You say: >I think we need to get real. Programming isn't something most people are interested in, it's not glamorous, it's factory work. Lets not delude oursel

You say: >I think we need to get real. Programming isn't something most people are interested in, it's not glamorous, it's factory work. Lets not delude ourselves that we're at the cutting edge of new discovery. And: >Building companies, developing products, increasing revenue are all far more interesting to everyone. But really, when was the last revolutionary product made? The last revolutionary way to increase revenue that was sustainable in the long run? I've seen many revolutions in software, distributed version control for instance ie git, LLVM, gaming and visualization software, and hybrid synchronization systems such as unison. In OSS the 'As Seen on TV' type products generally don't see the light of day, they get buried and forgotten, it's only incorporate environments I've seen crappy interfaces using decrepit languages from the 90's, because those programmers have stopped, or never even started caring about programming. If they did they might take the initiative to fix up the program themselves. As for the factory worker analogy, I think it's dead wrong. It's like saying anyone who ever designed a product is a factory worker. Programmers have to do a lot of design decisions, interop systems etc. Yeah if you're programming in Java, or making corporate program in programing lanaguage du jour you're probably not really innovating, like any part of life programming is self driven, for every Google there's a struggling business and for every git, there's a Java webapp for accessing an Access database. Also if you think AI is going to obsoltete us in 50 years you're kinda pushing reality there. MS tried to make programming easy and accessible, look at every <LANGUAGE> in 24 hours book. They download MSVS and run through a bunch of wizards, that's about as much control and innovation as you get with a Java webapp that accesses an access database. Not only that but who's going to program all the new features into the the automation AI? When Intel releases a new architecture or something happens there's always gonna be a need for programmers to translate and incorporate data. Essentially if what you said was true SIGGRAPH wouldn't exist. There's plenty of innovation going on in the software field, and plenty of talent, like every field there's a bunch of idiots, and plenty of talent. If anything you made a great social commentary, people care only about money, the easiest way to get money is working in middle management. Your argument is based on the corporate programmer stereotype.