinator.com/item?id=10361913
There were a number of other submissions over the last 12 years but no others with many comments.
Solving a Mystery Behind the Deadly ‘Tsunami of Molasses’ of 1919 - https://news.ycombinator.com/item?id=13043559 - Nov 2016 (20 comments)
Great Molasses Flood - https://news.ycombinator.com/item?id=10361913 - Oct 2015 (5 comments)
Looks like these are the other ones with comments:
The Great Molasses Flood (1919) - https://news.ycombinator.com/item?id=25911280 - Jan 2021 (1 comment)
In 1919, a 40-foot molasses wave slammed into city’s North End, killing 21 - https://news.ycombinator.com/item?id=18913799 - Jan 2019 (1 comment)
How the Boston Molasses Disaster Ushered in the Era of Modern Regulation - https://news.ycombinator.com/item?id=18863555 - Jan 2019 (2 comments)
[1] https://en.wikipedia.org/wiki/Uncle_and_the_Treacle_Trouble
Does it take some scientists to figure this out? If so, are there any papers to explain their findings?
Why would a munitions plant need an enormous quantity of molasses? Was it used in the production of munitions?
> unquenchable thirst for the industrial alcohol needed to produce cordite—a smokeless gunpowder used in ammunition and artillery shells
"...a small portion of molasses would be distilled into grain alcohol for rum, but most of it, more than 80 percent, would be distilled into industrial alcohol that would be used as a major ingredient in the production of munitions, especially dynamite, smokeless powder, and other high explosives."
From https://archive.org/details/darktide00step/page/10/mode/2up?...
Of the two specific examples listed, dynamite is unambiguously a high explosive. From what I can tell, smokeless powder is usually classified as low by default, because that's how it behaves during storage, shipping, and handling. But during actual use, it occupies a grey area between the two classifications. I've found several sources saying that it can act as a high explosive under the right circumstances.