Slashdot led me to this article inTech Review in which Bjarne Stroustrup says, among other things, "People reward developers who deliver software that is cheap, buggy, and first. That's because people want fancy new gadgets now." I've agreed with this for quite some time. The converse of it is that, for areas where the pace of innovation has slowed, or where bugs will cause serious inconvenience to people who use the software, quality can be (and is) much higher. Early cars, or even later cars like the ones my parents drove in the 1960's, were unreliable by today's standards, and don't even ask me about the TVs of the 1960's. Most of software is in that era now. In a few decades it'll work much better, not because of dramatic advances in technology, but because customers just won't want to pay for sloppy workmanship any more.
So why did people buy those cars and TVs? Why do they use today's software? Simple. What came before was even worse. There's an interesting book called "To Engineer Is Human" (find it at your favourite book place) which argues that engineering, as a discipline, makes progress by learning, and it learns by failing. I've been working in the software industry for twenty years and I've still got a lot of learning to do!
07 December 2006
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment