9 ms·

Surely at some stage it's better to write off this mess and start again from scratch? It really seems like they're throwing good money after bad. It's a nice

by Twirrim 12y ago

Surely at some stage it's better to write off this mess and start again from scratch? It really seems like they're throwing good money after bad. It's a nice idea to have a completely flexible plane capable of doing every role, but it's clearly not working. What we're ending up with is the proverbial camel designed by a committee who was designing a horse.

mercurial 12y ago
It's guaranteed subsidies to the military-industrial complex for decades, not to mention employment. Why would you stop that?
- mkr-hn 12y ago
  We could spend the money employing people to do useful things like repairing our failing infrastructure.
  - jiggy2011 12y ago
    Not the same people.
    - ihsw 12y ago
      So?
      - jiggy2011 12y ago
        They won't vote for you if you take their jobs, especially if they are in a niche industry.
        CamperBob2 12y ago
        So give them different jobs.
        engi_nerd 12y ago
        I am one of those people. I've worked in the aircraft industry for about 5 years total of my career, and with rockets for another 2 years or so. I have two engineering degrees, neither one of which are directly related to the kind of engineering I actually get paid to do. And I'm not really exceptional in regards to being flexible. The lead electrical designer for my department has a background in mechanical engineering. Our technical lead was a test engineer on the Delta IV program for years. It's common for maintenance leads to also have business degrees, or for an engineer to have a side job doing app development. Program goes away? We'll find other engineering work somehow.
  - Theodores 12y ago
    Can't we just go back to building pyramids? There must be some scheme that siphons off as much excess capital as possible so we can happily remain serfs and doesn't provoke arguments. The Egyptians managed to build pyramids for many, many, many decades. They are tried and tested unlike these new fangled war planes things.
- erglkjahlkh 12y ago
  This. They had well defined, implemented, and tested technologies to support creating extremely bug free and working software. For some reason they decided to abandon all that previous work. I can not believe they were that inept, they must have done that on purpose.
  - gaius 12y ago
    Indeed. The abandoning of Ada makes no sense, if your goal was to ship extremely reliable software...
    - tsotha 12y ago
      They abandoned Ada because they couldn't find people with any experience in the language. That's a big problem with the cyclical nature of military funding - when you don't build new things for a decade you start to lose the capability.
      - gaius 12y ago
        You can find people with COBOL knowledge, if you need them and can pay them - or can train them. If you are planning multi-decade projects like major defence procurements are these days, then training people up-front would be worth it. Incidentally, there are possibly millions of people with daily exposure to a subset of Ada - its the basis of Oracle's PL/SQL. A component of the PL/SQL parser is called DIANA, which true Oracle geeks know is Descriptive Intermediate Attributed Notation for Ada. See http://stackoverflow.com/q/7764656/447514 http://stackoverflow.com/q/7764656/447514
        tsotha 12y ago
        Of course they considered training a new generation of Ada programmers. In the end they decided other options were better. I think they encountered resistance to the idea from software developers - nobody wants to be the employable-only-on-defense-contracts Ada guy during the next round of budget cuts. There's no guarantee the software would have been any better were it written in Ada. Every generation is more complex than the last.
- fnordfnordfnord 12y ago
  Could be even worse if the thing worked. They might get the idea to go start a war somewhere.
bborud 12y ago
The tragedy is that they have known this approach doesn't work for at least as long as I have been alive. And I'm old enough to never have been the target demographic of garments designed to showcase my underwear.
smacktoward 12y ago
The problem is that defense procurement is kind of the polar opposite of "agile development." It takes 20+ years to get a new military aircraft design from the drawing board to active service. (The programs that eventually became the F-35 both started in the early '90s.) The F-35 has a lot of problems, but it's already made its way through most of that loooong process. Scratching it and going back to the drawing board would reset that clock, which means that even if 1) they started on the new design tomorrow and 2) it turned out much better than the F-35, it wouldn't start reaching squadrons until 2035 or even later. Which leaves a big, multi-decade gap where the various services have to keep their aging fighters flying until the New Thing shows up to replace them.
- engi_nerd 12y ago
  It's hard to even quantify what kind of development it is. The official chart that diagrams the steps in the process [1] is poster sized and is absolutely mind boggling. Even after multiple (mandatory!) courses explaining it to me I would find it impossible to explain to my 65 year old mother, which to me means that I do not really understand it. Probably no one really understands it as written except the instructors that the DoD hires to teach it to people. [1] http://www.wired.com/images_blogs/dangerroom/2010/09/atl_wall_chart.jpg http://www.wired.com/images_blogs/dangerroom/2010/09/atl_wal...
- GVIrish 12y ago
