The Titan Submersible Disaster Was Years In the Making, New … – Slashdot

Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!
Nickname:
Password:
Nickname:
Password:
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
“It was avoidable. It was inevitable.” is a contradiction in terms. Something that is inevitable by definition cannot be avoided. Perhaps that expedition leader meant “It was certain to happen [given the company leadership].”?
But hey, EditorDavid managed to turn the article’s female author [susancasey.com] into a “they”, so that counts for something?
This company sounds almost like the Theranos of deep-sea exploration … except their product worked for a little while. The only remotely positive result was that the CEO he

The only remotely positive result was that the CEO here removed himself in the implosion.
That’s not a positive outcome. Justice in general and the the family of the victims in particular would have benefitted from having him sitting in tribunal, be questioned, be shown evidence, experience shame, and know he will suffer in jail for the rest of his time.
Actually, revolutionary speaking, it _is_ a positive outcome. Also, jail time for non-intentionally bad engineering in an experimental field is rather rare.
This is more than “non-intentionally bad engineering” because he neglected the results of the test on the model sub that collapsed. I assumed he would be charged for negligent homicide. Assuming State of Washington where OceanGate is headquartered, penalty is ten years in jail https://www.lawinfo.com/resour… [lawinfo.com] , maybe adding for each of the 4 or 5 passengers that died.
was the test mandated by law? were there specs the hull had to meet?
The answers seem to be ‘no’.
Can you provide reference that negligent homicide requires the test to have been mandated by law? What I read is that it is about “disregard of a risk of harm led to the victim’s death” and the typical defence is “defendant may not have been aware of the unjustifiable risk” https://www.schmidtandclark.co… [schmidtandclark.com] and here he was made aware by the result of the previous test and by several internal reports in particular from his “director of marine operations”.
Indeed. The field is far too small to have any specific legal requirements. Legal requirements only emerge when it is a product many people use or the potential damage is very large.
That’s generally not the legal requirement for negligent homicide. Ignoring laws or regulations can make it easier to upgrade from say involuntary manslaughter to voluntary manslaughter, but isn’t necessarily required for either. It varies by state, but typically it’s a wanton disregard or neglect for safety that results in death.
And of course, nobody can say for certain whether they’d land a conviction, rather a jury would have to decide that.
I doubt that. There are no legal requirements for private submarine engineering and no best practice standards either. That means everybody is free to do whatever they like and there is not even any standard of skill or education you have to fulfill to do it.
Yes, I get the “Somebody has to pay!” mind-set, but it is stupid in such cases as the one at hand.

There are no legal requirements for private submarine engineering and no best practice standards either.
The first part of that is true. The second part I very much doubt. If you surveyed people who work for the companies that build them, you could probably compile a list of best practices in short order. They know because they have already built several iterations. If they had moved a little slower they could have hired some people with experience and gained the benefits of that kind of knowledge. Instead, they broke things with people inside of them.
There may not be legal requirements, but there are certainly class certifications and standards for these kinds of things, as there is for pretty much anything in the marine industry.
Youtube has a good documentary on building Gabe Newell’s “DSV Limiting Factor” and all the testing and certification they went through to build that.
It’s worse than that. The final production sub, the one that imploded, was intentionally fitted with a window rated to about 1/3rd the depth that the sub was repeatedly pushed to. It’s like the guy thought a hull breach would be a survivable failure.
From what has been released (ahead of the inevitable, if not already proceeding, lawsuits) the people responsible were warned, repeatedly, of the deficiencies and
The field of building private mini-subs in an agile-type way is very much experimental. AFAIK, nobody besides OceanGate ever even tried it.
Does something being “private” (i.e. for-profit) give it’s profit-takers the right to kill their customers?
I don’t see how justice is better served by long drawn-out lawsuits than having the ultimate punishment immediately administered by karma. (karma is literally “actions and consequences as one indivisible concept”, for all that it often eschews well-defined Western-style causal linkages between them.)
I mean, there’s financial compensation to consider, but I doubt any of the clients’ families really need it, and they’re still welcome to have the long drawn out lawsuit against his estate if they want to.

I don’t see how justice is better served by long drawn-out lawsuits than having the ultimate punishment immediately administered by karma.
Other than a warning to the next maverick that thinks their innovations are more important than safety?

I mean, there’s financial compensation to consider, but I doubt any of the clients’ families really need it, and they’re still welcome to have the long drawn out lawsuit against his estate if they want to.
If you want to talk about justice, an integral part should be that all victims are treated equally. Just because the victims are rich, you are advocating they deserve less justice.

