Results 1 to 6 of 6
  1. #1
    party pooper
    Posts
    13,025
    vCash
    500

    software and war machines

    The register has a nice article on "windows for warships" which is being deployed in the UK.
    http://www.theregister.co.uk/2007/02..._boxes_at_sea/

    Also, did anyone else hear mention of the CNN TV story that claimed that 12 F-22's heading for Japan from Hawaii had to turn back when the navigation systems for ALL 12 planes in the formation crashed hard as they crossed the international date line? The planes apparently had to follow their tanker back to base...good thing it was up there with them. Haven't seen it on CNN's web page yet...only a brief mention of it on slashdot.

    Anyway, all this makes me wonder....with soldiers, airmen, and sailors putting more and more of their job on software, and of course, deadlines to get systems deployed, I wonder how the process works of testing the code that our armed forces depend on and working all the bugs out happens. It makes me curious because it opens up a whole new realm of possibilities for things to go terribly wrong...things that likely, nobody ever thought of before.


    anyway, I was just wondering aloud.

  2. #2
    SoonerFans.com Elite Member
    Location
    Norman, OK
    Posts
    23,620
    vCash
    500

    Re: software and war machines

    I remember hearing/reading that our Arleigh Burke AEGIS destroyers run a version of Windows in the CIC.

    I'm not happy until you're not happy.

  3. #3
    Sooner All-Big XII-2-1+1-1+1 reevie's Avatar
    Location
    Oklahoma City
    Posts
    1,017
    vCash
    500

    Re: software and war machines

    Quote Originally Posted by Ike
    The register has a nice article on "windows for warships" which is being deployed in the UK.
    http://www.theregister.co.uk/2007/02..._boxes_at_sea/

    Also, did anyone else hear mention of the CNN TV story that claimed that 12 F-22's heading for Japan from Hawaii had to turn back when the navigation systems for ALL 12 planes in the formation crashed hard as they crossed the international date line? The planes apparently had to follow their tanker back to base...good thing it was up there with them. Haven't seen it on CNN's web page yet...only a brief mention of it on slashdot.

    Anyway, all this makes me wonder....with soldiers, airmen, and sailors putting more and more of their job on software, and of course, deadlines to get systems deployed, I wonder how the process works of testing the code that our armed forces depend on and working all the bugs out happens. It makes me curious because it opens up a whole new realm of possibilities for things to go terribly wrong...things that likely, nobody ever thought of before.


    anyway, I was just wondering aloud.

    Yes, software get tested before it is put in operational use. That doesn't mean it is perfect of course...

  4. #4
    party pooper
    Posts
    13,025
    vCash
    500

    Re: software and war machines

    Quote Originally Posted by reevie
    Yes, software get tested before it is put in operational use. That doesn't mean it is perfect of course...

    I know it gets tested...I just wonder how rigorous that testing is.


    The thing about software, is that it introduces a brazillian new failure options that nobody in their right mind would ever consider. and as software gets increasingly complex, the opportunities for unanticipated failure grows exponentially...like for instance, navigation equipment not being able to handle the international date line.

    It's pretty easy to make sure software does what it is supposed to under ideal conditions. It's a lot more difficult to examine all the possible (and often times stupid) ways that the software can break and make sure you protect yourself against those.


    As we start loading more and more complex software onto our war machines, especially if it becomes generalized software like windows, you start running into problems with dependencies, where suddenly your tanks targeting system depends on a library written for a completely different system that was created in 1982 (this is just a hypothetical, but plausible). That library may or may not interact with the hardware in the tank exactly like everyone thinks that it should, and could cause a lot of problems.

    Anyway, the things I wonder about are what manner of control of this kind of stuff does the military need to exert. Obviously they don't want to waste their time hiring their own software engineers to write their code...but if they don't do some of that, then they could easily wind up with some big problems on their hands in critical situations.

  5. #5
    Dirty bastard soonerboomer93's Avatar
    Location
    I'm in your spek, negging your ***
    Posts
    11,621
    vCash
    500

    Re: software and war machines

    yes, M$ is evil

    M$ is more evil then al queda
    “If a team is to reach its potential, each player must be willing to subordinate his personal goals to the good of the team.”

    Bud Wilkinson

  6. #6
    Sooner All-Big XII-2-1+1-1+1 reevie's Avatar
    Location
    Oklahoma City
    Posts
    1,017
    vCash
    500

    Re: software and war machines

    I was trying not to get detailed in my response. But software, like hardware, goes through developmental testing and operational testing. The rigourousness of the testing would be based on the complexity of the software and the systems it touches. That would all be addressed in the test plan.

    The government does hire software engineers. But regardless if it is a government product or a product bought from a contractor, it still goes through the same testing process.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •