Write Something You're A Little Too Dumb To Do Right

Posted: 7 months ago (2009-12-21 16:59:31 UTC )

The very best projects tend to come from guys who are very, very good, doing what they're good at, with some time pressure coming from a real project that needs to use what they're building.

If you're doing that right now, ignore this post and move on. You're golden.

For the rest of us, you're going to need to get smarter until you can be one of those guys, doing that. I've been there and it's awesome -- but now I need to get smarter so I can do it again, on a harder project. So do you.

The way to get smarter in a hurry is to write something that you're just slightly too dumb to do. For preference you should be under schedule pressure as well, from a real project. Having an application focuses your mind like nothing else, and it's especially vital when you're not smart enough to work it out ahead of time (pro tip: nobody is ever smart enough to work it all out ahead of time).

The nice thing about working out a hard problem is that you're forced to work just outside your comfort zone. The nice thing about having an application is that you find out if you were right about it. The nice thing about working only slightly outside what you can do is that you get to work on a lot of problems, with frequent feedback -- waiting too long between problems won't advance you as quickly. Yay for quantity over quality :-)