I mean, there’s financial compensation to consider, but I doubt any of the clients’ families really need it, and they’re still welcome to have the long drawn out lawsuit against his estate if they want to.
If you want to talk about justice, an integral part should be that all victims are treated equally. Just because the victims are rich, you are advocating they deserve less justice.
Justice is when the victims get what they need, not what they want. If victims always got what they wanted we’d be giving out the death penalty left and right. The rich don’t need the financial assistance, and the people who got on the sub made a choice to risk their lives. They either made an educated decision or chose to be ignorant, they were in a position to determine whether or not they were being hoodwinked and failed to do due diligence to protect their own lives.

Justice is when the victims get what they need, not what they want.
In what world is that justice? Nowhere in the definition of justice [wikipedia.org]: ” in its broadest sense, is the concept that individuals are to be treated in a manner that is equitable and fair” is there a clause of “only if the individuals are not wealthy”. That’s as idiotic as saying that someone who totals my car in a car accident due to their fault they do not need to compensate me for my car if I have two cars or if I can afford to buy another car right after the accident. How idiotic does that sound?

If victims always got what they wanted we’d be giving out the death penalty left and right.
And why ar
Keep in mind that the Titan submersible was owned by a corporate entity; OceanGate Inc. Going after Rush’s estate would be a more difficult affair, as it would mean having to pierce the “corporate veil”, which could really only be done in the event that there was fraud, some sort of co-mmingling or personal liability. Trying to sue Rushton’s estate would have to meet a pretty high bar by demonstrating that OceanGate Inc. and Stockton Rush were heavily co-mingled, and since there was at least one other inves
He was oceangate in all but name, he made all the decisions and he ignored safety advice even firing and threatening to sue the guy who pointed it out and it was his bad judgement that led to the deaths. His estate is there for the taking.

“It was avoidable. It was inevitable.” is a contradiction in terms.
You’re looking at this entirely the wrong way. This isn’t a contradiction: it’s words-stuffing to improve the article’s Google search ranking: there’s a bit of fatalism and optimism in the same bit of searchable text.
Either that or the author just isn’t very good – which Occam’s razor says would be a better explanation.

“It was avoidable. It was inevitable.” is a contradiction in terms. Something that is inevitable by definition cannot be avoided.
It was ironically both avoidable and inevitable because of the leadership. And the story centered directly around that leader. Not really a contradiction.
Yup, sure was the Theranos of the sea. But 10,001 more like it are operating today under the blind bloodlust of greed and fame. The prevalence of this behavior, with both society and govenment simply accepting it, is perhaps more the real story.

“It was avoidable. It was inevitable.” is a contradiction in terms.
“A paradox is a figure of speech that seems to contradict itself, but which, upon further examination, contains some kernel of truth or reason. Oscar Wilde’s famous declaration that “Life is much too important to be taken seriously” is a paradox.” [litcharts.com]
It’s a great lead sentence; paradoxical and intriguing.
https://www.thoughtco.com/what… [thoughtco.com]
https://literarydevices.net/pa… [literarydevices.net]
Hmm, I read it as
It was avoidable by not using the %$#@!ing bubble0gum and bailing wire sub.
It was inevitable if they kept using the sub.
Like, dying if you jump of a skyscraper is both easily avoidable (don’t jump), and inevitable if you do.
If it’s inevitable in practice, how come Jacques Cousteau died on land of a heart attack at the age of 87?
Yes, occasionally submarines sink. Most recently, the other notable sinkings were two Indian subs, one North Korean sub, and the UC3 Nautilus (an intentional scuttling). But mostly this kind of disaster is avoidable, and people have decades of experience in avoiding it. It’s just not cheap, and OceanGate wanted cheap.
I never understood that: Move fast and break things. That’s just a very quick way to end up with broken things.
The correct phrase should be: move fast and break things to make money quickly. That is probably true. You still end up with a bunch of broken things, but also a bunch of money, which I guess is the point.
Unless you get imploded to death, of course.
It should mean, ‘Go quickly to building test hardware. Work out how to test it without injuring people, and test it. Break it. Fix it, build it, test again.” Get to failing fast, so you can learn from your failures. But do it safely!
If they had been following the proper ‘move fast and break things’ philosophy, they would have built loads of test submarines and repeatedly dropped them, unpiloted, into the Mariana Trench. See how they fail, and learn from it. Cycle one repeatedly to its limit. They would have found out that carbon fibre composite makes a really bad submarine, and swiftly pivoted to something else.
You know, like SpaceX, who popularised the expression. They moved quickly into building Merlin engines, and blew them up until they made them work. Then the used them on Falcon 1, and lost 3 of them while learning about rockets. They moved on to Falcon 9, and with what they learned from Falcon 1 and some pretty good luck, got to space the first time. But they didn’t put people anywhere near their rockets until they had got them working very reliably.
Then go all in on developing a carbon fibre starship, building and breaking hardware while learning that carbon fibre isn’t good enough or econmoical enough at cryogenic temperatures. Switch to Stainless steel. Build and fly a water tank!. Keep building stuff and breaking it. Start building Raptor engines, and blow up lots of them learning what you need to fix.
Move fast. EXPECT things to break, so be ready for that!. Don’t build a sketchy prototype and go to sea in it!

