2 ms·
Thanks for this. I've been trying to come up with a good frame for an upcoming talk (about an open source programming project), and I think that putting it in
by jacobolus 17y ago
Thanks for this.
I've been trying to come up with a good frame for an upcoming talk (about an open source programming project), and I think that putting it in terms of the scientific method is just the thing. That is, in making code, just like in making science or in any other kind of "hacking", it's essential to first and continually "get your hands on the machinery", figure out by fiddling how things work, stop worrying about breaking things, and just see what happens when you push something here or there (of course, that doesn't mean we should throw caution to the wind, but just that we should take some risks), and then after that, after the parts are understood, take a step back and build a little bit, and then dive back in.
One important part (to writing code, and therefore a subject for my talk) is figuring out how to layer abstractions one on the other, so that we can avoid duplication of effort, and can separate out logic into clearly understood parts, small enough to be effectively tinkered on, and independent enough so that each one doesn't break the others. But we can't let worrying about the purity of those abstractions, and analyzing them, ultimately get in the way of actually diving in to fiddle, and figuring out what smells, first hand.
"You don't always understand what you are doing; if you do, it probably means that you aren't really near the forefront of knowledge."