Last night, I was sitting in bed reading the latest issue of TapeOp (music recording magazine). I used to be moderately involved in recording music, but these days I mostly just follow the trends and try to stay sharp. TapeOp has a lot of interviews with recording engineers and producers, and it’s great to hear what their thoughts were when they made some of their more famous recordings.

I feel sort of stupid that it took me until last night to notice (yet another)  interesting parallel with music and software. Recording is mostly a waterfall process. You record, then you mix, then you master. Some iteration is possible – you can record one song or a whole album before you mix – but most of the time, you finish recording, then you mix. When you’re dong mixing, you master. What’s interesting, is that there are a massive number of opinions on how to do each of these activities. Which mics are “best”? What rooms are best for recording a jazz combo? Do you record rock guitars with mics perpendicular, or at an offset? When should you use multiple mics? Where do you add eq? How loud do you make the vocals.

Then, there’s mastering – which in my opinion is awful on almost every pop or rock recording made in the last 10 years. Mastering (IMO) ruined the latest Metallica and Springsteen albums (and probably many others that I haven’t bothered listening to).

Whatever I think, the albums sold millions, and were (AFAIK, critically acclaimed). You know why – because despite the mastering – despite the fact they may have not used the best microphones or mic placements possible, it’s what the customer wanted. You can take the most well-rehearsed band in the world – use top notch equipment and fantastic production to recreate their sound exactly. You can add just the right punch and pop and remove any harshness and engineer the best recording ever.

But it doesn’t mean it will sell. Customers want something different, and if you don’t give them what you want, all you have is something that you are proud of, and not something that puts dinner on the table. Along the same lines, you can’t ignore the technical part of the process. Engineering quality still makes a difference, as long as you’re doing the right thing.

Same thing as my current day job.

In music, the better you are at the basics, the better you are on the bandstand. Even the pro musicians I know practice almost every day. I think testers (and developers) forget the value of practice too often.

In The Passionate Programmer, Chad Fowler suggests doing the exercises on CodeKata. I checked them out, and sure enough, the Kata are great, and I plan to start working through them. A few years ago, I solved a bunch of problems on project euler as an exercise to keep myself sharp.

As a tester, it’s sometimes hard not to practice. As I interact with software, I often ask myself “what if” …then I try it and see what happens. But this is only “sort of” testing – it’s my tester DNA seeping out into my every day life.

I’ve been thinking about other ways to practice testing. I’m a member of uTest, but I haven’t taken the time to test anything. I suppose I could volunteer to test a non-profit’s web site or find a product I like to seriously beta-test – or I suppose I could look into volunteering a few hours a week in a MS product group.

How else do you practice testing?