Who?

🔗 OceanGate's Contribution to the World

text

First published: .

Much has been said on the implosion of OceanGate's Titan submersible. Much is still being said. The media is currently on a crusade of vilifying the company's founder & CEO, Stockton Rush, and explaining away the disaster as an inevitable outcome of his nefarious actions. The same media that gladly advertised the company when the oceans were calmer. This disappoints me, because for a short while there, it really seemed like the media, and people in general, were starting to catch on to the real issue at heart.

For all its (many) faults, OceanGate did expose to the world just how unreliable these "tech companies" often are, how their businesses don't really make any financial sense, how what they deliver only slightly resembles what they promise, and how the existing, well-established solutions in the market are often superior. These sentiments were very common during the first days of the submersible's disappearance, but all but disappeared after the sub's remains were found and the media began pushing its villain narrative.

The way I see it, Stockton Rush was a run-of-the-mill founder of a venture capitalist backed tech company. I believe that every action, every choice, every decision made by Rush/OceanGate can be attributed to this simple fact. Here's why:

Privately owned, VC funded tech companies are founded based on an idea that is either supposed to appeal to a gigantic amount of people, or, more often than not, promises to revolutionize an industry, making it much cheaper and more accessible to the general public (which is a nicer way of saying "cornering a market"). The idea that these things can be made much cheaper is based either on nothing, or on a false belief that established solutions are all made by incompetent idiots building bloated solutions and milking the public of its money. And the founders of these startups usually have no prior knowledge or training in the field. Stockton Rush may have been an engineer and a former pilot, but had no relevant knowledge to the deep sea exploration industry.

The ideas are often grandiose and require a lot of money. However, the companies usually start their lives with a relatively small "seed funding" from these VCs (or investment banks, funds, etc.) that is a far cry from the amount of money actually needed to achieve their goals. This is different than "normal businesses", which often begin with a loan that is large enough to get the business up and running.

After raising seed funding, the companies start hiring employees, purchasing assets, and developing their solutions. The money, however, is not enough and can only last a certain amount of months/years, for a solution that can take many years to develop. This means that a subsequent "funding round" from the investors (whether new or the same) is required. This funding round, called "Series A", will also only be enough for a certain amount of months/years, so more funding rounds will be needed, Series B, Series C, Series Z44D. VC-funded companies are, therefore, in a constant state of financial loss, and a constant need of investments. This need drives everything the companies do, and leads every decision process, for years and years. Probably more years than you'd expect.

The paths undertaken by these companies are often similar. To get first funding, they will try to find "design partners" - other companies who have expressed interest in the proposed solution and have agreed to provide their input in the design process, in return for getting the product for little to no money when it's available, and show those agreements to potential investors. For the next funding round, the company will need to show investors that it has finished an initial version with several POC "customers". For the next funding round, they'll need to show a GA version and a certain amount of paying customers. For the next round, more customers, a product expansion, and so on, and so forth.

Very quickly, the companies learn that money has an annoying tendency to run out. R&D is expensive, sales and marketing are expensive, assets are expensive, everything is God-damn expensive, and here they are trying to make things cheaper. Between funding rounds, therefore, the money is always trending down quickly, so the companies are in a hurry. They need money, and quickly, otherwise the whole revolution goes kaput. And since they need to show all I mentioned in the previous paragraph in order to get that money, they need to do all those things ASAP. That's why every startup seems to be working in a constant state of hysteria; everything has to happen yesterday, despite the fact that absolutely no one is really waiting for the solution.

And because everything has to happen yesterday, compromises have to be made:

These "attributes" can be a hard pill to swallow for many employees, which leads to high turnovers, in turn leading businesses to prefer employees less likely to leave: young, inexperienced, willing to work 12 hour days, and unlikely to say "No". Months ago, in a Reddit conversation about Tesla with a smart and experienced person working in a safety-critical industry, he explained this quite well:

Developing these types of safety-critical systems, especially a largely novel safety-critical system, will require enormous investments in capital and time before a product can be offered to the general public. Validation is extremely expensive - both initially and continuously. Musk (and Tesla) wanted to pocket something right away - and they clearly did. In such a business environment, engineers that are competent in safety-critical systems are persona non grata.

