7 ms·

Given the size and complexity of such a device, and the limited number of them that are constructed (which means they are always a one-off), it would be a mirac

Given the size and complexity of such a device, and the limited number of them that are constructed (which means they are always a one-off), it would be a miracle on the order of Moses parting the Red Sea for one of them to be built __without__ such mistakes happening. In you are a developer, think of the number of times you have seen a successful SAP deployment i.e. on time, on budget, meets specs. Then multiply that by about 10k to reach the complexity of an aircraft carrier.

hwillis 7y ago
Engineering has a much stronger and more specific focus on avoiding "unexpected" problems like this. At its most general you use things like dimensional analysis to identify and avoid issues that you can't specifically foresee. This is, bare minimum, an embarrassing failure of design. No system, especially on a military ship, should have no redundancy. A single point of failure should not bring down every toilet on the ship. It's a damn aircraft carrier- splurge on separate sewage in the aft and fore.
- dreamcompiler 7y ago
  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 the contract then send bills that are 2x, 4x or 10x the bid, knowing they have the DoD over a barrel. The solution for this nonsense is competition. Real, honest, open, transparent competition. Combined with laws that dictate refusal to pay in excess of the bid, instant scrapping of programs that exceed cost estimates, and a 10-year timeout window between when people in DoD leadership can transition out and get jobs with DoD contractors.
  - ceejayoz 7y ago
    > instant scrapping of programs that exceed cost estimates So you're gonna build 75% of an aircraft carrier, and... then what?
    - jakear 7y ago
      Realize you’ve screwed your self over for underbidding and not do it again?
      - ceejayoz 7y ago
        Given the timeframes on a project like an aircraft carrier, both sides are equally screwed here.
        jakear 7y ago
        Better to screw both sides than only screw the military/taxpayers as is.
    - burlesona 7y ago
      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.
      - ceejayoz 7y ago
        That sounds like a great way to make sure no company ever bids to build an aircraft carrier.
        Analemma_ 7y ago
        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?
  - TazeTSchnitzel 7y ago
    How often does a “competitive” bidding process for being handed a temporary monopoly work out? The competition flies out the window the moment the contract is signed. What if the Navy builds the ships itself?
  - hwillis 7y ago
    > 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 the contract then send bills that are 2x, 4x or 10x the bid, knowing they have the DoD over a barrel. As a government contractor, I agree and fully endorse this. > The solution for this nonsense is competition. Real, honest, open, transparent competition. Combined with laws that dictate refusal to pay in excess of the bid, instant scrapping of programs that exceed cost estimates, and a 10-year timeout window between when people in DoD leadership can transition out and get jobs with DoD contractors. ...And I disagree with this fully. The solution is for the government to return to in-housing, relax the rules they hold themselves to, and aggressively expand their own capabilities. The things the government makes are far and away superior to anything in the private sector. At its best, the government does it better, cheaper, and longer-lasting than anyone else. Real competition for the things the government needs done is just not possible. Roughly 50% of contracts are based on just flat-out lies, and it's even worse for shitty contracts like SBIRs. I'll tell you, I have seen some real dumb shit. The BUYER needs technical expertise in order to not get ripped off, and if you already have expertise it makes much more sense to go the extra miles and build it yourself. The market for government contracts would need to be way more lucrative to allow companies to fail honestly, and even then, why would they when lying is more profitable? Plus, this doesn't even cover the worst part of large contracts, which is that they're effectively negotiated by politics, either congress itself or inter-DoD factionalism. That, more than anything else, creates large-scale rot.
    - dreamcompiler 7y ago
      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.
- grandinj 7y ago
  Any very large project is expected to have a number of embarrassing failures. You just don't hear about most of them. Not surprisingly, on a __military__ ship, sewage is not considered mission critical and required to have multiple redundancy. You can, after all, take a poop in a bucket and throw it over the side. Redundancy engineering is reserved for truly mission-critical things, like being able to find and shoot the enemy, and surviving a certain amount of damage. You don't waste valuable redundancy resources on things like this.
  - throwanem 7y ago
    It's a shame they also haven't wasted redundancy resources on the electromagnetic catapult system used to launch aircraft, and which as of 2018 had a failure rate consistent with a 70% chance of completing a full day of flight operations without downtime. (1) It's easy to imagine that the toilet failures here under discussion represent an isolated concern and a teething problem for the kind of new, nonessential system that you expect to see in a new class of capital warship which, after all, still gets the fundamentals right. The trouble is that, once you start really looking, that's just not borne out by the evidence. A lot of the ideas in play with these ships, as with the LCS or the F-35, are fundamentally good. But the US military procurement process has become so deeply corrupt that we as a country are no longer capable of implementing these good ideas well. The multiple-decade, trillion-dollar boondoggle has become table stakes, and that doesn't deserve to be excused by anyone. (Also, a Ford-class carrier has a crew of something like four thousand. They're gonna need a lot of buckets...) (1) https://breakingdefense.com/2018/06/navys-troubled-ford-carrier-makes-modest-progress/ https://breakingdefense.com/2018/06/navys-troubled-ford-carr...
    - hwillis 7y ago
      > It's a shame they also haven't wasted redundancy resources on the electromagnetic catapult system used to launch aircraft, and which as of 2018 had a failure rate consistent with a 70% chance of completing a full day of flight operations without downtime. (1) Note that this system is a privately-contracted one and was supposed to work as delivered.
bane 7y ago
I mean it's not like humans have been putting toilets in ships for a while and have past metrics they can use to predict things like pipe diameter or anything.

