At the beginning of the 1980s, shareware was everywhere. It seemed that there was an application to do anything you needed, ready to run, and only asking for $5

At the beginning of the 1980s, shareware was everywhere. It seemed that there was an application to do anything you needed, ready to run, and only asking for $5-$50 in return for its continued use. Thus, I was ready to swear on a stack of Bibles that there was NO MONEY WHAT-SO-EVER to be made writing computer software. I wanted to be a programmer, but I hedged my bets starting to earn an EE degree in college. Remember, this was right as Bill Gates rose to be one of the richest people on the planet. So now, it seems there is no future in software, as there didn't seem to be then. It's an illusion. Software is complex, and thus has very complex requirements. Those requirements can't be specified... they can only be discovered. Your job is (as it always has been) to match the problem to the available tools. In electrical engineering circles, this is called impedance matching, and it's a black art. In that world, there are only 2 independent values to worry about, resistance and reactance, but in software, there are an almost unlimited set of variables, so it's always going to be quite complex, and require you or someone like you, to make those optimizations.