This is an astute observation, and something I often observe in my dealings with startups and tech companies, but it wasn't until he wrote this that I was able to fully understand what drives it. Domain experts are notoriously absent from tech companies because they are more likely to say "No" and hinder the founders' goals.

Laws and regulations are often ignored in the "tech" industry, with company executives believing they can because "we are innovating", a ridiculous claim that makes less sense than the supposed innovations themselves. Problem is, many jurisdictions turn a blind eye, allowing tech companies to do as they wish. But only for a while. Eventually, the law catches up with the companies.

Remember all those electric scooters that were illegally dumped on the sidewalks of your once nice and walkable city by the thousands, littering the streets, blocking sidewalks, crosswalks, public benches, building entrances, and ignoring every possible city ordinances and traffic laws? These ridiculous startups wasted gigantic amounts of VC money to build these fucking scooters, and lobbied to avoid regulations with a "we are revolutionizing the something something whatever industry" excuse. Well, the law eventually did catch up with those startups, and in most jurisdictions the number of scooters dwindled to almost nothing, and they can no longer be "parked" anywhere you bloody want, taking away one of their major selling points, proving that the entire idea made zero sense.

So where does OceanGate come in? The company was founded in 2009, but according to Crunchbase, only raised $500K in seed funding 2014, a laughable amount of money you couldn't develop a rubber bathtub submersible with. Six years later, in 2020, they raised a further $18M, with the supposed intention to build a fleet of submersibles, something that seems never to have happened.

OceanGate was founded after a severe financial crisis that probably affected its ability to raise funds. But adding the fact that the seed funding was so paltry, that it took 6 years for a more serious funding, and that the investors preferred to remain anonymous, points to OceanGate having had serious trouble raising funds. As reported by the New York Times, Karl Stanley's email to Stockton Rush in April 2019, almost a year before its substantial second funding round, and sent after Stanley dove with the Titan and heard loud cracking noises, hints at those financial troubles:

“A useful thought exercise here would be to imagine the removal of the variables of the investors, the eager mission scientists, your team hungry for success, the press releases already announcing this summer’s dive schedule,” wrote Mr. Stanley, according to a copy of the email seen by The New York Times. “Imagine this project was self funded and on your own schedule. Would you consider taking dozens of other people to the Titanic before you truly knew the source of those sounds??”

Out of all the internal communications exposed since the disaster, this is, in my mind, the most revealing one. Karl Stanley was clearly implying that Stockton Rush was ignoring safety concerns and known issues in a rush (pun not intended) to please and obtain investors. There's obviously no rush (pun intended) to make Titanic dives publicly available, who cares? Why is there a tight schedule, if not because the money is running out, and investors are wary of putting their money into this enterprise?

OceanGate also displayed another common attribute of startups with grandiose ideas making bad choices: those bad choices tend to snowball. It can be very hard for some people to admit they were wrong, especially when large sums of money are involved, and it's even harder for startups to pivot successfully. OceanGate made a decision to build a submersible out of a material that many engineers and experts in the industry knew was unfit for the intended use case. It was warned about it, lost employees and sued over it, but did not back down.

OceanGate may have been able to redesign their sub to address these concerns, but with money running out, it's possible doing that would have been a death blow for the company. So what did OceanGate do? It doubled down, adding ridiculous patches with the hopes they will help overcome the glaring issues with the design, such as the company's absolutely bonkers decision to strap on an "acoustic monitoring system" meant to alert the crew when a catastrophic destruction was imminent. It's bonkers because an alert system can't fix a bad design, and yet it was touted as what made the Titan safe. Even more than that, the system itself was unproven technology, which Rush tried to spin as another exciting innovation. One bad decision often leads to other ones.