hwillis 7y ago

Engineering has a much stronger and more specific focus on avoiding "unexpected" problems like this. At its most general you use things like dimensional analysis to identify and avoid issues that you can't specifically foresee. This is, bare minimum, an embarrassing failure of design. No system, especially on a military ship, should have no redundancy. A single point of failure should not bring down every toilet on the ship. It's a damn aircraft carrier- splurge on separate sewage in the aft and fore.

dreamcompiler 7y ago
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 the contract then send bills that are 2x, 4x or 10x the bid, knowing they have the DoD over a barrel. The solution for this nonsense is competition. Real, honest, open, transparent competition. Combined with laws that dictate refusal to pay in excess of the bid, instant scrapping of programs that exceed cost estimates, and a 10-year timeout window between when people in DoD leadership can transition out and get jobs with DoD contractors.
- ceejayoz 7y ago
  > instant scrapping of programs that exceed cost estimates So you're gonna build 75% of an aircraft carrier, and... then what?
  - jakear 7y ago
    Realize you’ve screwed your self over for underbidding and not do it again?
    - ceejayoz 7y ago
      Given the timeframes on a project like an aircraft carrier, both sides are equally screwed here.
      - jakear 7y ago
        Better to screw both sides than only screw the military/taxpayers as is.
  - burlesona 7y ago
    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.
    - ceejayoz 7y ago
      That sounds like a great way to make sure no company ever bids to build an aircraft carrier.
      - Analemma_ 7y ago
        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?
- TazeTSchnitzel 7y ago
  How often does a “competitive” bidding process for being handed a temporary monopoly work out? The competition flies out the window the moment the contract is signed. What if the Navy builds the ships itself?
- hwillis 7y ago
  > 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 the contract then send bills that are 2x, 4x or 10x the bid, knowing they have the DoD over a barrel. As a government contractor, I agree and fully endorse this. > The solution for this nonsense is competition. Real, honest, open, transparent competition. Combined with laws that dictate refusal to pay in excess of the bid, instant scrapping of programs that exceed cost estimates, and a 10-year timeout window between when people in DoD leadership can transition out and get jobs with DoD contractors. ...And I disagree with this fully. The solution is for the government to return to in-housing, relax the rules they hold themselves to, and aggressively expand their own capabilities. The things the government makes are far and away superior to anything in the private sector. At its best, the government does it better, cheaper, and longer-lasting than anyone else. Real competition for the things the government needs done is just not possible. Roughly 50% of contracts are based on just flat-out lies, and it's even worse for shitty contracts like SBIRs. I'll tell you, I have seen some real dumb shit. The BUYER needs technical expertise in order to not get ripped off, and if you already have expertise it makes much more sense to go the extra miles and build it yourself. The market for government contracts would need to be way more lucrative to allow companies to fail honestly, and even then, why would they when lying is more profitable? Plus, this doesn't even cover the worst part of large contracts, which is that they're effectively negotiated by politics, either congress itself or inter-DoD factionalism. That, more than anything else, creates large-scale rot.
  - dreamcompiler 7y ago
    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.
grandinj 7y ago
Any very large project is expected to have a number of embarrassing failures. You just don't hear about most of them. Not surprisingly, on a __military__ ship, sewage is not considered mission critical and required to have multiple redundancy. You can, after all, take a poop in a bucket and throw it over the side. Redundancy engineering is reserved for truly mission-critical things, like being able to find and shoot the enemy, and surviving a certain amount of damage. You don't waste valuable redundancy resources on things like this.
- throwanem 7y ago
  It's a shame they also haven't wasted redundancy resources on the electromagnetic catapult system used to launch aircraft, and which as of 2018 had a failure rate consistent with a 70% chance of completing a full day of flight operations without downtime. (1) It's easy to imagine that the toilet failures here under discussion represent an isolated concern and a teething problem for the kind of new, nonessential system that you expect to see in a new class of capital warship which, after all, still gets the fundamentals right. The trouble is that, once you start really looking, that's just not borne out by the evidence. A lot of the ideas in play with these ships, as with the LCS or the F-35, are fundamentally good. But the US military procurement process has become so deeply corrupt that we as a country are no longer capable of implementing these good ideas well. The multiple-decade, trillion-dollar boondoggle has become table stakes, and that doesn't deserve to be excused by anyone. (Also, a Ford-class carrier has a crew of something like four thousand. They're gonna need a lot of buckets...) (1) https://breakingdefense.com/2018/06/navys-troubled-ford-carrier-makes-modest-progress/ https://breakingdefense.com/2018/06/navys-troubled-ford-carr...
  - hwillis 7y ago
    > It's a shame they also haven't wasted redundancy resources on the electromagnetic catapult system used to launch aircraft, and which as of 2018 had a failure rate consistent with a 70% chance of completing a full day of flight operations without downtime. (1) Note that this system is a privately-contracted one and was supposed to work as delivered.

bane 7y ago

I mean it's not like humans have been putting toilets in ships for a while and have past metrics they can use to predict things like pipe diameter or anything.