  > The F-35 has a lot of problems, but it's already made its way through most of that loooong process. Scratching it and going back to the drawing board would reset that clock, which means that even if 1) they started on the new design tomorrow and 2) it turned out much better than the F-35, it wouldn't start reaching squadrons until 2035 or even later. Which leaves a big, multi-decade gap where the various services have to keep their aging fighters flying until the New Thing shows up to replace them. The reason this program has taken so long is because its goal is to build an advanced, cutting-edge fighter that eclipses any of the other 4th gen aircraft in the world while simultaneously meeting strongly conflicting requirements. If instead the military developed 2 variants (carrier-borne and land-based), scrapped the STOVL requirement, and perhaps scrapped the low-observable requirement, they could end up with a far cheaper, far better performing aircraft in far less time. For example, the F/A-18 only took about 10 years to go from development to production. It would still leave a gap of about 10 years but that's better than spending the most money in military history and being stuck with a turkey for 3 or 4 decades.
tsotha 12y ago
Why would you think the next program would be any better? These are the normal teething problems every new advanced fighter goes through. The F-22 went through them. The F-16 and F-15 went through them, though it was so long ago nobody remembers. Nearly every project reaches a point where people say "This will never work. Let's cancel it." But there's no plan B. We can't fly the F-16 for another 25 years. In any event, the gun doesn't matter that much. Gun kills are very rare, and usually they happen because the pilot wants a gun kill. They probably wouldn't even have included the gun but for some bad experiences in the '60s with the F-4. By the way, the "flexible plane idea" has been incredibly successful in the F-18.
- GVIrish 12y ago
  > Why would you think the next program would be any better? Because if the powers that be were smart, they drastically simplify the requirements and drop the USMC variant completely. Any time the military tries to design a super-advanced swiss-army knife it invariably fails. This is the worst example of that in history. > These are the normal teething problems every new advanced fighter goes through. The F-22 went through them. The F-16 and F-15 went through them, though it was so long ago nobody remembers. Nearly every project reaches a point where people say "This will never work. Let's cancel it." Sure but the problems with the F-35 are waaayyy beyond teething issues. The software is way behind schedule. The fuselage has very poor rear visibility, so a targeting helmet was supposed to be developed but that's way behind schedule and may never work. The plane was designed without proper onboard lightning strike protection so it currently can't fly in bad weather. The plane can't be loaded with fuel higher than 80 degrees in temperature so it can't currently operate in a hot environment until retrofits to cooling trucks. The jet wash from the USMC variant melts the decks of the amphibious carriers it is meant to operate from. It has a poor turning and acceleration performance compared to Gen 4 fighter aircraft. It has poor loiter time. It doesn't have the basic video uplink functionality for close air support that even the A-10 has. It is already so far over budget that the unit price may approach that of the F-22. Some of the partner nations that were supposed to buy it are backing out, thereby increasing the unit price. The main problem with the F-35 is that the military took a lot of short cuts in drafting the requirements for this program, and when the predictable problems and schedule slippage occurred, they doubled down by relaxing testing requirements. This is supposed to be the backbone of American air power for DECADES. It would be one thing if this system would eventually perform as billed but it looks unlikely that it will and it will cost twice as much to purchase and operate as originally projected. It doesn't make sense to cripple our air power for the next generation becaus of the sunk cost fallacy. > By the way, the "flexible plane idea" has been incredibly successful in the F-18. Yes, but the F-18 is a much, much simpler concept. Carrier-based strike aircraft. Then look at the F-35: stealthy land-based, carrier-based, and stovl aircraft that will perform strike, cas, and interceptor roles. Right off the top the stovl requirement makes it very difficult to design a stealthy aircraft that will also have good enough aerodynamic performance to win a an air-to-air engagement after ther merge. If the F-35 was only a carrier-based strike fighter that added stealthiness, maybe we'd have a winner. But it is literally trying to be all things to three different services (one with a VERY difficult implement STOVL requirement) then throwing a difficult to implement low-observable requirement of which no other nation in the world has successfully implemented even a first generation of.
hga 12y ago
And we don't learn from history. The same exact thing happened with the '60s TFX (https://en.wikipedia.org/wiki/General_Dynamics_F-111_Aardvark#Tactical_Fighter_Experimental_.28TFX.29 https://en.wikipedia.org/wiki/General_Dynamics_F-111_Aardvar...), except in planned roles it had air superiority and not V/STOL. It ended up with one model for deep interdiction, then the various services were able to get back to procuring more specialized planes like the F-14 and F-15. In the meanwhile the Air Force had to swallow its institutional pride and fly Navy designed F-4s.
genwin 12y ago
> It really seems like they're throwing good money after bad. When making rich people richer is the primary goal, it all makes sense.

