Comment by jmyeet | Hacker News Reader

Comment by jmyeet | Hacker News Reader

Boeing is (yet another) cautionary tale for what happens to a company when the finance people take over, which is essentially inevitable once a company has a "monopoly". Yes I know there's Airbus but a duopoly can be pretty functionally similar to a monopoly.

The oft-quoted Steve Jobs quote on why Xerox failed [1] is always relevant.

Starliner, SLS and Artemis both behave exactly like a jobs program gone bad. Now I'm not opposed to a jobs program per se. We could honestly use more of that. But the difference between Boeing's efforts and SpaceX's efforts is just astounding.

Starliner launches on the fully expendable Atlas V N22 rocket, which reportedly costs $164 million [2] to launch. Falcon 9 seems to cost ~$62 million per launch [3] before you even factor in reuse.

I don't know how (or even if) a company can be rescued when it gets to this stage. It becomes a rent-seeking Frankenstein of tiny warring fiefdoms. We've also seen this with Intel's complete inability to internally collaborate as mid-level VPs seemingly engage in turf wars rather than share resources and information.

EDIT: fixed typo (Xerox)

[1]: https://www.youtube.com/watch?v=tGKsbt5wii0

[2]: https://en.wikipedia.org/wiki/Atlas_V

[3]: https://nstxl.org/reducing-the-cost-of-space-travel-with-reu...

> Now I'm not opposed to a jobs program per se.

In the 1990s we managed to get enough evidence that useless jobs are worse than just giving people money in every way. So that idea died almost everywhere.

Governments putting money on unprofitable but useful things is great. But they really should try to be efficient on those.

> In the 1990s we managed to get enough evidence that useless jobs are worse than just giving people money in every way.

Because the government money never goes equally to the workers. Would there be better outcomes if government money only went to co-ops and worker unions?

There would be better outcomes if the money were used to pay down debt, to reduce future inflation, or give rebates, to boost growth somewhere.

> to pay down debt

Government debt? That one is not clear at all. My best guess is that it depends on lots and lots of factors.

> to reduce future inflation

There exist an optimum number for inflation, making it smaller makes things worse. That optimum number is almost always greater than 0.

> or give rebates

Nope, sending the money to poor people is almost always better than rebating taxes. (Taxes rebates is such a weird concept. I've only seen it discussed on the US.)

> to boost growth somewhere

Giving money to poor people is a very effective way of achieving this.

Fun fact: There's actually no solid evidence that 2% is a good level of inflation. For all we know it could be 1%, 0% or -2%.

The computer industry has been experiencing inflation around -10% a year for almost a century - if deflation is such a problem, why hasn't it collapsed yet? People delay their purchases a few months, but the foreseen total dry-up of purchasing just doesn't happen. (Except to Osbourne.)

possibly, and this is just a thought, because it is an industry, and not a country?

90s-10s Japan seems a better comparand, and it does not support your thesis

For many computation tasks, the purchaser needs the answer today, not the same calculation done twice as fast for hardware bought 2 years from now.

For very computationally expensive tasks such as the Human Genome Project, the answer is indeed to delay the purchase.

(This does lead to an interesting question on the motivations for current very-expensive AI training runs today: for those who are convinced their specific model is the future, is there really so much value in doing this right now rather than next year at 70% of the price?)

And the same is true for most things in the world. If money deflated, people wouldn't put off purchases forever, especially because it wouldn't be expected to keep deflating.

Really what keeps economies stable is broken expectations. If a certain thing was guaranteed to make lots of money, everyone would do it, but because no financial games can be reliably expected to, even the things that are making lots of money at a certain point in time (like holding it during deflation) they don't.

My knowledge on this area is limited (AKA, I've read few scientific articles and meta studies), but AFAIK (and heavily paraphrasing) you will get more economic growth the smaller the entities you hire, and more poverty reducing the poorer are the richest people getting the money. (Co-ops and unions do not automatically lead to poor, but avoiding having a well-paid CEO on the way has huge practical effects.)

Nothing beats giving money to poor people in either account, except if the results of the labor are useful.

A jobs program isn't necessarily useless jobs. We don't need to pay people to dig a ditch and then other people to fill in that ditch.

As a society, we get caught up in the idea that a government service has to turn a profit. This comes up all the time with the post office and public transit. Yet we never have a conversation about the fire department or roads turning a profit.

These services don't lose money. They simply have a cost.

We could have people who fix the roads, for example. Directly. Not by having a tender process that allows some extra entity to leech government funds, not do the thing and cost 5 times as much as they promised.

We could be training people in mental health services and providing those services so we can appropriate deal with the problems of homelless people and people having mental health episodes, rather than calling the police, which tends to be about the worst thing you can do as the person having an episode has a pretty high chance of simply ending up dead. Best case scenario, they end up in the criminal "justice" system.

> In the 1990s we managed to get enough evidence that useless jobs are worse than just giving people money in every way.

For my own edification, what was the evidence? I was a toddler then.

Jobs is pretty much right on there. Engineers are denigrated as "product guys" or at US auto companies, "car guys."

As if that's a bad thing.

(btw, it's Xerox, not "Zerox.")

If you took the same amount of money spent on the starliner and spent it on education poverty eradication and just universal income Americans would howl.

The total cost of starliner (including Boeing’s expenses beyond what NASA paid for) is about $6.7 billion over the last decade.

We spend several hundred times that on education and poverty eradication every year.

Umm, spread across 330 million people in the US, $6.7 billion would amount to a UBI of about $20 per person. Since it's over a decade or so, that's about $2/year/person.

Hardly enough to even be noticed, let alone make a difference.

The argument "we shouldn't go to space before solving problems on earth" is only for people who don't understand actual large (budget) numbers. Unfortunately, this is a very large portion of the voting population. (And, no I am not saying that $6.7B would be meaningless if deployed in a program with effective results leverage, or that we shouldn't demand better than Boeing is giving us, only that the soace & science initiatives are worth far more than the single-digit-percentages allocated to them)

Reply on news.ycombinator.com