4 ms·

Before you tear down a fence, learn why it was built in the first place. The real question is, why were approvals processed in serial? It's very possible that t

Before you tear down a fence, learn why it was built in the first place. The real question is, why were approvals processed in serial? It's very possible that the input for some reviews requires the output from others, for example. Think of an analogy: If someone said 'we'll speed up X software by running it concurrently', you'd have a lot of questions before you believed it. I've seen this mistake before. For example, when the F-35 was being planned people had lots of similarly 'obvious' cost-saving and time-saving ideas: Instead of all these different services and countries buying different planes, why not all buy a common plane and get enormous economy of scale in production, maintenance, training, etc.? Instead of doing the waterfall method and locking in all the specs and developing all the tech before production, why not do it more iteratively, save time and get other benefits? We learned the answers: 1. It's hard enough to meet the demanding (existential) requirements and bleeding edge tech development needs of one service; building one solution to please 3 services and a dozen countries is impossible. 2. That works for software, not $100 million hardware. I'm not saying it's a bad idea; I'm saying that we don't know and that the reason for the fence is the interesting, valuable information.

throwaway84742 9y ago
I bet that if powers that be decided not to unify F35, we’d end up with 3 specialized stealth fighter projects costing $1T each, and without the ability to stockpile just one set of parts for sustained maintenance.
- philwelch 9y ago
  The Air Force and Navy both need (or want) a light, stealthy fighter-bomber. Historically, there's no good reason not to just make the Air Force use the same planes the Navy does. They just never actually agree on which plane they want and it's easier to let both services have their own way. The only tricky part is that the Marines need a STOVL close-air-support bomber to replace the Harrier, and the Royal Navy needs a STOVL fighter-bomber to replace the Harrier since they don't have full-sized carriers. There's some commonality between those requirements and the Navy/Air Force requirements--there's some rationale for stealthy CAS planes, and your STOVL plane needs to be a full-featured fighter-bomber for the Brits to want it anyway--but STOVL is a lot more complex than normal horizontal flight.
  - forapurpose 9y ago
    > there's no good reason not to just make the Air Force use the same planes the Navy does Perhaps the Navy plane is designed to fly from carriers; to operate in a wet, salty marine environment; and to operate where resources (fuel, parts, etc.) can be limited? Why would the Air Force want to make all those trade-offs in a $100 million plane? Remember that any trade-off will result in a certain number of additional dead pilots, failed missions, and lost expensive capital in a war. Perhaps the services need different ranges, have different refueling capabilities, have different functional needs (hunting subs, for example). They do different jobs, despite the fact that they both use small planes.
    - philwelch 9y ago
      Tons of air forces use the F/A-18 Hornet and Super Hornet--just not ours.
  - paulmd 9y ago
    > The Air Force and Navy both need (or want) a light, stealthy fighter-bomber. Historically, there's no good reason not to just make the Air Force use the same planes the Navy does. This is not actually true. The Navy needs a heavy chassis that can withstand repeated 4G+ carrier takeoffs and landings, the Air Force wants a lighter chassis that leaves more room for payload and allows more maneuverability. The Navy wants dual-engine configurations for reliability for over-water flight, the Air Force prefers single-engine for greater simplicity and weight efficiency. You can make something like the Tomcat or Hornet work, but it's not as agile as a dedicated interceptor, can't carry as much weight as a dedicated strike aircraft, and is heavier than a dedicated light-multirole aircraft. It's also more maintenance-intensive and has higher fuel consumption/shorter range than all of these. Money is no object for the US military, we get the best aircraft for the role, period. You can disagree about whether that should be the policy, but it is. The Marines' STOVL requirement is a third set of requirements that is totally disjoint to the other services, agreed. I don't know enough about it other than a general sense that it's possibly the complex task to engineer out of all of these. Airplanes go forward, not up and down. From a financial perspective, we might be better off just telling the Marines to fly helos instead and operate light strike aircraft off improved strips once we have a beachhead, but again, money is noobject, so they get a rube-goldberg airplane that can go up and down and land on their pipsqueak carriers.
- pg_bot 9y ago
  I think you've learned the wrong lesson from the F35's development. It likely would've been more cost efficient to build three different planes that are specialized to the needs of each branch of the military. One reason that the plane has been such a boondoggle (1.5 trillion in development cost) is that it is trying to be everything for everyone. Optimizing a fighter jet for replacement parts is monumentally stupid. It should be optimized for its ability to perform its job in the field and it turns out each branch needs a different type of plane. Stockpiling several sets of parts is of least concern and is unlikely to cost more money.
lopmotr 9y ago
Perhaps there were done sequentially so that if a project fails one stage or has to be changed to comply, all the other stages' work isn't wasted. Like a branch misprediction in a CPU. That might make the approval more expensive overall but having things finished faster is also an advantage to weigh against it.
- TheCoelacanth 9y ago
  But that would only work if you can arrange to have the more likely to fail stages earlier in the process. If the last of ten steps fails 5% of the way through, you have wasted 95% of the approval cost. If all the steps are done at the same time and one of the steps fails 5% of the way through you have only wasted 5%.

