« Microsoft Xbox runs as a University Web Server | Home | Microsoft Office versus OpenOffice (For Students) »

Ship’s Autopilot Fails due to Windows Error

By DivineOmega | June 17, 2006

This is quite old news, but none the less it is funny.

One of the Navy’s ships used a Windows NT powered network to automate many of the functions of the ship, is meant to reduce the number of people needed to operate a ship. This dependance on Windows NT brought predictable results: the USS Yorktown suddenly became dead in the water when it experienced a cascading failure in its network of NT-powered machines. This error was obviously a programming error (or at least an oversight) due to its nature. The error was ‘can not divide by zero’. How this can cause an entire network to fail is a complete mystery!

“Because of politics, some things are being forced on us that without political pressure we might not do, like Windows NT… If it were up to me I probably would not have used Windows NT in this particular application. If we used Unix, we would have a system that has less of a tendency to go down.”

Ron Redman, a deputy technical director of the Navy

Tags: , , , , |

One Response to “Ship’s Autopilot Fails due to Windows Error”

  1. Curt Says:
    June 19th, 2006 at 2:16

    The event is true…however, the “USS” stands for United States Ship.” The YORKTOWN was (now decommissioned) CG-48, the second ship of the AEGIS Cruisers.

    I was at the first meeting of the “Smart Ship” project in Carderock, MD, in the spring of 1993. The CNO, Adm Mike Boorda, had put together the project to see how we could do more with far less people on ships in the future. The rules were: All current policies, programs and equipment were fair game to be questioned, disposed of or overhauled. That was a pretty bold directive for the Navy, and it worked. The BOA went out in Commece Business Daily, and invited everyone to participate, but without a promise that any business would come their way as a result.

    I transfered before I could get more involved, but I know a good number of great ideas were tested, and there were some truly interesting discussions held about why things were the way they had been….but…that’s another day’s story. The significantly reduced manning of the DD(X) being discussed for construction for our Navy is a direct result of Adm Boorda’s foresight and leadership.

    As a software safety person, I hated the idea that anything besides admin stuff in the military would be run on top of Windows (or any commercial OS), because there isn’t a large company that will give you the rights to look inside to the code, just to make sure no one says “What was that ‘WHOOSH!’?” at a critical moment in operations.

    The engineering plant of the DD-963s and the CG-47s all ran on a serial data loop, which was a pretty neat system, for the time, not to mention pretty bullet proof in operation. I also was an Engineer Officer for 28 months on a DD-963.

    While there was some good ideas about putting another layer of computerization on the system, this case (of the YORKTOWN) obviously wasn’t done too well, or…one of the “undocumented features” of the Windows OS wasn’t fixed…and Lord knows there are a few (thousand) of those…

    Cheers!

Comments