I have been designing and hacking my own languages (to varying degrees of completion) for almost as long as I have been programming. A lot of the time, their genesis is a thought like, "what if language X did Y?" or, "I've never seen a language that does this, this, and that... I wonder if that's because they're insane things to do?"
When you're working on a system, how do you approach the question, "Is this really useful, or am I spinning my wheels chasing a conceit?" Is the answer as simple as try it out and see what happens? Or do you have some sort of heuristic that your many years of experience has proven to be helpful?
replies(1):