In my previous post, I discussed the semantic gaps that afflict current programming languages. These gaps are caused by tools focusing on syntax and parsing, and mostly neglecting human factors. I’m not just talking about the fact that languages are clumsy for us to use (more about this later); I’m saying that they ignore our need to talk about important realities of software development: security, coding habits, testability, maintenance plans, dependency management, requirements, intellectual property, and much more.
All this stuff falls within our scope of concern, but none of it is describable in our languages. That’s weird. Imagine we hired a general contractor to build our house, and he was great at swinging hammers and leveling studs. But as soon as we asked him questions about building permits or hiring subs or choosing the right kind of concrete for the foundation, he acted like he didn’t have a clue what we were talking about. We’d be likely to end up with lots of false starts, poorly met requirements, endless kludges, tons of frustration, a heavy QA burden. Hmm… That sounds familiar.
I call this lack of semantic continuity the lacuna humana — the human gap.
The good news is, gaps can often be bridged.
I promised I would describe a bridge that has a lot of virtues, and I’m going to begin that work here. It might take us a couple posts to get all the way across, though. Thanks for hanging with me…