wheatpenny opened this issue on Oct 12, 2006 · 141 posts
kuroyume0161 posted Mon, 16 October 2006 at 1:42 AM
Quote - I wrote programs for various computers and they were tested before end-users got their hands on them. But then I was a very good programmer. Not like some programmers who have no pride in their own workmanship. A lot of programmers don't actually use their own software they program. Watch out for those guys.
I don't care how good you are (even if you are). I don't care how good anybody is. And I do use my own software - but I can't test and retest every facet of it as it is being developed and modified. This is what beta-testers are for. This is what updates are for.
Two things of note here:
Any system of sufficient complexity behaves chaotically (this is a fact - pick up a book on Chaos Theory). That means that there is no possible means to guarantee that an application (a complex system of non-linear computer instructions) will not exhibit chaotic (erratic/unexpected) behavior. There isn't anything that any developer can do to avoid it (like do a check - if (program == chaotic behavior) close(quietly). You can test the application on a hundred different computer configurations and still end up with a significant set of replicable problems when released for public consumption.
Remember when Poser 5/6 were released. There were these unending topics: "Poser crashes!", "This doesn't work!", and "That doesn' work!". What was a portion of the usual responses: "Works here", "No problems for me", "That works, but this doesn't". Hmmm, why is that? Believe it or not, the company that makes Poser actually beta-tests it. How could these things possibly slip by? I've already explained it.
C makes it easy to shoot yourself in the
foot. C++ makes it harder, but when you do, you blow your whole leg
off.
-- Bjarne
Stroustrup
Contact Me | Kuroyume's DevelopmentZone