…You know, like SpaceX, who popularised the expression…. They moved on to Falcon 9, and with what they learned from Falcon 1 and some pretty good luck,
and some lessons in quality control from NASA, who was funding them…

got to space the first time.
He didn’t do it properly since that would have cost too much money. Instead he was one of those entrepreneurs who wanted to profit instead of actually innovating.
Get it certified. Nope, it cost too much money.
Give tours within country borders. Nope, too many inconvenient laws and regulations to follow.
Purchase new carbon fiber shells. Nah, why do that when I can buy a perfectly unused* one from the great Boeing aerospace company?
Bond titanium and carbon fiber in a clean room. “No, no, no, you see our new soon-to-be-patented maverick method doesn’t need a clean room. It’s so good only household tools are required.”
*materials are expired
I am wondering about the expired carbon fiber. What happens to it over time that it requires an expiration date? There has to e a reason because it is done like that, but still, it’s not like carbon fiber rusts, but something has to happen to it.
You forgot things like, “add a failsafe escape mechanism to explosively open the bolted-on-from-the-outside hatch so that if the sub surfaces with failed communication, the passengers don’t slowly asphyxiate in a sealed container bobbing on the waves because no one knows where they are.”
Or, “build a sphere instead of a cylinder because spheres are stronger.”
Or, “use a viewing port that’s rated to the necessary depth instead of one rated for a much shallower level.”
Or, “design a reliable communications system with the mother ship.”
Or, “include a locating beacon.”
Or, “don’t use carbon fiber in a design that puts it under compression.”
Or really stupid things like, “add comfortable seating that normal people will be able to use for extended periods.”

“Go quickly to building test hardware. Work out how to test it without injuring people,”
should be “

go quickly to designing test hardware which can be tested without killing (or preferably, involving) people
“, then lather, rinse, repeat as desired. If your design for test hardware involves exposing people to significant harm, then it is far from ready for prototyping.
Last night someone mentioned screaming into a vacuum to me in an SF sense, and I reminded them that several people had indeed survived that experience in the early 60s, testing early versions of pressure suits and vacuum suits for NASA. But long before they put people inside suits and into chambers
The silliest episode was the destruction of the concrete launch base, which threw trash everywhere and destroyed property like cars and such. Now they are building a proper one, which they could have done initially if they had wanted. Breaking the previous one was unnecessary for the sake of moving forward – a dangerous show.
I find it also hilarious how the nearby village is danger close and inhabitants need to evacuate themselves from the peace of their own homes during testing and operations. They consented to it, but still. I am not against space exploration at all and I have been a space buff since a little kid, but things have been done better.

You know, like SpaceX, who popularised the expression.
I think it was Facebook who popularized the expression “move fast and break things”:
* https://en.wikipedia.org/w/ind… [wikipedia.org]
* https://en.wikipedia.org/wiki/… [wikipedia.org]

It should mean, ‘Go quickly to building test hardware. Work out how to test it without injuring people, and test it. Break it. Fix it, build it, test again.”
It should mean that. It sometimes does mean that. But frequently it is used as cover by people who actually mean “Move fast and break people” and “transfer the risks to our partners/gig workers/customers”
That sounds more expensive than “cheap out and kill people”
Move fast and break things can work, if the things you break in the process are cheaper than it would be to test and verify carefully. They even did that during the space race when trying to figure out the correct spacing for some holes to drill (I forgot for what part of the rocket engines it was). Just drill some holes, fire it up and see if it runs better. Today we’d probably have the calculating power to just figure it out in simulations, back then it was simply faster to build a mockup and blow it up o
You might NOT want to take that approach… switching my morning coffee for a morning beer was not a good idea.

