32 comments

[ 2.7 ms ] story [ 71.6 ms ] thread
> When OceanGate was founded the goal was to pursue the highest reasonable level of innovation in the design and operation of manned submersibles. By definition, innovation is outside of an already accepted system. However, this does not mean that OceanGate does meet standards where they apply, but it does mean that innovation often falls outside of the existing industry paradigm. [1]

Christ they even used the word “paradigm.” Put this down in the list of crazy startup stories that needs to be in the Silicon Valley reboot.

Anyone who works in DSVs needs to take the Ballad of James Cameron very seriously:

    No budget too steep, no sea too deep
    Who's that? It's him, James Cameron!
[1] https://web.archive.org/web/20220812024916/https://oceangate...
It’s unclear to me what the “63 years without a fatality” track record refers to. Private submarines?
Same here. There have been hundreds of documented fatalities on lost subs since 1960.
I think he's referring to subs diving specifically to the Titanic.
The Titanic wreck wasn't found until 37 years ago, so that doesn't make sense either.
>It (OceanGate) added that the classification agencies "slowed down innovation… bringing an outside entity up to speed on every innovation before it is put into real-world testing is anathema to rapid innovation".

As a private pilot who is considering building a kit aircraft, this is certainly true. It can be exceedingly expensive to prove safety standards have been met for new designs, which is why so few new certified aircraft are on the market and old designs are so expensive. This leads to the whole industry slowing to a crawl. But that absolutely doesn't mean safety standards haven't been met with new designs.

However, if I build a kit plane, the FAA will require it be emblazoned "Experimental" and commercial activity - such as paying passengers - is forbidden.

I hope the paying customers of this sub were well aware of the level of risk they were taking.

All that being said, I'm supportive of anyone's right to go and build a contraption that may end their life, so long as they don't take anybody or anyone else's property with them.

>However, if I build a kit plane, the FAA will require it be emblazoned "Experimental" and commercial activity - such as paying passengers - is forbidden.

unless you're a big company

https://www.reddit.com/r/flying/comments/ono50t/ever_see_a_m...

Reading the comments, that plane was actually fully certified, they just forgot to remove the sticker.
Ignorant response. Just because you see a sticker doesn't mean anything. The aircraft is certified to fly commercially.

Its the same as people freaking about speed tape.

To quote the CBS reporter who recorded an earlier journey with them, as he read the waiver form¹:

>...An experimental submersible vessel, that has not been approved or certified by any regulatory body, and could result in physical injury, disability, emotional trauma, or death. Where do I sign?

Probably not enough of a disclaimer, especially if they had people read and sign it once they were already miles out at sea. The passengers were technically advised of the risks, though.

¹: https://youtube.com/watch?v=29co_Hksk6o&t=2m38s

There are externalized costs besides death though, like the resources deployed in the search and rescue effort. In the case of aviation you can also cause collateral damage or death.
Yes, but you have to allow a certain amount of freedom or everything grinds to a standstill.

The approach the FAA takes is perfectly reasonable: you can do this dangerous thing, but you can't legally make money from it, and if you "forget" to tell your passengers, the aircraft itself is labelled "Experimental."

I am happy for my tax dollars to go to efforts like these. The world is cooler when we have people doing rad stuff.
This applies until the cost of people doing "rad stuff" is too high to bear socially.
which is why you most probably can't play with plutonium in your backyard
Search and rescue is often seen by the military as a high quality training exercise. They feel the stress of a real exercise where someone’s life is on the line.
> is anathema to rapid innovation

So what? Rapid innovation is not always what should happen. It always comes with pretty serious risks and not insignificant costs.

In a free country, yes, anyone has the right to do so. Not my place to judge if I'm going to continue watching Jackass movies.

But I'm not a fan of people like this guy vaguely shaming those who don't like to take lots of risks. I'm not a daredevil, if I got on a plane I would expect it to be safer than the trip to the airport.

The FAA, and the elevator industry, are two of the groups that actually do things the way I like to see them done. If there is one single crash, they will train pilots and update firmware or even retrofit hardware to stop that one specific scenario.

If there's a near miss, and someone could have crashes, they might do the same.

Anywhere else, we just say "Yeah, we know this kills X people a year, but if ya wanna be safe don't get up".

Innovation is largely overrated. A lot of the useful stuff seems to be held up on the basic science rather than willingness to do hands on experiments.

If there's tech to save fuel, they'll do it because it's cheap. If there's tech to make it safer, the FAA will probably require it.

If there's some really clever trick that does neither but makes it simple and cheap and interesting, I'm quite happy for it to just remain in the experimental stuff.

Whenever people talk about innovation being more important than safety it always sounds like typical hacker ethos "I want a world full of lots of different random technical ideas that are simple and can be understood"..... "love of ideas" more than love of engineering progress in the modern sense.

The hull was made from 5-inch thick carbon fibre. Carbon fibre is very lightweight for such a strong material, which makes it attractive, but it has a few critical downsides:

• it tends to fail catastrophically, without warning, after repeated stress

• it's very difficult to predict how many stress cycles it can endure

• it's very difficult to detect stress cycle damage

Considering every dive and resurface is a huge stress cycle, and the fact OceanGate didn't seem to have any sort of independent certification of their submersible's endurance, this kind of hull implosion event seems inevitable. Their real-time hull monitoring system was a boondoggle, n

Edit: In another interview, Cameron elaborated on these exact concerns: https://www.youtube.com/watch?v=V-yu_wixdgE&t=1m20s

An important note is that an employee was (allegedly) fired for making a point that the carbon fiber was two inches too thin.

They also apparently didn't even report it until 8 hours after noticing it didn't resurface once communications stopped.

If that's true, James Cameron's assessment rings true.

What's the benefit of carbon fiber's strength-to-weight ratio when making a submersible?

I guess you can have smaller motors to drive you around on the bottom, but the vast majority of the travel is controlled via ballast/buoyancy.

To me, planes and cars have much more obvious gains from lighter materials. Not so much for the sub.

(comment deleted)
I'm not sure why I highlighted that because indeed that wouldn't be useful, though as a proxy for density it is (apparently they picked it for its buoyancy).
The submersibles need to be less dense than water on average so they rise when you release ballast.

It's always a cost tradeoff - I think Challenger Deep (Cameron's former sub) was steel construction which required a large volume of syntactic foam that can resist high pressures to counteract the weight of the steel. That makes for a much larger vessel.

Titan's submersible for exploring the Mariana Trench I believe is largely titanium so much lighter than steel, but at a hugely increased cost.

> What's the benefit of carbon fiber's strength-to-weight ratio when making a submersible?

Not having to bond expensive foam to it to make it buoyant. I also think it may have something to do with the shape of the hull in this case, but I have not seen any commentary on that yet.

Right - the first thought which came to my mind - there was only one prototype sub, had it been tested for say a dozen trips? What happens? how much wear and tear, all that stuff you don't call out in the agreement to opt-in to death.
> "I knew that sub was sitting exactly underneath its last known depth and position. That's exactly where they found it," he continued.
Just him? I thought that was the general consensus.