6 ms·

If SpaceX is providing fixed pricing to the Air Force rather than cost plus pricing, their internal processes and methods are basically irrelevant. The only ti

If SpaceX is providing fixed pricing to the Air Force rather than cost plus pricing, their internal processes and methods are basically irrelevant. The only time that matters is when doing cost plus where management efficiency might have some kind of an impact on the price. The Air Force took a pretty silly position on the whole thing IMO.

foobarqux 12y ago
Certification isn't about cost, it's about safety, where internal processes and methods are paramount.
- bediger4000 12y ago
  You know, I wish I could believe that in the context of "aerospace" or "Defense Department". After working in aerospace in the mid-80s, I came to believe that certification was more about barrier-to-entry than anything else. Ever notice how the seat-belt buckles on passenger transport planes look like a 1948 design? They're certified, and it costs way too much to certify a new design or process. Sorry to be cynical, but I just don't believe the public justifications of "certified" any more.
  - foobarqux 12y ago
    First, I was correcting the putative point of the audit which the grandparent erroneously claimed was a cost audit. Second, I'm sure big corporations take advantage of certification process to keep smaller players out but certification is extremely important, suppliers would not follow many of the safety requirements otherwise.
  - rdtsc 12y ago
    > I came to believe that certification was more about barrier-to-entry than anything else Still the case with most government procurement processes. If you manage to jump through all the "certification" hoops, you can mark up your product to an insane rate and then lobby for certification to be expanded. Then the government has a list of "approved" vendors who are basically in the "certification" club and sometimes you can only find a product of a particular type only by a particular company and when you ask them for the price, your jaw drops (especially if there is a 20x cheaper COTS one on the market at the same time).
    - HeyLaughingBoy 12y ago
      You can add Medical to that list. Need a USB thumb drive for a Medical Device? That'll be 10x the cost of the exact same one you can buy on Amazon. OK, so maybe the vendor guarantees traceability. So now you just paid 10x the price to be safe in the knowledge that you and Amazon are buying from the same guy in a basement in Shenzen.
  - dingaling 12y ago
    > They're certified, and it costs way too much to certify a new design or process That buckle is a standard UI across every commercial aircraft in the World. That is very important in emergency situations. Billions of people know pull-to-release. There is really no advantage to changing the design unless we also change to three-point harnesses as in cars. But that won't happen for cost reasons ( ' safety is our number one priority, profit is our zeroth' ).
- wavefunction 12y ago
  Is that why the F-35 was cutting off oxygen to pilots? I'll believe your statement when I see it in action.
  - nickff 12y ago
    The F-22 (not the F-35) was the aircraft that was having pilot consciousness issues, and it has not been proven that the issue was actually oxygen being cut off from the pilot. Some have also posited that the problems were actually due to the high Gs at low pressure (as the F-22 is the most manueverable aircraft ever made at high altitude), which is also possible.
    - angersock 12y ago
      pilot consciousness issues What a marvelously tame way of describing such a thing.
- shabble 12y ago
  It's also about cost, in the form of reliability, since the payloads commonly cost > 1 billion USD each. Not to mention, if you botch the orbit they might have to throw missiles at it so it doesn't injure peopl^W^Wfall into the wrong hands[1] [1] https://en.wikipedia.org/wiki/USA-193#Destruction https://en.wikipedia.org/wiki/USA-193#Destruction
  - gaadd33 12y ago
    Aren't most commercial payloads insured by third party insurers? Why is the Air Force unable to do that?
    - cpwright 12y ago
      Maybe the Air Force doesn't want to provide the insurer with sufficient information about the payload to make a reasonable underwriting decision? Insurance also doesn't come free. The US government can also take on more risk than any particular insurer, so by purchasing those contracts would be in effect wasting taxpayer money. The consequences of something like a failed launch can be hard to quantify economically. There is clearly opportunity cost for missed intelligence not having the satellite in orbit when it is needed. What happens if the satellite crashes in an unfriendly place, losing the exclusivity to a bit of technology could be very damaging as well. I think you would find it hard to place a particular insurable value on those; but minimizing that risk obviously costs money. How much that risk minimization is worth is a question that I can't answer.
