Re^3: YAC (Yet Another Challenge): Oldest Useful Computer Text

We us a hardware clone (see below) of said machine to control the nuclear reactors and the associated boilers and valves at the Bruce "A" and Bruce "B" Power Generating Stations. They are opperated by Bruce Power.

Some parts have been replaced with more modern components. For example, the core memory has been replaced, but it still only has 32KW (64KB, 16 bit addressable) of RAM. We load binaries on the machine via a serial connection to a device pretending to be a paper tape reader to the Varian.

Comment on Re^3: YAC (Yet Another Challenge): Oldest Useful Computer Text

Replies are listed 'Best First'.
Re^4: YAC (Yet Another Challenge): Oldest Useful Computer Text by hsmyers (Canon) on Apr 17, 2006 at 16:12 UTC
I know that many live by If it's not broken, don't fix it... but this seems a little extreme. Anyone in your managment chain see this as a liability? --hsm "Never try to teach a pig to sing...it wastes your time and it annoys the pig."	[reply]
Re^5: YAC (Yet Another Challenge): Oldest Useful Computer Text by ikegami (Patriarch) on Apr 17, 2006 at 16:17 UTC
Why would it be? Do software and hardware designs go bad after a while? If you're worried about the hardware itself going bad, that can happen to new hardware just as easily. Maybe even more so since they are much more complex. Redundancy and fault tolerance are key factors no matter which hardware is used. "Fixing" it would involve rewritting entire bookshelves of assembler code (since assembler is machine specific). That includes completely redesigning the OS to take advantage of the extra memory and disk space a new machine would provide, and you'd have to do deep structural changes to all the applications to be able to run on this new OS.	[reply]
Re^5: YAC (Yet Another Challenge): Oldest Useful Computer Text by Anonymous Monk on Apr 17, 2006 at 16:39 UTC
I know that many live by If it's not broken, don't fix it... but this seems a little extreme. Anyone in your managment chain see this as a liability? It's politically difficult to tell people: "we're going to make changes to how we run our nuclear reactor" unless you have a very good reasons. The Chernobyl disaster was only possible because some manager decided that it would be a good idea to let scientists do some experiments on a live nuclear reactor (and then let them disable multiple fail-safe systems, so that they were running without backups -- then they reacted badly to a lack of instrumentation). Sometimes, a healthy dose of "leave it alone" syndrome is a good thing; especially when mistakes can kill thousands of people. -- Ytrew	[reply]