Steven Sinofsky, president of Microsoft's Windows division, spoke with TechFlash about "One Strategy," his upcoming book with Harvard Business School professor Marco Iansiti, which is based on Sinofsky's internal blog posts during the Windows 7 development process.

Read on for excerpts from the interview.

TB: You're not the most public person. But through this book people are probably are going to get a better sense for what you're like. What impressions will people come away with about your management style, and your approach to the business?

Steven Sinofsky: Of course, the book isn't really supposed to be about me. What Marco and I wanted to really capture was this notion of, when you're working on a project -- especially anything at a significant scale -- the biggest problem is figuring out what to do and getting it done. The matching of strategy with execution, and really trying to get those two to be the same. The alignment of those two things is something that I believe in, but really the book is just about a way that it worked for us based on the experiences we've had.

TB: One of the lessons that you talk about is the need for planning with input from across the group, but also the need to stay flexible enough to divert from the plan. How does that work?

Sinofsky: Certainly that's one element of how to align strategy and execution. First, you have to have a plan. There's a long history where companies would say, OK, we need a really good strategy, so now we're going to have a small group of people come up with the strategy. And then the people who do all the work every day look at what those people produce and say, well, that doesn't have anything to do with how hard it is to get stuff done or our resource constraints, and other things. The whole picture fails.

So there's this conventional wisdom that you have to have a master plan and you then just hand it off to people. But of course you could never plan at that level of detail. If you think about just remodeling your kitchen, you could go to the architect and make sure the architect puts on the drawings, "Screw goes here, nail goes here, glue in this line." But it would take them five years to come up with those plans, and then the contractor would just show up and still say, well, these materials don't even work with glue. So you need this inherent flexibility, and a big part of what we tried to do is figure out how do you have enough of a plan so that you can then be flexible in a rational way.

TB: These principles that you're talking about -- the planning, the flexibility -- how widespread are these practices inside Microsoft, and how unique is the Windows team in that regard?

Sinofsky:If you go back to the start of where the content really came from, when I change jobs, I go through the same thing that a lot of managers go through -- which is, OK, I've got a new job, I have to do all this new stuff, I have to learn all these new things, but we have to keep doing stuff. The whole team is not going to wait around for me to get up to speed. So how do you go through that transition? And that's pretty common, whether you're just a first-line manager or more senior or higher in an organization. So I started blogging because it seemed cool at the time. I'm actually really glad that Twitter wasn't really around then because I don't really know how I would have conveyed all of what we need to convey in 150 characters at a time.

TB: No, Twitter is not your medium, after reading your posts.

Sinofsky: [Laughs.] I'll take that in the positive sense that I'm sure you meant it.

I think that all projects, all teams, all companies, have ways of planning and doing stuff. What happens is that some things work well in certain situations and some things don't. And so what we tried to do was pull together the lessons of what have worked for the organizations that I've been part of, along with Marco's history of studying lots of organizations. Here's a guy who for 20 years has been visiting companies, talking to them when their plans weren't working. Whether it's Toyota or Dell or Sun or the hundreds of companies that he visits, he's studying them like an anthropologist, talks to everybody. He finds out that there's this huge disconnect between the people on the assembly line and the people in the fancy offices. Why is that, and what to do? What we'd always tried to do on our teams is reduce that disconnect and reduce the friction between the plan and the execution. So a lot of what's written is really just that history.

The point of the blogs was to share and for me to open up a dialogue with people. Not for me to just write speeches. Nearly every post is the result of something that happened in the hallway, in a group meeting, in some email that somebody sent me. They're not just, 'OK, now it's Day 17, time for the post on planning.' Everything was this ongoing dialogue and this conversation.

TB: To what extent do you think people will be expecting a behind-the-scenes look at Windows 7's development from this book?

Sinofsky: Well, I suspect that people will either want there to be, or hope that there is --

TB: A Windows 7 gossip book.

Sinofsky: Of course there's just no chance that's going to happen.

TB: Why not?

Sinofsky: Because we're a business and we have our own boundaries in a way, and that's not a story that we need to tell. What we want is to be judged not on the gossip or the salacious details but the results of the software that we make. We're going to make software that people like, and we're going to make software that isn't as much as we'd all wanted, and we're going to keep trying, and doing a better job. So it isn't a book about that, at all. In fact, it has very little to do with Windows 7 in particular, or even the Windows organization in particular. It's really about building software and managing a large project. I hope that people are able to see past the desire for inside-the-beltway kind of stuff.

TB: I'll have to work on that with my book.

Sinofsky: Yeah, good luck on that.

TB: At the same time, people will probably try to draw from your book some insights into Windows 7's development. Is there a specific incident or example you can give from the Windows 7 development process that particularly exemplifies the lessons you hope people take from the book?

Sinofsky: There were a couple small things. One of the challenges that a lot of the product teams have is, when do you start talking about your product when you know it's not done yet? There's always this desire to get out there. You've got competitors doing things, you just want to get feedback. How does that meld with having a high-integrity relationship with your customers and your partners.

