I was catching up Joel Spolsky's blog earlier and he had a series of posts containing the transcription of a talk he gave at Yale (part 1, part 2, and part 3). The posts are long but I thought they were worth reading. In particular there were two things he touched on that I thought were interesting.
In part 1 he talks about testing and the fact that there's no realistic way you could have automated tests cover everything. He argues that at the very core the only thing you can do is check to ensure that one program behaves like another or according to a set of tests. Those tests however can't possibly cover everything that needs testing.
He gives an example of the move at Microsoft to more of an automated test process and the rumored laying off of testers who don't know how to code to bring in SDETs who can program those automated tests. He points out that those tests can't test a number of things that require someone actually look at the product such as whether thing were laid out in an orderly and logical fashion and whether the product has a feeling of continuity.
You can test a lot of things through automated testing but there's always going to be a need for someone with a sharp eye to go through and make sure everything is ship shape.
In part 2 he writes about why an in-house programmer sucks. He gave three reasons and they are all very good but the one that really struck me was the second:
"So, the number two reason product work is better than in-house work is that you get to make beautiful things."
Numbers 1 and 3 are that you never get to do things the right way and that at a software company your business is software so there's more incentive for the company to treat developers like rock stars (and who doesn't want to be treated like a rock star now and again).
Number 2 stood out to me because it struck a chord somewhere and reminded me why I started getting interested in programming. I wanted to create something cool and useful and beautiful.
I played guitar for a long time and I didn't do it to pick up chicks (though I may have used it in that way a time or two), I did it because I wanted to create something. I wanted to write something that would touch someone or cause someone to open up their eyes to a new perspective or challenge an existing perspective. I wanted to create something beautiful. I used to sit and play for hours, now I only play around on the guitar here and there when my son (who is 18 months old on Saturday) points to it and either wants to watch me play and dance a bit here and there or wants to sit in the case after I've pulled the guitar out.
Currently I'm working on various small (on a scale of all development undertakings) projects. Some of them are pretty cool (like the one I'm working on now that I'll post about once it launches) and some of them aren't as cool (like forms). Though I can still work towards creating flexible and elegant solutions. In the grand scheme of things they may not be groundbreaking or hugely significant but they're still opportunities for elegance.
Someday I'd like to create something beautiful and significant. And I will.