[_zamorano_]

So I suposse building an aircraft involves standard bolts, procedures, testing and much more standarized ways.

In contrast a spacecraft like the shuttle, faces much harsher conditions and, as not many of these were built, I expect more manual procedures and tinkering while building the thing.

In the end, it's incredible these things didn't crash more often.

[alexpotato]

NASA has incredibly detailed records about every piece of their spacecraft down to things like:

- material used to make a bolt

- what the torque used to tighten the bolt was

- who tightened the blot

- when it was tightened

- etc

This allows them trace back through the history of each vehicle for debugging purposes.

They also applied this to the Space Shuttle software. This article from 1996 does an amazing job of describing the process: https://www.fastcompany.com/28121/they-write-right-stuff

It's interesting how modern some of the practices described are. Plus, some of the practices (E.g. the bug rate model), from my experience, only existed there.

[dotancohen]

  > It's interesting how modern some of the practices described are.

It should be noted that among famous NASA inventions, modern Project Management is listed among them.

[hinkley]

A reusable aircraft that faces that sort of intense vibration, I'm not at all surprised that we need to track when the last time each bolt was checked and by whom.

(One of the things you have to watch out for is that if the torque on a nut drops for no reason, it may be a hairline crack in the bolt it's attached to)

[rachofsunshine]

The risk of spaceflight is still very high. Wiki [1] lists 676 people as having traveled to space, of whom 19 have died in accidents as a result of that travel, meaning that going to space has about a 3% chance of killing you.

The average age of an astronaut is 34 [2], and most are male, so a look at an actuarial table [3] tells us that going to space is approximately as likely to kill you as literally every risk an ordinary person would take in their life up to that point (at 34 years of age, about 4.3% of men have died, and a large proportion of those deaths are due [4] to accidental injury).

[1] https://en.wikipedia.org/wiki/List_of_spaceflight-related_ac...

[2] https://en.wikipedia.org/wiki/NASA_Astronaut_Corps#Qualifica...

[3] https://www.ssa.gov/oact/STATS/table4c6.html

[4] https://www.ncbi.nlm.nih.gov/books/NBK600454/table/ch2.tab4/

[Gazoche]

> The risk of spaceflight is still very high. Wiki [1] lists 676 people as having traveled to space, of whom 19 have died in accidents as a result of that travel, meaning that going to space has about a 3% chance of killing you.

But 14 of those were caused by the shuttle alone. All the others were over 50 years ago. So far, all the spacecrafts still in use today have had a pretty good track record.

[Retric]

That 19 is a rather narrow list. It excludes the Apollo 1 mission where astronauts died in the spaceship during a rehearsal etc. In total 11 died during training including a cosmonaut in 1993 and a Spaceship 2 test pilot in 2014. “As of 2024, there have been over 188 fatalities in incidents regarding spaceflight.”

The shuttle also carried over half of all astronauts (355) on orbital missions, so if you’re excluding the shuttle it’s not that much safer.

Soyuz MS is a refined design, but Soyusz 11 killed 3 people and Soyuz 1 killed 1. Calling it a different design isn’t unreasonable but by that token it would be limited to 22 successful crewed missions and 1 in progress.

[wongarsu]

Out of 355 astronauts that have ever used the shuttle, which comes out to about 4%. Not that much worse.

The shuttle's lack of a launch abort mechanism is something NASA wouldn't accept in any modern human-rated spacecraft. But arguably the deadliest feature of the shuttle was that it was pushed as the single launch platform for all launches, even those that didn't require any crew. Putting crew on every single flight made many missions more risky than they had to be

[cfraenkel]

And also made missions at least 3X more expensive..... It wasn't just the on-paper launch costs. Everything had to be man-rated, meaning, among many other things, everything that operated during the launch had to be triple redundant, all the pyros had to be unpowered while onboard the shuttle (meaning you had to design another system to then power up the pyros and make it reliable), you needed three full launch crews (Cape, Johnson, plus wherever you actually ran your own ops) and all three launch crews had to support an endless set of rehearsals and launch delays.... The costs kept mounting. (source - was in program office of expendable launch comm sat, each satellite was ~$150M, launch was ~$80M. Roughly comparable mission down the hall cost ~$300M / satellite, ~$500M / launch.)

[dboreham]

We're there more Soviet accidents that we don't know about?

[bombcar]

Probably on the ground, but we likely know or highly suspect most of the ones in the air or in space.

[bell-cot]

> ... of whom 19 have died in accidents ...

Skimming your reference [1], I see 11 more who died in accidents during testing & training. Including the https://en.wikipedia.org/wiki/Apollo_1 fire on the launch pad (during a launch rehearsal test).

