0 Members and 1 Guest are viewing this topic.
Diminishing returns for the programmers is the issue - they could spend years trying to find the hard-to-find bugs that creep in over time to make operating systems and other programs go wrong, but if they don't occur so soon that you can't get a reasonable amount of work done first, you can just fix them by rebooting, so the programmers just leave you to do that while they go off to write new code to do something more productive. The questions they ask are, is it good enough to sell, and is it worth trying to inprove it a little rather than getting on with writing more important code which will make the old stuff redundant anyway. In most cases there will be many rare bugs which gradually disable a system, so hunting down one and fixing it won't even show up as there will be lots of other bugs which will still force you to reboot every now and then. These bugs will also be the fault of different people, and while some of these programmers may feel the need to put in the effort to eliminate the bugs they've caused, the rest can't be bothered, and that means that some who might have bothered won't bother because they don't see the point if the rest aren't going to put in the work. If the operating system and programs that run on it are intended to operate equipment that must not go wrong, more work will be done to check code and eliminate bugs for that, but the costs will go through the roof.
I laughed so much at this thread that I nearly cried; that computer is fabulous. Actually, at the BBC's Make it Digital show in Cambridge a few weeks ago they had one of the original consoles for PONG, sitting alongside an original Atari; we played bat n ball for about half an hour; ridiculously addictive...
*How Bill Gates managed to write a BASIC interpreter in machine language in a couple of days astonishes me. I guess that is why he has billions of dollars, and I don't.