Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

The U.S.'s $13B aircraft carrier has a toilet problem

popularmechanics.com

Post reply on HN
Print view View original

81–90 of 98 posts

Re: The U.S.'s $13B aircraft carrier has a toilet problem

#81

Post by dreamcompiler »

hwillis wrote: post #78
dreamcompiler wrote: post #63

Earlier quoted context omitted.

Engineering in the traditional sense plays very little role in modern defense systems. The overriding concern is profit: How can we extract the most money from the DoD? Good engineering that avoids recurring costs is discouraged in favor of gee-whiz technology that's easy to sell but expensive to build and maintain. The icing on the cake is that bids no longer matter; the largest contractors always underbid to get th…

> Engineering in the traditional sense plays very little role in modern defense systems. The overriding concern is profit: How can we extract the most money from the DoD? Good engineering that avoids recurring costs is discouraged in favor of gee-whiz technology that's easy to sell but expensive to build and maintain. The icing on the cake is that bids no longer matter; the largest contractors always underbid to get…

I have tried many times to break into government contracting and always failed. It's always one of two problems: The award is too small to cover the cost of the proposal and reporting paperwork (e.g. SBIR) or the award is huge and you need a full-time team whose sole job is to navigate all the red tape. And like you said, the people evaluating proposals often have no background in the field, so proxies for competence are used like slickness of marketing materials, whom you know in the organization, or how good-looking you are (not kidding).

After I graduated I thought I wanted to work for NASA. Then I found out most NASA employees spend most of their day not inventing or building things, but managing contractors who did the actual work.

I guess I'm saying I like your suggestion better than mine.

Re: The U.S.'s $13B aircraft carrier has a toilet problem

#82

Post by didericis »

hwillis wrote: post #78
dreamcompiler wrote: post #63

Earlier quoted context omitted.

Engineering in the traditional sense plays very little role in modern defense systems. The overriding concern is profit: How can we extract the most money from the DoD? Good engineering that avoids recurring costs is discouraged in favor of gee-whiz technology that's easy to sell but expensive to build and maintain. The icing on the cake is that bids no longer matter; the largest contractors always underbid to get th…

> Engineering in the traditional sense plays very little role in modern defense systems. The overriding concern is profit: How can we extract the most money from the DoD? Good engineering that avoids recurring costs is discouraged in favor of gee-whiz technology that's easy to sell but expensive to build and maintain. The icing on the cake is that bids no longer matter; the largest contractors always underbid to get…

> At its best, the government does it better, cheaper, and longer-lasting than anyone else

Can you cite examples and explain factors that you think lead to the government working at its best?

As much as I like free market competition, I think you’re probably right, at least when it comes to projects/pieces of projects where the only customer is the government. But when there’s a problem not well suited for competition to solve efficiently, that doesn’t guarantee a good government result: there’s a ton of variability in the quality of government responses.

My basic theory is this: the success of large government engineering projects depends primarily on whether or not everyone involved in the project believe it to be of existential importance (military projects during the cold war), whether its something that they will personally use and want to feel pride in (golden gate bridge), or whether or not the problem has some novel/interesting element that attracts the best and brightest.

Re: The U.S.'s $13B aircraft carrier has a toilet problem

#83

Post by hwillis »

didericis wrote: post #82
hwillis wrote: post #78

Earlier quoted context omitted.

> Engineering in the traditional sense plays very little role in modern defense systems. The overriding concern is profit: How can we extract the most money from the DoD? Good engineering that avoids recurring costs is discouraged in favor of gee-whiz technology that's easy to sell but expensive to build and maintain. The icing on the cake is that bids no longer matter; the largest contractors always underbid to get…

> At its best, the government does it better, cheaper, and longer-lasting than anyone else Can you cite examples and explain factors that you think lead to the government working at its best? As much as I like free market competition, I think you’re probably right, at least when it comes to projects/pieces of projects where the only customer is the government. But when there’s a problem not well suited for competitio…

USPS delivers packages much faster, more efficiently (fewer stops), significantly more cheaply, and with better employee pay and benefits. Medicare negotiates far more effectively than private insurance (although obviously that isn't a simple comparison) and has dramatically lower overheads. Government and even locally-owned infrastructure is cheaper than commercial infrastructure like roads or communications. Government-owned health services are cheaper and higher quality. Government engineering projects are superior. The GAO and IRS are extremely effective financial institutions.