I never understood that: Move fast and break things. That’s just a very quick way to end up with broken things.
The correct phrase should be: move fast and break things to make money quickly. That is probably true. You still end up with a bunch of broken things, but also a bunch of money, which I guess is the point.
Unless you get imploded to death, of course.
It is also a good strategy for _research_. It is completely unsuitable for any production use where there are real costs if it breaks.
Sounds more attractive now?
I think that in the testing phase, SpaceX and others have shown that you shouldn’t be afraid to break things, as long as you learn something from it. That approach clearly works for rockets, and I see no reason why it wouldn’t work for deep ocean exploration vessels as well.
However, rocket companies know that all they need is one manned rocket to explode and they’re done for, or at least in jeopardy. So people aren’t allowed anywhere near these rockets until they have a lot of confidence that they won’t exp

But in the deep ocean, the price of admission is humidity
There. Fixed that for you.

In a culture that has adopted the ridiculous mantra “move fast and break things,”
That stupid motto comes from the IT culture, and specifically software makers, and for one simple reason: they move fast to make money but you suffer the consequences of bad software.
Every developer seems to believe the old API/software has too many extra special cases/code in it that are unneeded. So they re-write it missing those special cases (and break a lot of stuff) and then spend the next several years adding most of those special cases back in. This all seems to be because developers love writing new code but hate debugging anyone’s code (even their own), and have overconfidence in the re-write.
Had they just fixed the original they would be way ahead. But arrogant and incompetent does make one think that the original developer was an idiot to do all that work to add all of those “unneeded” special cases, and that you can re-write it “correctly”.
My rule is if something weird was done there was probably a good reason. Sometimes maybe the reason was wrong, but often it really was a good reason, so don’t discount all of the special cases.
And cost for agile development seems to be paid for by the users.
Hahahahaha!
That’s got to be the first time Microsoft has been credited with not wanting to break things.
They don’t break backwards compatibility, even when that means leaving great broken fissures in their OS, but everything else is fair game. Especially anything having to do with competitor’s software working properly.

Which competitor’s software did they break?
Caldera DR-DOS.
Microsoft deliberately prevented Windows 95 from running on anything other than MSDOS 7. I know that for a fact because I worked on proving that Windows 95 could run just fine on top of DR-DOS with a little TSR to implement the undocumented INT21h called to check that it ran on MSDOS.
Microsoft got sued over that and lost [wikipedia.org]. Well, they didn’t lose, since no large corporation ever loses a court case in the US: they settled for $180M – and as we all know, settlement is proof of guilt when the settling party has the money to keep fighting in court.

Other than in the legal arena, or their market manipulation practices?
Do you know why I included this caveat? Because the context of the article we’re discussing is talking about the “move fast and break things” culture. We’re talking about the engineering side only. Being anti-competitive has nothing to do with that.
Microsoft doesn’t care about breaking things. Just look at how they handle updates now. They deploy them first to a small, random subset of their customer base, then wait and see if the update breaks anything. If it doesn’t, then they’ll roll it out to everyone. If it does, well sucks to be you if you were one of their unwitting beta testers.
If anyone suffers from my bad software, it’s me. But the approach is still sensible.
When I write stuff, my goal actually is to break things. Welcome to security. If code I write actually breaks something, it’s a happy coincidence and a security flaw I (inadvertently) found.
“Move fast and break things” has its right to exist. The difference between being dumb and being smart is knowing when to use it.
Using it when you bet your life on something is Dumb. Yes, with a capital D. And in boldface.
Open source. Breaking, fixing, quickly and convincing a thousand eyes it’s a feature.
You are mistaken. That is the usual _commercial_ strategy these days. Oh, sure, there are stupid FOSS projects that try to emulate the commercial way, but sane FOSS does it the “if it is not broken, do not fix it” way.
Nonsense. FOSS rarely gets rewrites or refactoring because nobody pays for them. Bad examples like the failed “Systemd” design notwithstanding, although RedHat paid for this crap because they hoped to have a central Linux component under their control. And unfortunately a significant part of the Linux community was stupid and let them do it.
I took a shit yesterday. That was 4 billion years in the making too.
Yeh. I finally went when I had a couple of grilled cheese bean burritos at taco bell.

To me, the more stuff you’ve broken, the more innovative you’ve been.
No one was preventing you from innovating. But innovating doesn’t mean you don’t do testing. The ocean can’t decide to reduce the pressure on innovative vessels. You test your vessel on the conditions, or you don’t. Not doing them is not innovation. Thinking you’re writing safe code just because you marked other code as unsafe doesn’t mean you don’t do testing.
Indeed. This is also a case of “cheaper than possible” engineering. Often “agile” just means “doing it on the cheap” and that can come at a very high price.

