Allan McDonald refused to approve Challenger launch, exposed cover-up (2021)

I used to run the nuclear power plant on a US Navy submarine. Back around 2006, we were sailing somewhere and Sonar reported that the propulsion plant was much, much louder than normal. A few days later we didn't need Sonar to report it, we could hear it ourselves. The whole rear half of the ship was vibrating. We pulled into our destination port, and the topside watch reported that oil pools were appearing in the water near the rear end of the ship. The ship's Engineering Officer and Engineering Department Master Chief shrugged it off and said there was no need for it to "affect ship's schedule". I was in charge of the engineering library. I had a hunch and I went and read a manual that leadership had probably never heard of. The propeller that drives the ship is enormous. It's held in place with a giant nut, but in between the nut and the propeller is a hydraulic tire, a toroidal balloon filled with hydraulic fluid. Clearly it had ruptured. The manual said the ship was supposed to immediately sail to the nearest port and the ship was not allowed to go back out to sea until the tire was replaced. I showed it to the Engineer. Several officers called me in to explain it to them. And then, nothing. Ship's Schedule was not affected, and we continued on the next several-week trip. Before we got to the next port, we had to limit the ship's top speed to avoid major damage to the entire propulsion plant. We weren't able to conduct the mission we had planned because the ship was too loud. And the multiple times I asked what the hell was going on, management literally just talked over me. When we got to the next port, we had to stay there while the propeller was removed and remachined. Management doesn't give a shit as long as it doesn't affect their next promotion.

Don't even get me started on the nuclear safety problems.

Saying "no" is easy and safe in a world where there are absolutely no external pressures to get stuff done. Unfortunately, that world doesn't exist, and the decision makers in these kinds of situations face far more pressure to say "yes" than they do to say "no".

For example, see the article:

> The NASA official simply said that Thiokol had some concerns but approved the launch. He neglected to say that the approval came only after Thiokol executives, under intense pressure from NASA officials, overruled the engineers.

Not in my experience. Saying no to something major when others don’t see a problem can easily be career-ending.

There's a big difference between "complaints" because something is not optimal, and warnings that something is a critical risk. The Thiokol engineers' warnings about the O-rings were in the latter category.

And NASA knew that. The summer before the Challenger blew up, NASA had reclassified the O-rings as a Criticality 1 flight risk, where they had previously been Criticality 1R. The "1" meant that if the thing happens the shuttle would be lost--as it was. The "R" meant that there was a redundant component that would do the job if the first one failed--in this case there were two O-rings, primary and secondary. But in (IIRC) June 1985, NASA was told by Thiokol that the primary O-ring was not sealing so there was effectively no redundancy, and NASA acknowledged that by reclassifying the risk. But by the rules NASA itself had imposed, a Criticality 1 (rather than 1R) flight risk was supposed to mean the Shuttle was grounded until the issue was fixed. To avoid that, NASA waived the risk right after reclassifying it.

> at some point, one needs to say "yes" and take risks, otherwise nothing would be done

Taking calculated risks when the potential payoff justifies it is one thing. But taking foolish risks, when even your own decision making framework says you're not supposed to, is quite another. NASA's decision to launch the Challenger was the latter.

Even in the case of the Challenger, no single article say WHO was the executive that finally approved the launch. No body was jailed for gross negligence. Even Ricahrd Feynman felt that the investigative comission was biased from the start.

So, since there is no "price to pay" to make this bad calls they are continuously made.

The safety posture of that whole program, for a US human space program, seemed bad. That they chose to use solid rocket motors shows that they were willing to compromise on human safety from the get-go. There are reasons there hasn't ever been even one other human-rated craft to use solid rocket motors.

Every kind of meaningful success involves negotiating risk instead of seizing up in the presence of it.

The shuttle probably could have failed in 1,000 different ways and eventually, it would have. But they still went to space with it.

Some risk is acceptable. If I were to go to the moon, let’s say, I would accept a 50% risk of death. I would be happy to do it. Other people would accept a risk of investment and work hour loss. It’s not so black or white that you wouldn’t go if there’s any risk.

Do they? Even if risks are not mitigated and say risk for catastrophe can't be pushed below ie 15%? This ain't some app startup world where failure will lose a bit of money and time, and everybody moves on.

I get the political forces behind, nobody at NASA was/is probably happy with those, and most politicians are basically clueless clowns (or worse) chasing popularity polls and often wielding massive decisive powers over matters they barely understand at surface level.

But you can't cheat reality and facts, not more than say in casino.

Or: at some point, one decides to launch the thing.

You are reducing the complaints of an engineer as something inevitable and unimportant, as if it happened in every lunch, and in every lunch someone decided to went ahead, because it was what was needed.

Do they though? If the Challenger launch had been pushed back what major effects would there have been?

I do get your general point but in this specific example it seems the urgency to launch wasn’t particularly warranted.

I'm wondering how the two astronauts on the ISS feel about that while Boeing decides if/when it is safe to return then to Earth.

https://www.cnn.com/2024/06/18/science/boeing-starliner-astr...

Read the comments (especially from NASA engineers). It's pretty interesting that sometimes it takes courageous engineers to break the spell that poor managers can have on an organization.

Nixon even had a 'if they died' speech prepared, so someone had to put the odds of success not at 100.

We don’t see software engineers behave ethically in the same way.

Software is filled with so much risk taking and there’s few if any public pushback where engineers are saying the software we’ve created is harmful.

Here’s a few examples:

- Dark patterns in retail

- Cybersecurity flaws in sensitive software (ie. Microsoft)

- Social media and mental health

- Social media and child exploitation / sex trafficking

- Social media and political murder (ie. Riots, assassinations)

This stuff is happening and it’s just shrugs all-around in the tech industry.

I have a ton of respect for those whistleblowers in AI who seem to be the small exception to this rule.

True, but that is for cases where you take the risk yourself. If the challenger crew knew the risk and were - fuck it - it's worth it it would have been different than a bureaucrat chasing a promotion.