- humanrebar 12y ago
  > where internal processes and methods are paramount I don't buy that. Process is CYA first and foremost. If something happens, you get to say, "We adhere to the strictest processes and best practices... blah, blah, blah." Was there a lack of process in the space shuttle disasters? The flight 9525 crash? The Titanic? You want safety? Get a third party (insurer, certification body, etc.) to assume responsibility (financial, legal, political, etc.) if anything goes wrong, regardless of the root cause. You might not get perfect safety, but you'll get responsibility, which is the next best thing. Rigid processes are about diffusing responsibility, not taking it.
  - panzagl 12y ago
    Existence of process does not mean the process is correct. Process failure was a big part of the shuttle crashes.
    - humanrebar 12y ago
      By that definition, everything is a process mistake and the term has no meaning.
      - panzagl 12y ago
        You're right, everything should be done by engineering ninja rockstars, and process is just what unawesome people use to keep their jobs.
        pyre 12y ago
        Well, you're ignoring blind adherence to process. E.g. it doesn't matter how good an idea is in a 'traditional' Japanese company, if the person presenting it doesn't have seniority, it will be ignored.
    - LoSboccacc 12y ago
      wasn't it management meddling? process called for redesigning the challenger oring, as corrosion was triple than expected, and process called for inspecting the columbia tiles. you can find all of these details in the reports, and you can find the name of the managers overriding engineering processes in both instances. challanger: http://www.onlineethics.org/cms/12722.aspx http://www.onlineethics.org/cms/12722.aspx http://www.engineering.com/Library/ArticlesPage/tabid/85/ArticleID/170/The-Space-Shuttle-Challenger-Disaster.aspx http://www.engineering.com/Library/ArticlesPage/tabid/85/Art... columbia: http://www.aero-news.net/index.cfm?do=main.textpost&id=9fe7f63a-d780-403a-adcf-a51622a8a9cf http://www.aero-news.net/index.cfm?do=main.textpost&id=9fe7f... (notice that engineering requested hard photographic data to assess the damage, and her used gut feeling)
Shivetya 12y ago
I would tend to think that the officers in charge took the safe route, as in followed the established bureaucracy in place so as not to appear to be rocking the boat. Leaving established protocol is always a risk not worth taking unless you have been instructed to do so. Having been in the service I now see similar traits at my work and in the words of others as their place of work. For many the job is to get the task done and do so only through the established framework
- sputknick 12y ago
  I would normally agree with you, having also been in the Air Force, but as OP pointed out, these types of reviews are inappropriate for this contracting structure. The Air Force isn't supposed to care how it's done, just that it is done to listed specifications. I think more likely explanation is some officer dusted off a review for ULA and just said "eh, this is how we perform these reviews in the past, I'll just do the same thing for SpaceX."
  - tsotha 12y ago
    They have to care how it's done, because they have to evaluate the risk of the launch failing and destroying the satellite. That's not just a money issue - a launch failure means the capability you thought you would have won't be available until some later date. Maybe years later.
panzagl 12y ago
The difference between a space launch and an intercontinental ballistic missile is basically intent. Internal processes and methods are definitely a factor here.
- CapitalistCartr 12y ago
  And 3 miles per second versus 5 miles per second. Plus 3 Mk 12A nuclear warheads. I guess those go to intent, though.
- wolf550e 12y ago
  ICBMs (and SLBMs) are allowed to fail more often than spysat launchers or human rated rockets. For important targets, you can afford to launch 2 missiles, but NSA can't afford 2 $4B birds, and can't afford the lead time for a replacement if one is lost during launch. 1 year on ISS is launching on Soyuz in 40 minutes!