Windows has what I think is a fairly unique challenge, which is that it's really a part of an entire ecosystem of software and hardware and it can't really stand on its own. Without all the new PCs and without all the new devices, and the new software, Windows is not nearly as exciting. On the other hand, these are all companies, and they have their own plans, and their own budgets, and their own timelines, and so we can't announce something, and then slip it, and not deliver it, or change our mind. But on the other hand, we can't just surprise everybody. We can't show up and say, OK, so it's Oct. 20, new release of Windows in two days, is everybody ready?

TB: You're Steven Sinofsky, not Steve Jobs.

Sinofsky: Well, we have a different business model. We're not in control of every single element of the thing, literally from the components to the store where you buy it. We're actually just one part of the entire chain. So it turns out we have to go and tell Best Buy when we're going to be ready just as much as we have to go and work with Seagate or work with Intel, then work with software companies that are big and small, and then work with HP and Dell and Acer and so on. And so, in a way, we had to learn some of the lessons from past projects where we got out too early and talked about things. But how are we going to help the team understand this notion, and that's really where this notion of integrity came in. Because it was, how do you have a high-integrity relationship with all of those customers and partners.

One day, there were some (internal) blog posts about other topics, and all of a sudden some elements of one show up on some web site. Off the internal blog, which of course was supposed to be the internal blog. It's on a Sharepoint site internally, it's not on MSDN. That should be kind of a warning sign. It's on a server that looks like the Windows source code server. Nobody would think about posting the Windows source code. It was a very interesting moment, because it challenged me.

So I actually wrote a whole post on our partners and why we do all of this. It was really interesting, because it's the sort of thing that when people outside read it -- the whole thing, as opposed to the small parts of it that were leaked -- they'll go, oh, you know, what's not to agree with? Of course you don't want to go out there and tell people a bunch of stuff that then you later have to recant. So we're going to be deliberate about how we talk to people and what we say. And we're not going to play games.

This whole thing, "Oh, we're going to underpromise and overdeliver, because the old world overpromised and underdelivered." No. Why don't we just do the rational thing that we just promise and deliver? That became this whole meme for how we wanted to work as a team. It really encapsulates that notion of integrity and just getting done something that we say we're going to get done.

TB: Does that person still work in Windows?

Sinofsky: It was a leak. It wasn't like they printed t-shirts that said, "Hi, I'm the leaker."

TB: You use the phrase "dilithium crystals" in the book. I think I know what that means in the Star Trek context. What does it mean in software development?

Sinofsky: We were having this discussion in our team, constantly: What's the difference between having a technology focus and just saying, hey, let's write this cool piece of code, where cool is technically cool, it does some nifty technical thing, vs. hey, let's solve this end-user scenario, or let's solve this IT professional scenario. So there's this difference between scenarios and cool technologies. How do you bridge that?

For me, it reminded me of this story when I was little. I got this Star Trek technical manual in 1974 or 1975, and I opened it up, and I was like, this is so cool. It showed how to make a phaser. So I went to the electronics surplus store, and it's just bins of parts, and I'm literally walking up and down the aisles with my father buying things that I don't really understand. Oh, a resistor! It says to have a resistor. So I have this handful of these 3-cent parts. And then I'm like, well, now where's the dilithium crystals? It turns out, you need them for the phaser and the communicator and the tricorder. They all have dilithium crystals in them. It was this huge letdown that I had everything but dilithium crystals. And I left all dissapointed.

In a way, scenarios are a lot like that. There is always a magic ingredient to making a scenario complete, and you can't just have all the parts. You do really have to stitch them together. To make an end-to-end, get-your-photos-on-the-Internet scenario, you have to do all the steps, and you can't leave the person hanging. It could have been "batteries not included." It's the same kind of thing.

TB: Are you satisfied with the volume of dilithium crystals in Windows 7?

Sinofsky: That's a specific thing about the product, which has little to do with the book. What I would say about Windows 7 is, don't ask me, ask all the other people who have written about it. Because that's the much better judge. It's our product, I worked on it, so I'm not really objective about it.

TB: You're associated in some ways with what some might see as a turnaround of Windows. Some people may look at this book in that way. Do you see yourself in that way, and could you see yourself writing other management books?

No, I don't see that at all. Building software or just building things is what I like to do. In a way the book is just an offshoot of that. It's not my goal to do other things. I really wouldn't get into, did we rescue the product, or whatever. All products go through these lifecycles where you have releases that are great, you have products that are great, then you have ones that aren't as good as you had hoped. You have ones that exceed your expectations. You have surprising things that do really well that you weren't expecting.

So that's just part of the natural organic nature of product development. I think people ascribe too much certainty to product development. If product development were easy, there would be no reason to write books, or there would just be one book. I view our contribution as just that -- a contribution to the body of experience and knowledge in building products.

Editor's Note: For the record, as evidenced above, Sinofsky politely declined our creative efforts to raise topics that strayed too far beyond the book, including Windows 7, Windows 8, and his career aspirations.