Adam Barr posted a comment on my entry on Joel's now obviously massively ancient article on interviewing at Microsoft, with some amplified comments in response to Scoble and even more amplified comments in his review of "How Would You Move Mount Fuji?" on Slashdot. (Whew. Way too much hyperlinking going on there.) It seems like there are some deeper issues between Adam and Microsoft that are going on in the background - I'm not quite sure how one jumps from my entry to the idea that I believe that "I work at Microsoft therefore I'm a genius at all things including evaluating talent" - but it is worth noting that I actually agree with many of the points that Adam brings up. Since I was just riffing on a thought I had about Joel's article, I didn't really cover everything I think about interviewing. So here are a few more thoughts that I think are worth saying...

To start with the practical side of things, I'll cop to the fact that I don't think I'm a great interviewer or all that good a judge of talent. I'm a pretty good programmer and a decent writer (I hope) but I make no claims to be Lamont Cranston. To be even more honest, interviewing people has always kind of terrified me. After a lot of conscious work I've gotten to the point where I can do it and feel reasonably secure in doing it, but if SteveB called me up tomorrow and told me he'd written me a Get Out of Jail Free card for interviewing from now until doomsday, I wouldn't be sad.

Anyway, despite the stories I've heard on the web, all of the people who I've been on interview loops with at Microsoft have been genuinely interested in the candidates and have tried pretty hard to be being fair, open and honest. But Adam's right that Trust Your Instinct by itself is hardly sufficient to conduct a decent interview. Too much belief in your own instincts can easily lead to arrogance and condescension. That principle always has to be counterbalanced with the equally important principle of Be Humble or, put an even better way, There But For The Grace.... (This applies not only to interviewing, obviously, but to the rest of work and life. Hubris has destroyed many a Microsoft competitor, and we've certainly had our share of near-misses.) The trick is finding the right path between Scylla and Charybdis - too much self-confidence or too little self-confidence can be equally deadly. Do I always get the balance right? Suuuuuuure I do...

One of the things that Adam will probably find most surprising is that I tend to agree with Poundstone's "fifteen second" evaluation insight. More often than not, the initial impression I get from the very beginning of the interview is pretty much where I end up at the end of the interview, even though I consciously put my first impressions out of my mind. (One is certainly free to draw the conclusion that I do not really succeed at putting them out of my mind, but I would assure you I've gotten pretty good at it.) I've thought for a long time that the first thirty seconds of the interview are often the "real" interview and then the other fifty nine minutes and thirty seconds are follow-up to catch those times when the first impression is wrong. Which happens, certainly - sometimes people take a little while to get "warmed up" or may be having a bad day of it. But it is amazing how much the non-verbal cues that we're hardly even aware of can say so much about a person. (I would point to another excellent Malcolm Gladwell piece from the New Yorker on this subject.)

And, finally, I'd completely agree that Microsoft's style of interview is not a good intelligence test, mainly because that's not what it's intended to be. When I interview someone for Microsoft, I don't really care whether they're brilliant. I don't really care if they're an incredibly hard worker. I don't really even care if they like computers and/or Microsoft and/or their old bosses. All I care about is, in order: a) can they do the job?, b) can they do the job well?, c) can they do the job well at Microsoft? The goal of the interview, in my eyes, is to figure that out and nothing more. There are really smart people who are nonetheless completely unable to get anything practical done. There are really hardworking people who nonetheless lack the creativity needed to solve difficult problems. Choosing for a single skill or talent or way of working is a really, really, really bad idea. The important question, indeed the only question, is whether the person is a good fit for the job and the company.

OK, one more point. In the case of the intern interview, Adam wondered aloud whether it would really be so bad to "give the kid a chance" and if he didn't work out then "kick him to the curb and no permanent harm done." The problem with that is pretty simple if you think about it for a moment. My team gets one intern for the summer (if we're lucky) and we interview a bunch of college students for that position. It's not a question of whether the kid is smarter than me or smarter than Bill Gates (in the case of the former, it's often an even bet), it's whether he's going to do a better job than the rest of the kids who are vying for the position, pretty much all of whom are deserving of a break one way or another. I'd love to give 'em all a tryout, but there's only so much one team can do...

[Hmmmmmm.... "Mr. Stud Interviewer." I like the ring of that... Darn, now I'll have to get some new business cards printed up!]