diy

Many years ago, I applied for a job at Microsoft, was flown up to the Mountain View campus, and was put through eight hours of infamous Microsoft interview questions. You know the type: brainteasers intended to reveal to the interviewer how I solve problems. Most of the questions were standard riddles, like, How do you match up three light switches in one room to each of the three light bulbs that they control in another room while making as few trips as possible between rooms? (Hint: there's a way to do it in one trip.)

But one question, toward the end of the day, caught me off guard. The interviewer asked, "Why do the drainpipes of sinks have P-traps on them?" (P-traps are that curve in the pipe under the sink.)

At the time, I thought it a pretty strange and irrelevant question. (Yes, more so than the one about the ratio of blue paint to red paint in two buckets where the same spoon has been dipped into the red one and then the blue one and then back into the red one.) After all, why would someone applying for a job at Microsoft need to know how the plumbing under a sink works?

If I remember correctly, I figured that since the trap often gets clogged -- like the time I tried to put a whole colander of old spaghetti down the garbage disposal -- the trap must be for, well, trapping chunks of food before it gets into the sewer system. Not quite, the interviewer explained. Turns out that getting clogged is actually an unwanted side effect of the P-trap. Its real function is to stay full of water so that sewer gases don't come back up into the house.

I didn't get the job -- not just because I didn't know what a P-trap is for, but also because I wasn't much for brainteasers back then, or perhaps because I generally wasn't all that bright at the time. But anyway, for the most part, I soon forgot about the seemingly silly P-trap question.

It wasn't until not too long ago, when G and I bought our house, and when I suddenly realized that I really like do-it-yourself projects -- and suddenly found myself fixing leaking drains and pulling toilets off their bases and rigging washing machines so as not to overflow -- that I remembered the P-trap question again. And it was then that I finally understood what an interviewer could gain from asking such a question.

See, if I were to answer the question incorrectly, as I had, then it would be no big deal: the interviewer could assume that I simply had lived in apartments and dorms for a long time and never had to deal with plumbing. However, if I were to answer correctly, then the interviewer could immediately imply a handful of things about me -- like that I had at some point fixed a sink myself, and therefore had had to teach myself a new skill in a limited amount of time, and that I was the type of person who would do it himself rather than wait for a plumber to show up (and that I probably could have waited for that plumber but was just too cheap to want to pay one.)

In other words, the interviewer could find out if I was a do-it-yourselfer. And in doing so, they'd find out quite a bit about my personality.

It's clear to me now why I like meeting other do-it-yourselfers so much. Sure, we share a hobby in common. But also, I can be pretty sure that they're self-reliant people. They believe they can learn new things. They like to learn new things. They like to solve problems. They like to make things better. Sounds like the perfect qualities to look for in a new hire.

If I'm ever interviewing you and you want to impress me, find a way to tell me about how you spent the past weekend tiling your kitchen floor.

c:\history\ancient

Found this while cleaning off bookshelves in our bedroom today. MS-DOS 6, copyright 1994. It pre-dates the Web (at least, the graphical version that we all know now.) Never removed from the shrink wrap. Even has the certificate of authenticity on the back. Classic.

victoria

Parliament Building, Victoria, BC, Canada

brainstorming gone bad

Marc Andreessen blogged recently about brainstorming, pointing to research quoted in the The Medici Effect that states that groups that get together to brainstorm consistently underperform groups that split up and brainstorm ideas individually and then come back together later to share those ideas.

It's an interesting insight, and I've certainly had my share of meetings where I've just wanted to go away and think quietly for a while. But I'd say the bigger problem facing most companies is not generating ideas, but rather knowing when to stop brainstorming and start doing.

I think this is especially a problem in the tech industry, where it's easy to talk about all the cool things you could do with a product. Over and over again, I've seen teams come together to solve a problem and walk away with a half dozen other projects to think about, each with their own problems.

One organization I worked for called a meeting to discuss what should have been a fairly simple request to modify a checkbox on its Web site. They wanted the box to be pre-checked when users first saw it, in hopes that it might raise subscription rates to their e-newsletter. Sounds like something that probably could have been handled over email or in a phone call? Nope. The meeting lasted an hour, with six people representing a variety of different departments and levels in attendance. By the end of the meeting, the original reason for "checking the checkbox" had long been forgotten and replaced with eight new "wouldn't it be great if" ideas. And the associate who had initially been asked to make the change was left sitting at the table wondering aloud if she was still supposed to make the change or not. Awesome.

Organizations like these aren't stagnant behemoths in need of new processes to tease out the next brilliant idea. They're teeming with ideas -- and no ability to focus on just some of them. What they need is not someone to say, "Hey, let's change the way we brainstorm" but rather "Hey, we're past the brainstorming phase on this one and into the decision-making phase."