foobarqux 12y ago

Certification isn't about cost, it's about safety, where internal processes and methods are paramount.

bediger4000 12y ago
You know, I wish I could believe that in the context of "aerospace" or "Defense Department". After working in aerospace in the mid-80s, I came to believe that certification was more about barrier-to-entry than anything else. Ever notice how the seat-belt buckles on passenger transport planes look like a 1948 design? They're certified, and it costs way too much to certify a new design or process. Sorry to be cynical, but I just don't believe the public justifications of "certified" any more.
- foobarqux 12y ago
  First, I was correcting the putative point of the audit which the grandparent erroneously claimed was a cost audit. Second, I'm sure big corporations take advantage of certification process to keep smaller players out but certification is extremely important, suppliers would not follow many of the safety requirements otherwise.
- rdtsc 12y ago
  > I came to believe that certification was more about barrier-to-entry than anything else Still the case with most government procurement processes. If you manage to jump through all the "certification" hoops, you can mark up your product to an insane rate and then lobby for certification to be expanded. Then the government has a list of "approved" vendors who are basically in the "certification" club and sometimes you can only find a product of a particular type only by a particular company and when you ask them for the price, your jaw drops (especially if there is a 20x cheaper COTS one on the market at the same time).
  - HeyLaughingBoy 12y ago
    You can add Medical to that list. Need a USB thumb drive for a Medical Device? That'll be 10x the cost of the exact same one you can buy on Amazon. OK, so maybe the vendor guarantees traceability. So now you just paid 10x the price to be safe in the knowledge that you and Amazon are buying from the same guy in a basement in Shenzen.
- dingaling 12y ago
  > They're certified, and it costs way too much to certify a new design or process That buckle is a standard UI across every commercial aircraft in the World. That is very important in emergency situations. Billions of people know pull-to-release. There is really no advantage to changing the design unless we also change to three-point harnesses as in cars. But that won't happen for cost reasons ( ' safety is our number one priority, profit is our zeroth' ).
wavefunction 12y ago
Is that why the F-35 was cutting off oxygen to pilots? I'll believe your statement when I see it in action.
- nickff 12y ago
  The F-22 (not the F-35) was the aircraft that was having pilot consciousness issues, and it has not been proven that the issue was actually oxygen being cut off from the pilot. Some have also posited that the problems were actually due to the high Gs at low pressure (as the F-22 is the most manueverable aircraft ever made at high altitude), which is also possible.
  - angersock 12y ago
    pilot consciousness issues What a marvelously tame way of describing such a thing.
shabble 12y ago
It's also about cost, in the form of reliability, since the payloads commonly cost > 1 billion USD each. Not to mention, if you botch the orbit they might have to throw missiles at it so it doesn't injure peopl^W^Wfall into the wrong hands[1] [1] https://en.wikipedia.org/wiki/USA-193#Destruction https://en.wikipedia.org/wiki/USA-193#Destruction
- gaadd33 12y ago
  Aren't most commercial payloads insured by third party insurers? Why is the Air Force unable to do that?
  - cpwright 12y ago
    Maybe the Air Force doesn't want to provide the insurer with sufficient information about the payload to make a reasonable underwriting decision? Insurance also doesn't come free. The US government can also take on more risk than any particular insurer, so by purchasing those contracts would be in effect wasting taxpayer money. The consequences of something like a failed launch can be hard to quantify economically. There is clearly opportunity cost for missed intelligence not having the satellite in orbit when it is needed. What happens if the satellite crashes in an unfriendly place, losing the exclusivity to a bit of technology could be very damaging as well. I think you would find it hard to place a particular insurable value on those; but minimizing that risk obviously costs money. How much that risk minimization is worth is a question that I can't answer.
humanrebar 12y ago
> where internal processes and methods are paramount I don't buy that. Process is CYA first and foremost. If something happens, you get to say, "We adhere to the strictest processes and best practices... blah, blah, blah." Was there a lack of process in the space shuttle disasters? The flight 9525 crash? The Titanic? You want safety? Get a third party (insurer, certification body, etc.) to assume responsibility (financial, legal, political, etc.) if anything goes wrong, regardless of the root cause. You might not get perfect safety, but you'll get responsibility, which is the next best thing. Rigid processes are about diffusing responsibility, not taking it.
- panzagl 12y ago
  Existence of process does not mean the process is correct. Process failure was a big part of the shuttle crashes.
  - humanrebar 12y ago
    By that definition, everything is a process mistake and the term has no meaning.
    - panzagl 12y ago
      You're right, everything should be done by engineering ninja rockstars, and process is just what unawesome people use to keep their jobs.
      - pyre 12y ago
        Well, you're ignoring blind adherence to process. E.g. it doesn't matter how good an idea is in a 'traditional' Japanese company, if the person presenting it doesn't have seniority, it will be ignored.
  - LoSboccacc 12y ago
    wasn't it management meddling? process called for redesigning the challenger oring, as corrosion was triple than expected, and process called for inspecting the columbia tiles. you can find all of these details in the reports, and you can find the name of the managers overriding engineering processes in both instances. challanger: http://www.onlineethics.org/cms/12722.aspx http://www.onlineethics.org/cms/12722.aspx http://www.engineering.com/Library/ArticlesPage/tabid/85/ArticleID/170/The-Space-Shuttle-Challenger-Disaster.aspx http://www.engineering.com/Library/ArticlesPage/tabid/85/Art... columbia: http://www.aero-news.net/index.cfm?do=main.textpost&id=9fe7f63a-d780-403a-adcf-a51622a8a9cf http://www.aero-news.net/index.cfm?do=main.textpost&id=9fe7f... (notice that engineering requested hard photographic data to assess the damage, and her used gut feeling)

Shivetya 12y ago

I would tend to think that the officers in charge took the safe route, as in followed the established bureaucracy in place so as not to appear to be rocking the boat. Leaving established protocol is always a risk not worth taking unless you have been instructed to do so. Having been in the service I now see similar traits at my work and in the words of others as their place of work. For many the job is to get the task done and do so only through the established framework

sputknick 12y ago
I would normally agree with you, having also been in the Air Force, but as OP pointed out, these types of reviews are inappropriate for this contracting structure. The Air Force isn't supposed to care how it's done, just that it is done to listed specifications. I think more likely explanation is some officer dusted off a review for ULA and just said "eh, this is how we perform these reviews in the past, I'll just do the same thing for SpaceX."
- tsotha 12y ago
  They have to care how it's done, because they have to evaluate the risk of the launch failing and destroying the satellite. That's not just a money issue - a launch failure means the capability you thought you would have won't be available until some later date. Maybe years later.

panzagl 12y ago

The difference between a space launch and an intercontinental ballistic missile is basically intent. Internal processes and methods are definitely a factor here.

CapitalistCartr 12y ago
And 3 miles per second versus 5 miles per second. Plus 3 Mk 12A nuclear warheads. I guess those go to intent, though.
wolf550e 12y ago
ICBMs (and SLBMs) are allowed to fail more often than spysat launchers or human rated rockets. For important targets, you can afford to launch 2 missiles, but NSA can't afford 2 $4B birds, and can't afford the lead time for a replacement if one is lost during launch. 1 year on ISS is launching on Soyuz in 40 minutes!