Microsoft's Iain McDonald: It's the end of a (Longhorn) era

Iain McDonald, a guy with the humble title of Director of Project Management for Windows Server, has had a long and deep impact on Microsoft's Windows organization.

On May 15, the day that Microsoft officially announced the "Windows Server 2008" product name, McDonald blogged about the end of the Whistler/Longhorn/Blackcomb era of Windows codenames:

"I do have to say that the Longhorn name going away is a little sad to me – it’s the end of the naming that I started the week after we shipped Windows 2000 in December 1999. At the time we had the Neptune & Odyssey projects that had to become one. (Former Senior VP of Windows Development) Brian Valentine told me to take a week off & come back with a new project name. I went to Whistler with some friends (hey dave, hey rob!) & it was when I was learning to jump on my snowboard – also before I used a helmet which is dumb.

"I was in the terrain park on whistler – I lined up a jump & had a thought on the way onto the jump. “What will we call the next release” has nothing to do with that & thinking often is not the best idea when going for a jump. I blew the jump bad & landed on my head. This is not an experience to have - wear your helmets, kids. Lying on the ground for 10 minutes made me realize next release would be called Whistler. At least I stopped thinking… (now is the time where (Group Manager) Jack Mayo will say i stopped thinking until 2005).

"The idea that Blackcomb wasn’t much further but it’s much better kinda fell by the wayside & grew soon afterwards. (Current Corporate VP of Entertainment and Devices ) Joe Belfiore decided to create a little release in the middle & named it after the bar between the mountains – Longhorn. so much for the cool names."

I had a chance to chat with McDonald at the Windows Hardware Engineering Conference (WinHEC) 2007 in Los Angeles earlier this year. We didn't talk a whole lot about Windows Server 2008. Instead, we touched on everything from his hobbies (motorcycles and guitars), to the "worst project ever run," -- which, in McDonald's opinion, was Windows 2000.

Here's Part 1 of a few of the interesting bits from my tape:

MJF: I know you used to be a musician. But how did you ever end up coming to work at Microsoft?

McDonald: I was playing music. And anyone who says they're making money playing music isn't kind of telling the truth. I dropped out of uni and did this kind of little computer schools course that a friend of mine had recommended. And I did a bunch of software contracts where you go and work for three months.

I was working at this one place that was a credit union. We made this system. The teller had to go and report how much money they had in their drawer at their station. We were sitting there one afternoon and we worked out that they could go on and send it to the next teller and say "Hey, I've got eight dollars in the drawer."

About two weeks later we realized there were a lot more messages in there than the teller sessions that we had. We went and had a look at them. They were like, "What pub are we going to this afternoon?" and "Nah, we don't want to go there today." And it turns out they'd started using this system we had put there, a reporting system, as a social network and a means of communicating between them.

So that led me onto the idea that e-mail was a good idea. And so we did some stuff there and expanded that. Then I went to work for Northern Telecom and we did an e-mail system that we spread across Asia and we created lots of stuff there.

MJF: Had you had any real training in coding or software development, at that point?

McDonald: Well, I'd been writing software for mainframes. (Guys like Former Windows 2000 General Manager) Frank Artale Brian and me….we just didn't fit the mold. We did fit the mold of the people who came in in the mid-80s. I don't know that any of us would fit the mold of what's there today -- like the kids who are coming in now.

So yeah, we had to write a whole bunch of stuff. We had Unix with a whole bunch of Mac clients. In the telecom world, Apple was big at the time. And then I was thinking I was gonna leave. I saw two jobs: There was one with Novell and one with Microsoft. And a friend of mine said "You're not going to fit in at Novell. With that kind of work, you're just not gonna fit in." The Microsoft one -- I went for a networking job and in speaking with the guy who originally hired me -- he had a big problem he had to fill and I told him I'd been doing that for four years.

I started with PSS (Microsoft Product Support Services). At the time, there wasn't any Microsoft Consulting. I met a guy who convinced me to come to America. I came to America. And I realized I had come have way around the world to be at the center of the universe for what I was doing. The guys on the dev team said, "Why'd you come over to support? We would have hired you." But I did my commitment out there (in PSS). And I was really good at that job. That was probably the job I was best at in my whole life. It's funny. Bill Laing's group has now the three original escalation engineers. Now there are hundreds of them worldwide, but originally the three of us were neighbors.

Then I did MS Mail and Exchange. Then Brian (Valentine) went to Windows. He said they didn't have a day-to-day run-the-project person. And he asked me to do that, and I said sure. I loved to work with Brian. I worked for (Brian) for nine years. He is a great in-the-trenches leader. He's so good at the big leadership thing. We swapped mail a couple of weeks ago and we're gonna go and ride motorbikes sometime. Except we can't find a time to do it.

It's funny now. I look back at all the travails of the past couple of years. But nothing was worse than Windows 2000. People outside are going and saying, "Vista this and Vista that." But nothing was worse than Windows 2000. They just don't remember. That was the worst-run project of all time.

MJF: What made it the worst?

McDonald: We finished that project and I spoke to someone who said I worked 30 seven-day weeks this year. That was someone who worked for me. I knew I worked a lot more seven-day weeks than she did. It was dumb. You should never run a project like that. In the long run, there were some great things we did, but we also made some just fundamentally stupid decisions. Like the security one. Code Red in August 2000: It's amazing how stupid that was in retrospect. We just didn't know.

We did a lot of stuff. We wrote XP. And you know, XP was not all that well-received. I remember in November that year, when the first security vulnerability came up for it, I was on holiday in Australia. And I got a call from Brian (Valentine). He was whipping me. He said, this will be on the front page of the paper, you know.

We've gone through the equivalent amount of time with Vista and it hasn't got any of those black eyes, which says to me, I think we've done pretty well with that.

I think security is a ten-year job. We started in 2002, so we've still got years to go on that.

MJF: What do you do day-to-day? What's your job as head of project management?

McDonald: I watch my time really jealously. I keep a running average of what I've done in the last four months. I watch where I spend my time here and there. I spend 15, sometimes 18 percent of my time doing customer stuff directly. I don't know what I'm supposed to do, but I've learned so much (from them.)

The great thing about making software … is customers go and take it way beyond what we can ever imagine. And I realize way back to MS Mail and the phone company I was with in Australia. We built that system. And then we found out what people would go and do with it.

Then I spend time on old things, current things, next things. And also managing my people and mentoring my people. Then there's the whole dealing with our friends in Europe…(the European Commission).

Q: Ah, you get to do that, too?

Stay tuned for Part 2 of my Q&A with McDonald later this week to hear what he said to that one. And more...