This past week was a blur – lots and lots going on. A few highlights:
I had lunch with Robert Scoble and had a great talk about the power of blogs in building community. We’re in the midst of planning more frequent drops of Whidbey bits, and two of the things we’ll likely do are get the word out informally re: quality/problems/expectations via high visibility blogs people are already reading (e.g. Brad Abrams, Sara Williams, etc.), and provide some sort of Whidbey blog aggregator that provides a semi-official voice from the teams within DevDiv re: what problems people can expect, any workarounds, etc.. This does not mean that we’ll turn these blogs into mktg spam, but just encourage these folks (who would probably do it anyway) to feel free to comment on the builds.
Also on the subject of more frequent build availability, we’re looking at how to make it easier for people to report and track bugs. For the near future, we’ll continue to rely on bug reporting through betaplace, even though betaplace has a poor reputation within our division – it’s too manual and customers often have a bad experience waiting for days or weeks for their registration to go through. We’ve limited access to this tool largely due to the limitations and because the registration experience is so poor in an age where instant access is standard. In the near future we’ll push those limitations some and open it up more. Further out (meaning months) we’re going to have a much more open process that enables more collaboration re: bugs/feedback between our development teams and customers, and even between customers.
There was a Jim Allchin review last week of the community efforts across his organization. It’s clear we have a lot of work to do, but we’re making progress in some areas. I almost caused permanent damage to my career by implying a ship date for Whidbey that was later than his expectation – I said we are looking hard at customer-reported bugs that have been postponed many months before we ship (the point being: what kind of message does that send about whether we value the feedback?) and didn’t explain “many months” very well. After the mtg stopped and Jim said “what? WHAT?” a couple times I figured out what he had heard and corrected the problem – I was referring to some bugs people reported last July that got postponed. I’m still here, but I assume I’m on double-secret probation now 🙂
Speaking of Jim, this is my second stint at MS. In my first stint here I got to work on Windows1 as a developer. I remember a time when I linked my Windows application directly into Windows and it all ran in 64K of memory. (Hopefully some Monty Python fans are thinking “You were lucky! My family grew up in a shoebox.”) At some point in the OS/2 days, I switched over to program management, and ultimately was the group program manager when we shipped NT3.1. I worked briefly in the Cairo team, and left pretty soon after that. After 7 years of doing some interesting, non-software-related things, I came back in early 2001 to work with Mark Lucovsky on HailStorm. (There’s a story). I moved into Visual Studio about 18 months ago, ran the program management team in the smart devices team (VS for PPC / CE-based devices), and moved over into this job a couple weeks ago. It’s a very big company compared to when I started, and one of the things that attracted me to this job is the the big disconnect between public perception of MS and the fact that people who work here do care about our customers succeeding. People have always cared, but in growing big we lost some of the connection we had with developers. We want to strengthen the connection and enable our customers to be more successful.