The biggest problem with these tech companies and their modus operandi, however, is that they are encouraged to maintain these bad attributes. You see, startups don't actually need to succeed commercially in order for the founders and investors to succeed financially. Startups and tech companies aren't generally founded to revolutionize or improve anything, they are founded to make the founders rich via a lucrative "exit". Have you ever stopped to think why the word "exit" is used to describe the sale of one company to another one? It comes from the sold company exiting the market, ceasing operations. Most startups that sell do just that: disappear off the face of the earth. The purchasing company may integrate the startup's product(s) into its own, but often entire products are reduced to a checkbox in the purchaser's product's settings, or the products are completely scrapped.

Most startups are literally built with the purpose of shutting down. Massive gains can be made out of extremely unprofitable businesses. Here's how founders get rich from terrible ideas, and forgive me if I'm repeating a few sentences from earlier in the text:

  1. A small group of founders found a new company. They own 100% of the company. They don't distribute shares because there's no one to distribute shares to (except themselves), but they also have no money to fund the company's activities, hiring employees and purchasing assets.
  2. The company looks for investments from investment banks, venture capitalists, individual/angel investors, friends and family. The founders relinquish a certain percentage of their ownership of the company to these investors, and distribute shares that reflect those percentages. Usually, the founders will still have a very large percentage of ownership after this first funding round.
  3. With the new funds, the company hires employees, purchases assets, does research and development, etc. But this money doesn't last, and a new investment round is necessary quickly.
  4. In a second investment round, the company looks for more funds from either completely new investors, the existing investors, or both. Existing investors are offered one of two options: they can either increase their stake by buying shares from you/other investors, or they can decrease their stake by selling some of their shares to new investors. Of course, the company can also issue more shares, but it will always affect existing shareholders. All parties agree to a share price (and company value) based on various factors that for lack of a better word are completely imagined.
  5. One or more of these investors buys some of the founders shares. This purchase comes with some restrictions meant to prevent the founders from simply running away with the money and forgetting about that company they founded, such as agreeing to continue on in their capacity for at least X more years, having the money dispersed over a period of time or via stock options, etc. Regardless, in this second round, most of the money from investors goes to the company, but some will go to the existing shareholders. The founders are now rich, even though the company is still not making any money. Previous investors may already have a positive return on their investments by being bought out by new investors.
  6. Rinse and repeat. This can literally go on for years, and many funding rounds. SpaceX went through at least 30 of them.
  7. The company launches an IPO, allowing the general public to buy out the founders/existing investors (this is how Elon Musk became the richest asshole in the world and been able to purchase Twitter), or exits the market by selling to another company which buys out all existing shareholders (i.e. investors/founders).

This strategy is only different than a Ponzi scheme in that new investors know they are bailing out previous investors (and vice versa), an action they take gladly in the expectation that they will be bailed out themselves later on. This fact has allowed a large number of people to become obscenely rich by building terrible businesses with terrible products, ignoring laws, regulations, safety concerns, public health, and more. OceanGate is merely an example of how dangerous this strategy is: the longer it takes to reach the exit, the bigger the chance that the entire house of cards will collapse. And OceanGate's house of cards collapsed spectacularly.

OceanGate's website now simply says that "OceanGate has suspended all exploration and commercial operations." Many people think this is because the company is going to be buried in the courts, which may be part of it, but it's much more likely that OceanGate is suspending all operations because it has no money, and no prospects of ever getting any money, because no investor in their right mind will bail out this company right now.

By the way, isn't it absolutely hilarious how when the submersible that was going to revolutionize deep sea diving went missing, absolutely no one in the entire fucking world batted an eyelash when actual, real, well designed submersibles were sent to find it? Nobody found it weird that these subs already existed and were able to dive down 3,800 meters in (effectively) a moment's notice to perform complicated search operations. That's another common attribute of tech companies: they like to ignore the fact that the problems they're solving have already been solved, better, safer, and cheaper, and as long as the public and investors are also willing to ignore that, everything is good.

So OceanGate did expose a bit of the inner workings of revolutionary tech companies, but people are quick to forget. In the end, though, OceanGate's real contribution to the world will probably be exactly what it set out to destroy: more laws and regulations, to be heeded by the good players in the market, and ignored by the bad ones.