> But when there’s a problem not well suited for competition to solve efficiently, that doesn’t guarantee a good government result: there’s a ton of variability in the quality of government responses.

Can you find me an example? By and large, the idea that the government is inefficient is a myth. There are boondoggles like Star Wars, but those are failures independent of government implementation. There are "failures" like Amtrak that are unprofitable, but they are kept running to offer services at a given price point because we want that regardless of what the market says.

Government projects are generally subject to the exact same motivations as private projects. They make profits. Employees can be fired and promoted. Innovation is rewarded, at the individual level. Underperforming projects are cut, often much more aggressively than in the free market.

Much more important than a sense of patriotism is the fact that governments are subject to intense internal and external scrutiny. Literally the most powerful people in the country are constantly looking, specifically to cut individual programs. There is no safe position in the government and institutional rot is constantly being gutted even when it's unwise, eg CDC, EPA, IRS lately. You will not find that pressure at IBM.

Re: The U.S.'s $13B aircraft carrier has a toilet problem

#84

Post by Analemma_ »

ceejayoz wrote: post #73
burlesona wrote: post #72

Earlier quoted context omitted.

It’s possible to put performance guarantees / insurance etc. into contracts such that if the contractor cannot build the entire thing within some percentage of budget then the entire cost is refunded, and the contractor is left with scrap metal.

That sounds like a great way to make sure no company ever bids to build an aircraft carrier.

Then it's a wide open field for a company who can do it to get the contract at whatever price they want. See how markets work?

Re: The U.S.'s $13B aircraft carrier has a toilet problem

#85

Post by sillywalk »

I'd have to say my favorite problem with the ship is that it has to stop flight operations if the new arresting gear needs repairing, and if one catapult needs repairs, it cannot be independently shut-down, all four need to be turned off, which takes ~90 minutes.

Re: The U.S.'s $13B aircraft carrier has a toilet problem

#87

Post by xyzzyz »

hwillis wrote: post #83
didericis wrote: post #82

Earlier quoted context omitted.

> At its best, the government does it better, cheaper, and longer-lasting than anyone else Can you cite examples and explain factors that you think lead to the government working at its best? As much as I like free market competition, I think you’re probably right, at least when it comes to projects/pieces of projects where the only customer is the government. But when there’s a problem not well suited for competitio…

USPS delivers packages much faster, more efficiently (fewer stops), significantly more cheaply, and with better employee pay and benefits. Medicare negotiates far more effectively than private insurance (although obviously that isn't a simple comparison) and has dramatically lower overheads. Government and even locally-owned infrastructure is cheaper than commercial infrastructure like roads or communications. Govern…

> USPS delivers packages much faster, more efficiently (fewer stops), significantly more cheaply, and with better employee pay and benefits.

USPS is fast because its mail is hauled by Fedex. Fedex is also significantly cheaper than USPS at delivering packages if you have a large contract with them. USPS is sometimes cheaper for retail customers, though, and it's also cheaper for standard mail.

Re: The U.S.'s $13B aircraft carrier has a toilet problem

#88

Post by closeparen »

hwillis wrote: post #83
didericis wrote: post #82

Earlier quoted context omitted.

