With respect to a stable, useful life for software, while I agree that it would be nice if, as an industry, software were mature enough to not have to continually struggle to keep software running and just, plain get things to work for now, much less for an extended period of time, it would be a shame to do so at the expense of industry progress. New frameworks, platforms, languages, etc. I see as a necessary element for the progression of the industry - something that the automotive industry you compare it has shown to be distinctly lacking. The 1989 Geo Metro performed about as well as today's advanced Toyota Prius at a fraction of the price and complexity; by and large, I'd say the automotive industry has been very busy indeed, but made little real progress. As for reliability, I'd venture a guess that there are more fix-it shops out there for busted cars than there are for software. I suppose I am equally dissatisfied with the lack of progress in established industries such as automotive and banking as you are with the excessive activity in software...