by Twirrim 12y ago

mercurial 12y ago

It's guaranteed subsidies to the military-industrial complex for decades, not to mention employment. Why would you stop that?

mkr-hn 12y ago
We could spend the money employing people to do useful things like repairing our failing infrastructure.
- jiggy2011 12y ago
  Not the same people.
  - ihsw 12y ago
    So?
    - jiggy2011 12y ago
      They won't vote for you if you take their jobs, especially if they are in a niche industry.
      - CamperBob2 12y ago
        So give them different jobs.
        engi_nerd 12y ago
        I am one of those people. I've worked in the aircraft industry for about 5 years total of my career, and with rockets for another 2 years or so. I have two engineering degrees, neither one of which are directly related to the kind of engineering I actually get paid to do. And I'm not really exceptional in regards to being flexible. The lead electrical designer for my department has a background in mechanical engineering. Our technical lead was a test engineer on the Delta IV program for years. It's common for maintenance leads to also have business degrees, or for an engineer to have a side job doing app development. Program goes away? We'll find other engineering work somehow.
- Theodores 12y ago
  Can't we just go back to building pyramids? There must be some scheme that siphons off as much excess capital as possible so we can happily remain serfs and doesn't provoke arguments. The Egyptians managed to build pyramids for many, many, many decades. They are tried and tested unlike these new fangled war planes things.
erglkjahlkh 12y ago
This. They had well defined, implemented, and tested technologies to support creating extremely bug free and working software. For some reason they decided to abandon all that previous work. I can not believe they were that inept, they must have done that on purpose.
- gaius 12y ago
  Indeed. The abandoning of Ada makes no sense, if your goal was to ship extremely reliable software...
  - tsotha 12y ago
    They abandoned Ada because they couldn't find people with any experience in the language. That's a big problem with the cyclical nature of military funding - when you don't build new things for a decade you start to lose the capability.
    - gaius 12y ago
      You can find people with COBOL knowledge, if you need them and can pay them - or can train them. If you are planning multi-decade projects like major defence procurements are these days, then training people up-front would be worth it. Incidentally, there are possibly millions of people with daily exposure to a subset of Ada - its the basis of Oracle's PL/SQL. A component of the PL/SQL parser is called DIANA, which true Oracle geeks know is Descriptive Intermediate Attributed Notation for Ada. See http://stackoverflow.com/q/7764656/447514 http://stackoverflow.com/q/7764656/447514
      - tsotha 12y ago
        Of course they considered training a new generation of Ada programmers. In the end they decided other options were better. I think they encountered resistance to the idea from software developers - nobody wants to be the employable-only-on-defense-contracts Ada guy during the next round of budget cuts. There's no guarantee the software would have been any better were it written in Ada. Every generation is more complex than the last.
fnordfnordfnord 12y ago
Could be even worse if the thing worked. They might get the idea to go start a war somewhere.

bborud 12y ago

The tragedy is that they have known this approach doesn't work for at least as long as I have been alive. And I'm old enough to never have been the target demographic of garments designed to showcase my underwear.

smacktoward 12y ago

The problem is that defense procurement is kind of the polar opposite of "agile development." It takes 20+ years to get a new military aircraft design from the drawing board to active service. (The programs that eventually became the F-35 both started in the early '90s.) The F-35 has a lot of problems, but it's already made its way through most of that loooong process. Scratching it and going back to the drawing board would reset that clock, which means that even if 1) they started on the new design tomorrow and 2) it turned out much better than the F-35, it wouldn't start reaching squadrons until 2035 or even later. Which leaves a big, multi-decade gap where the various services have to keep their aging fighters flying until the New Thing shows up to replace them.

engi_nerd 12y ago
It's hard to even quantify what kind of development it is. The official chart that diagrams the steps in the process [1] is poster sized and is absolutely mind boggling. Even after multiple (mandatory!) courses explaining it to me I would find it impossible to explain to my 65 year old mother, which to me means that I do not really understand it. Probably no one really understands it as written except the instructors that the DoD hires to teach it to people. [1] http://www.wired.com/images_blogs/dangerroom/2010/09/atl_wall_chart.jpg http://www.wired.com/images_blogs/dangerroom/2010/09/atl_wal...
GVIrish 12y ago
> The F-35 has a lot of problems, but it's already made its way through most of that loooong process. Scratching it and going back to the drawing board would reset that clock, which means that even if 1) they started on the new design tomorrow and 2) it turned out much better than the F-35, it wouldn't start reaching squadrons until 2035 or even later. Which leaves a big, multi-decade gap where the various services have to keep their aging fighters flying until the New Thing shows up to replace them. The reason this program has taken so long is because its goal is to build an advanced, cutting-edge fighter that eclipses any of the other 4th gen aircraft in the world while simultaneously meeting strongly conflicting requirements. If instead the military developed 2 variants (carrier-borne and land-based), scrapped the STOVL requirement, and perhaps scrapped the low-observable requirement, they could end up with a far cheaper, far better performing aircraft in far less time. For example, the F/A-18 only took about 10 years to go from development to production. It would still leave a gap of about 10 years but that's better than spending the most money in military history and being stuck with a turkey for 3 or 4 decades.

tsotha 12y ago

Why would you think the next program would be any better? These are the normal teething problems every new advanced fighter goes through. The F-22 went through them. The F-16 and F-15 went through them, though it was so long ago nobody remembers. Nearly every project reaches a point where people say "This will never work. Let's cancel it." But there's no plan B. We can't fly the F-16 for another 25 years. In any event, the gun doesn't matter that much. Gun kills are very rare, and usually they happen because the pilot wants a gun kill. They probably wouldn't even have included the gun but for some bad experiences in the '60s with the F-4. By the way, the "flexible plane idea" has been incredibly successful in the F-18.

GVIrish 12y ago
> Why would you think the next program would be any better? Because if the powers that be were smart, they drastically simplify the requirements and drop the USMC variant completely. Any time the military tries to design a super-advanced swiss-army knife it invariably fails. This is the worst example of that in history. > These are the normal teething problems every new advanced fighter goes through. The F-22 went through them. The F-16 and F-15 went through them, though it was so long ago nobody remembers. Nearly every project reaches a point where people say "This will never work. Let's cancel it." Sure but the problems with the F-35 are waaayyy beyond teething issues. The software is way behind schedule. The fuselage has very poor rear visibility, so a targeting helmet was supposed to be developed but that's way behind schedule and may never work. The plane was designed without proper onboard lightning strike protection so it currently can't fly in bad weather. The plane can't be loaded with fuel higher than 80 degrees in temperature so it can't currently operate in a hot environment until retrofits to cooling trucks. The jet wash from the USMC variant melts the decks of the amphibious carriers it is meant to operate from. It has a poor turning and acceleration performance compared to Gen 4 fighter aircraft. It has poor loiter time. It doesn't have the basic video uplink functionality for close air support that even the A-10 has. It is already so far over budget that the unit price may approach that of the F-22. Some of the partner nations that were supposed to buy it are backing out, thereby increasing the unit price. The main problem with the F-35 is that the military took a lot of short cuts in drafting the requirements for this program, and when the predictable problems and schedule slippage occurred, they doubled down by relaxing testing requirements. This is supposed to be the backbone of American air power for DECADES. It would be one thing if this system would eventually perform as billed but it looks unlikely that it will and it will cost twice as much to purchase and operate as originally projected. It doesn't make sense to cripple our air power for the next generation becaus of the sunk cost fallacy. > By the way, the "flexible plane idea" has been incredibly successful in the F-18. Yes, but the F-18 is a much, much simpler concept. Carrier-based strike aircraft. Then look at the F-35: stealthy land-based, carrier-based, and stovl aircraft that will perform strike, cas, and interceptor roles. Right off the top the stovl requirement makes it very difficult to design a stealthy aircraft that will also have good enough aerodynamic performance to win a an air-to-air engagement after ther merge. If the F-35 was only a carrier-based strike fighter that added stealthiness, maybe we'd have a winner. But it is literally trying to be all things to three different services (one with a VERY difficult implement STOVL requirement) then throwing a difficult to implement low-observable requirement of which no other nation in the world has successfully implemented even a first generation of.

hga 12y ago

And we don't learn from history. The same exact thing happened with the '60s TFX (https://en.wikipedia.org/wiki/General_Dynamics_F-111_Aardvark#Tactical_Fighter_Experimental_.28TFX.29 https://en.wikipedia.org/wiki/General_Dynamics_F-111_Aardvar...), except in planned roles it had air superiority and not V/STOL. It ended up with one model for deep interdiction, then the various services were able to get back to procuring more specialized planes like the F-14 and F-15. In the meanwhile the Air Force had to swallow its institutional pride and fly Navy designed F-4s.

genwin 12y ago

> It really seems like they're throwing good money after bad. When making rich people richer is the primary goal, it all makes sense.