A Murder of Yaks

Aug 20, 2013

An occupational hazard of spending my days working on several large, fast-moving codebases is that I've started to evaluate every new tool or idea in the context of a large codebase, maintained diffusely by a lot of people. In short: I've become incredibly conservative about new ideas and new pieces.

This is made quite a bit worse by the fact that your standard "hello world" usually shows the golden path in its best light, and it is often very hard to find better information on how to use these tools in anger: what happens when you aren't running a demo application? What if you have a broad mix of code across a big site?

In short: my life is a series of yaks, and every tool is measured by it's ability to - out the gate - help me shave them.

I've been struggling with this for some time, and have tried various approaches to getting past it. The first was vicarious: read a lot of articles on Hacker News, and bask in the warm glow of other people's accomplishments. This has the virtue that its compatible with my RSS feed and commute. It has the downside that I have ended up a dilettante on far too many ideas and tools.

The second was a serious of ambitious side projects: take a new idea, implement it with new tools! Yay! Sadly (and predictably) I fell foul of the two-things-at-once problem: either solve a new problem, or use new tools. Doing both compounds the likelihood that you will end up with no solution at all.

Both of these approaches have not really worked for me. My experience with exercise suggests that I learn best by focusing on a specific subject, and working through a lot of small examples.

So.

This entry was tagged Experimentation and Tech