I had to chuckle this afternoon when Steven Sinofsky exploded upon the Internet with one of his missives describing the Windows 7 Team on the Engineering Windows 7 blog.
Steven has been blogging on the internal network since late 2006, and his blog posts have never been accused of being brief. In fact, I keep expecting him take his blog and publish it as a book (The Tao of SteveSi). It’s already over 300 pages long and about 140,000 words, so he’s got a great start.
Personally, I really appreciated the level of detail that Steven always went into about his process and his decisions when I was in that “Internet Explorer” feature team that he mentions in his list of Windows feature teams. In this respect, Steven raised the bar over his predecessors (and every other senior management that I’ve ever worked for, with or around).
Windows under VPs like Jim Allchin and Brian Valentine was a cowboy-culture – ambitious, technology-driven and always brought to the finish-line by blood, sweat and tears (to be fair, that included the VPs who didn’t spare themselves the long nights and hard work). This culture successfully delivered 6 or 7 releases of one of the largest and most successful software projects on the planet. But that couldn’t last. It was inevitable that the size of the project would outgrow this method of running a project.
Steven came to Windows from running the Office organization where he built a reputation for running a tight ship and getting the product out on time with the right set of features. Interestingly, from my standpoint outside of Office, I always saw the Office team as a team I’d never like to work on – it seemed stodgy where Windows seemed fluid; it seemed dictatorial where Windows seemed collaborative; it seemed boring where Windows seemed exciting.
Down in the troops, when Steven was named to take over Windows, there was, shall we say, trepidation. But Steven took the smart route. He spent the last few months of the Vista project quietly going around talking to people. And when he moved, he moved decisively – making sweeping organization changes that challenged the traditional structure of the Windows organization. Clearly, such changes might be difficult for the team to accept.
What made all of the difference is that Steven started blogging and he blogged regularly. His blog explained in sometimes excruciating detail the tradeoffs in his decisions and why he chose one way over another.
The bottom line was – you could disagree with him, but you couldn’t say that there wasn’t a well-thought out method to every decision he made. He showed clarity and thoughtfulness in his actions. He showed that he wasn’t making changes for the heck of it, but that there was a reason why he did everything. His thinking showed a long-term perspective on the product that had been completely missing from the previous management.
His blog enabled him to communicate more clearly and widely about these decisions to those who wanted to understand. It gave people a place to voice concerns and ask questions (the true open door policy in a company where going to his actual office would require a 15 minute trek across campus). Not everyone read it, to be sure (and even those who wanted to were often daunted by the length), but if you wanted to know, the information was there to be had.
In large part, that blog is why I accepted a position in his old team, Office. His development methodology is a proven system for how to bring a thousand-strong development team together while creating the freedom to develop and ship the features and ideas that I want. It may not be perfect, but it’s better than anything that the previous management of Windows had ever done.
I’m fascinated to watch the next step in this journey. I’m not completely convinced that his style will work in a public forum, but I’m glad that he’s doing this.