throwaway84742 9y ago

I bet that if powers that be decided not to unify F35, we’d end up with 3 specialized stealth fighter projects costing $1T each, and without the ability to stockpile just one set of parts for sustained maintenance.

philwelch 9y ago
The Air Force and Navy both need (or want) a light, stealthy fighter-bomber. Historically, there's no good reason not to just make the Air Force use the same planes the Navy does. They just never actually agree on which plane they want and it's easier to let both services have their own way. The only tricky part is that the Marines need a STOVL close-air-support bomber to replace the Harrier, and the Royal Navy needs a STOVL fighter-bomber to replace the Harrier since they don't have full-sized carriers. There's some commonality between those requirements and the Navy/Air Force requirements--there's some rationale for stealthy CAS planes, and your STOVL plane needs to be a full-featured fighter-bomber for the Brits to want it anyway--but STOVL is a lot more complex than normal horizontal flight.
- forapurpose 9y ago
  > there's no good reason not to just make the Air Force use the same planes the Navy does Perhaps the Navy plane is designed to fly from carriers; to operate in a wet, salty marine environment; and to operate where resources (fuel, parts, etc.) can be limited? Why would the Air Force want to make all those trade-offs in a $100 million plane? Remember that any trade-off will result in a certain number of additional dead pilots, failed missions, and lost expensive capital in a war. Perhaps the services need different ranges, have different refueling capabilities, have different functional needs (hunting subs, for example). They do different jobs, despite the fact that they both use small planes.
  - philwelch 9y ago
    Tons of air forces use the F/A-18 Hornet and Super Hornet--just not ours.
- paulmd 9y ago
  > The Air Force and Navy both need (or want) a light, stealthy fighter-bomber. Historically, there's no good reason not to just make the Air Force use the same planes the Navy does. This is not actually true. The Navy needs a heavy chassis that can withstand repeated 4G+ carrier takeoffs and landings, the Air Force wants a lighter chassis that leaves more room for payload and allows more maneuverability. The Navy wants dual-engine configurations for reliability for over-water flight, the Air Force prefers single-engine for greater simplicity and weight efficiency. You can make something like the Tomcat or Hornet work, but it's not as agile as a dedicated interceptor, can't carry as much weight as a dedicated strike aircraft, and is heavier than a dedicated light-multirole aircraft. It's also more maintenance-intensive and has higher fuel consumption/shorter range than all of these. Money is no object for the US military, we get the best aircraft for the role, period. You can disagree about whether that should be the policy, but it is. The Marines' STOVL requirement is a third set of requirements that is totally disjoint to the other services, agreed. I don't know enough about it other than a general sense that it's possibly the complex task to engineer out of all of these. Airplanes go forward, not up and down. From a financial perspective, we might be better off just telling the Marines to fly helos instead and operate light strike aircraft off improved strips once we have a beachhead, but again, money is noobject, so they get a rube-goldberg airplane that can go up and down and land on their pipsqueak carriers.
pg_bot 9y ago
I think you've learned the wrong lesson from the F35's development. It likely would've been more cost efficient to build three different planes that are specialized to the needs of each branch of the military. One reason that the plane has been such a boondoggle (1.5 trillion in development cost) is that it is trying to be everything for everyone. Optimizing a fighter jet for replacement parts is monumentally stupid. It should be optimized for its ability to perform its job in the field and it turns out each branch needs a different type of plane. Stockpiling several sets of parts is of least concern and is unlikely to cost more money.

lopmotr 9y ago

Perhaps there were done sequentially so that if a project fails one stage or has to be changed to comply, all the other stages' work isn't wasted. Like a branch misprediction in a CPU. That might make the approval more expensive overall but having things finished faster is also an advantage to weigh against it.

TheCoelacanth 9y ago
But that would only work if you can arrange to have the more likely to fail stages earlier in the process. If the last of ten steps fails 5% of the way through, you have wasted 95% of the approval cost. If all the steps are done at the same time and one of the steps fails 5% of the way through you have only wasted 5%.