Until spaceflight is "buy your ticket, show up, get in your seat, wait, exit at your destination", I'd argue that we should include the testing & training risks in the risk of human spaceflight.

[lobochrome]

Small populations make for terrible statistics.

[bell-cot]

Often true.

But when that many people die, in that many separate incidents, across a variety of nations & launch vehicles - then the "The risk of spaceflight is still very high" thesis is statistically solid.

[Aachen]

An estimate is better than no clue at all

[leoedin]

To what extent is that true when the cause of the fatalities is the technical design of completely unrelated systems?

If one space agency built a rocket which always immediately exploded after launch, and another space agency built one which always worked, you could say the odds of failure for the next astronaut was 50%. But of course the two rockets are essentially unrelated. The chance of success of each rocket is a function of design, engineering process, organisational culture of that organisation.

Telling the astronaut strapped to the top of the explody rocket that there's a 50% chance of exploding is actually less help than no estimate. Because actually there's a 100% chance of them exploding. An estimate is only as valuable as the assumptions that drive it.

[jajko]

Or taking a guided climb of Mt Everest, roughly same numbers (compared to Aconcagua which sits around 30%)

[dghlsakjg]

Aconcagua is not notoriously difficult, and around 1k people per year summit. I can't find anything that says that hundreds of people per year are dying.

Do you mean Annapurna? It, at one point had a death rate above 30%, but is now below 20%. K2 has taken over the crown for deadliest mountain with a death rate of 24%

[ghaff]

>compared to Aconcagua which sits around 30%

??? Aconcagua doesn't have an especially high fatality rate.

[krisoft]

> at 34 years of age, about 4.3% of men have died

I think what you are saying is that 34 years after being born 4.3% of male individuals are dead.

In my understanding if someone dies when they are 10 years old they will never be "34 years of age". This probably feels nitpicky but it has thrown me into a loop of trying to understand what you are saying.

(Not even mentioning that I read the table you linked as 4.2% not 4.3%)

[chasd00]

Saying 95.7% of men live to at least 34 would be more clear. (i have no idea if that stat is legit or not, i'm just writing it differently)

[krisoft]

That sounds even better! Thank you for that rephrasing.

[pc86]

This seems needlessly pedantic as the meaning is still very clear.

And you did mention it, by saying you weren't mentioning it.

[krisoft]

> This seems needlessly pedantic as the meaning is still very clear.

As I said it wasn’t clear to me. The two meaning which was fighting in my mind were the one i wrote and that the percentage is the probability of a male dying in their 34th year of life. Had to consult with the table to figure out which one they mean.

> And you did mention it, by saying you weren't mentioning it.

Well spotted. Exactly because the discrepancy troubles me. It either means that I don’t understand how to read or what to read in the table (in which case I would love to be corrected) or that the commenter made a typo (which doesn’t matter at all). If i were certain it is a typo I wouldn’t mention it. But since I can’t be certain that the error is not “in my equipment” i shared the observation hoping to get clarification.

[goodcanadian]

That was part of the rationale behind the space shuttle in the first place. It was to create a space plane with aircraft-like operations in order to fine tune processes and technology to bring down the cost of space flight. Unfortunately, NASA never managed the operational cadence required, in part, because of the per-flight cost (which was, in turn, high, in part, because of the low flight cadence). It was a fine idea, but it didn't work out so well in practice.

[bsder]

It also didn't help that the design was compromised.

In order to get funding from the miltary, the shuttle had to be able to switch to a polar orbit which is why it had those stupidly large engines that serve no purpose otherwise.

If you get rid of that, you actually can design a reusable space plane.

[lupusreal]

Turns out, making something look like an airplane doesn't mean you can treat it like an airplane.

[dotancohen]

The space shuttle's appearance has little to do with its flight cadence - the airplane-like flight cadence (and thus, airplane-like reliability and cost) just never manifested. Especially after January 1986.

[pfdietz]

Also, the smaller number, and the smaller number of flights, means much less experience to wring out the low probability gotchas.

Today's extremely reliable airliners got that way on a long, long string of accidents and near accidents.

[jccooper]

Though a large batch and large amount of flights by space standards, Shuttle was basically all prototypes by normal manufacturing standards. The 135 STS flights wouldn't even make a dent in an airliner certification and test campaign. It's not surprising they kept encountering problems.

[hinkley]

Even in commercial aerospace, every part is tracked like a library book. There's rarely a question about whether a particular part is the right part or more importantly a used part. Because there's a chain of custody for each one.

You also have some parts that are destined for QA purposes, and those have a tagging system that is meant to prevent them from being recycled onto a real aircraft once they've been used for stress testing.

[kelnos]

> In the end, it's incredible these things didn't crash more often.

I think two catastrophic space shuttle failures is more than enough :-/