> At its best, the government does it better, cheaper, and longer-lasting than anyone else Can you cite examples and explain factors that you think lead to the government working at its best? As much as I like free market competition, I think you’re probably right, at least when it comes to projects/pieces of projects where the only customer is the government. But when there’s a problem not well suited for competitio…

USPS delivers packages much faster, more efficiently (fewer stops), significantly more cheaply, and with better employee pay and benefits. Medicare negotiates far more effectively than private insurance (although obviously that isn't a simple comparison) and has dramatically lower overheads. Government and even locally-owned infrastructure is cheaper than commercial infrastructure like roads or communications. Govern…

But because of all that scrutiny, the government often has to behave suboptimally:

a) Processes can be a matter of law, requiring an extraordinarily heavy bureaucratic lift to adapt to new information or fix bugs.

b) Processes are calibrated to minimize trust. Contracting must be as obtuse as it is, because the alternative is to let civil servants make judgement calls, which the public cannot abide. This probably mitigates some serious downside risk, but it also precludes a lot of potential upside. Other institutions succeed when competent and trustworthy people just do what they think is right. (I daresay this is the only way an organization ever accomplishes anything difficult). Government agencies are only allowed to do that in emergencies, or in tiny edge cases like special forces.

c) Where leaders do have latitude, the fear of embarrassment weighs unusually heavy on their decision-making. It's harder to take good risks. I'm sure there was waste at IBM that would not survive in government. I'm equally sure there was innovation at IBM that could not have happened in government.

In some ways "government can't do anything right" is a self-fulling prophecy. The controls we implement in response to that perception make it come true. More fundamentally, I think any engineer can identify with how having to balance so many stakeholders could be incompatible with doing their best work. Making government less accountable probably isn't the answer either.

Re: The U.S.'s $13B aircraft carrier has a toilet problem

#89

Post by didericis »

hwillis wrote: post #83
didericis wrote: post #82

Earlier quoted context omitted.

> At its best, the government does it better, cheaper, and longer-lasting than anyone else Can you cite examples and explain factors that you think lead to the government working at its best? As much as I like free market competition, I think you’re probably right, at least when it comes to projects/pieces of projects where the only customer is the government. But when there’s a problem not well suited for competitio…

USPS delivers packages much faster, more efficiently (fewer stops), significantly more cheaply, and with better employee pay and benefits. Medicare negotiates far more effectively than private insurance (although obviously that isn't a simple comparison) and has dramatically lower overheads. Government and even locally-owned infrastructure is cheaper than commercial infrastructure like roads or communications. Govern…

Institutional rot is extremely expensive, and large institutions eventually wither away because of it. That pressure exists whether or not companies are aware of it/proactive about cutting it out.

I think the problems with government controlled projects are better generalized as problems with incentives and accountability in large organizations. It’s possible to get it right, but the bigger an organization gets, and the less existential pressure it feels/the more it can let problems fester without consequence, the more problems will occur.

The most stereotypical example is the soviet union. There was no accountability, and production and quality problems were endemic to pretty much every sector under government control. Cars, food, housing, clothing, energy, etc, were all behind the west, and became increasingly behind the longer things went on.

The soviet union was a particularly horrible government, but the US government also has plenty of inefficiencies. The SLS project has been dragging for eons, while SpaceX has been accelerating rapidly. Virtually no one pays taxes directly through government interfaces; they all use private services like TurboTax because they’re much simpler to use and more user friendly that what the government provides. Publicly funded job training programs have abysmal success rates, whereas private apprenticeships a schools that earn by pairing well trained students with their first job do much better.

Once you become a mega corporation, I think accountability and incentive problems start to kick in, just like in many governments. If they also lack the internal scrutiny that you describe within the US government, they will do worse (assuming they don’t have as many differing political objectives as a government).

Because there are also examples of huge government funded successes, like the Apollo program, the Golden Gate Bridge, the internet, etc, there is some sort of difference between the government projects that work and the government projects that don’t. I think its important to explicitly determine what that difference is, rather than reduce it to a binary opinion on government projects.

Re: The U.S.'s $13B aircraft carrier has a toilet problem

#90

Post by downrightmike »

unixfg wrote: post #2

Wait, what kinda of acid and how much? $400,000 seems like a lot.

Odds are that a good part of that expense is just storage for the massive amounts of the product. x100 of course. But when you need it, you'd better be able to get it and that's what's more than likely being paid for.
Post reply on HN