People seem to forget this wasn’t the first dive.
No one is forgetting that. The point that you missed is that the many were warning that the vessel would fail catastrophically eventually due to the design. The use of carbon fiber was questioned because it was not a good material for this purpose. And accidents also happen with submersibles or vessels from big established companies, it’s not like there haven’t been any accidents with military subs.
And those accidents were not caused by using carbon fiber. If your mechanic tells you that your brakes need t
Definitely not a lossless algorithm.
Well, it has a 100% loss ratio but also a 100% compression factor.
Also, the whole “disaster” is not very interesting because it was very much and very obviously predictable. There was a small number of casualties, the whole activity was clearly unsafe and risky, and there was no major general impact. Why do we even care? This really is a case of “Those that do stupid things with dangerous objects often die.” and just confirms that sound engineering and risk-management practices are in no way obsolete and that “agile” can (and often will) come at a high price when safety and/or security are a factor.
Well, agreed.
To me what makes the story so interesting is exactly the obviousness. This guy had been getting warnings for years from experts in the field that what he was doing was unsafe, but he ignored them. It’s not that he was innovating in isolation as a pioneer of deep sea exploration and one of his designs failed. There were people that had been developing deep sea vessels for decades coming to him and telling him there would be a catastrophic failure, that his designs were unsafe. There were tests that showed th
Hubris is my guess. And yes, another nice entry for the history of engineering failure. Although I hesitate to call this “engineering”. The approach used is basically an insult to any competent engineer. Fortunately, the joke was on OceanGate in the end.
Everything old is new again. What certain buzzword addicted business drones call “agile”, we used to call “cutting corners”.
Please don’t think my comment is intended to apply to you, by the way. I noted your use of quotation marks to explicitly exclude yourself from that slack-jawed group of language abusers.
He got his Darwin award, and those he fooled to believe in him.
To quote Arthur C. Clarke: “The only way of discovering the limits of the possible is to venture a little way past them to the impossible.” [nsf.gov]
So, turns out if you go beyond the limits of the possible, things break.
Yep. Which is why you don’t put people on your vehicle when testing beyond the limits of what is known to be possible.
That’s why the US Navy has the SUBSAFE program [wikipedia.org].
The environment that submarines operate in really in unforgivably harsh.
At least they learn from their mistakes.
That’s more than can be said for most corporations.
You are angry that this guy had children? There’s no reason to suppose that they will be making the same mistakes, especially as they just lost their Dad.
The good news is, this kind of idiocy isn’t likely to be genetic.
We all have areas where we choose *not* to follow in the footsteps of our parents, because we disagree with them. His kids will likely note the fatal consequences of his approach to safety, and learn a forceful lesson from it.
Considering his son was on the sub… maybe it is genetic.
At that point, his son had (unfortunately) not had the opportunity to see the catastrophic result of his father’s methods. It’s unfortunate that he had to pay the price personally.
Perhaps not genetic, but still heritable. There’s a strong tendency for wealthy children to de indoctrinated with the arrogance and stupidity of their parents. That’s presumably where Stockton Rush got his. And if he had stuck with something business-oriented, where influence and abuse of power can compensate for a multitude off sins, he’d probably have done quite well for himself, offloading the price of his failures onto less influential participants. Just as so many others do.
But he made the mistake
Fun fact, one of these was involved in the rescue of a sunk sister ship back in the 1970s:
https://en.wikipedia.org/wiki/… [wikipedia.org]

“The more stuff you’ve broken, the more innovative you’ve been”
Sounds like a case of “fake it ’til you break it”…
I haven’t heard of a significant fire threat. What have you heard?
AIUI, the chamber atmosphere was surface pressure air, so what you need for maintenance is an air pump, an oxygen bottle, a CO2 scrubber, a couple of sensors, and some glue logic. Essentially, take the well used control systems from a (couple, for redundancy) SCUBA rebreather system and stow it under benches along the walls. Note the count of hull penetrations : zero. Source : COTS ; use two different manufacturers to avoid both failin
“Pale, really isn’t a verb.”
Um…really, it is:
https://www.merriam-webster.com/dictionary/pale [merriam-webster.com]
There may be more comments in this discussion. Without JavaScript enabled, you might want to turn on Classic Discussion System in your preferences instead.
Linux’s Marketshare on Steam Still Higher Than Apple macOS
Are Scrums a Cancer?
“When anyone says `theoretically,’ they really mean `not really.'” — David Parnas

source