Friday 23 May 2008

Vista Hurt By User Feedback?

Well things are quite quiet on the tech news front which gives me the opportunity to look at the Windows 7 debate again. I read Ed Bott's artcle on Windows 7 here. He makes a two points regarding beta cycles and feedback that I don't agree with.

[Update: After re-reading the post it appears that I misread it. I agree with what's being said in these points. Suppose that's what happens when I post after a late night]

Long beta cycles make better products. Oh really? If you count the infamous “Longhorn reset,” Windows Vista had arguably the longest beta cycle in the history of software development, with tens of thousands of outside testers. And look how well that worked out. Getting advance access to new Windows releases might make some outsiders feel like insiders, but it doesn’t make for a better product.


Users need time to give feedback about design decisions. Microsoft is getting plenty of feedback about the design decisions it made with Windows Vista. I don’t think there’s been any shortage of suggestions on what needs to be fixed in Vista, do you? Presumably, that feedback is being incorporated into components and features of Windows 7, including User Account Control, Windows Explorer, the Network and Sharing Center, and Internet Explorer. But there’s a cold, hard reality with all those design decisions: You can’t please everyone. One of the weaknesses of the Vista beta cycle was that the UI designers kept changing things up until the very last minute. For Windows 7, they need to get the design right (or nearly so) the first time.

I suppose this comes down to how much of a failure you believe Vista is but for those who do believe Vista failed to deliver I ask them to remember that the features you saw in the final release is only a small subset of the features that were supposed to be in it. It was the large amount of public [Update: Not public, private testing] testing that told Microsoft that the new features were too complicated and too far from what the users understood and Microsoft had to remove them. The Vista that would've been released if it hadn't been for the comprehensive testing and feedback might've been a little more complicated but it certainly would've been feature rich and different enough from XP to make it a must have. Microsoft seemed to be saying at the time that the features that were removed would be slowly added over the next few OS versions to make the transition easier on the end user. I don't think added user feedback would help Windows 7, actually the opposite.


  1. You're misreading what I wrote. Go back and read the whole post in context and you'll see that I started that list by saying "I think there’s something healthy about undoing some of the old assumptions about how Windows should be developed. Here are three of those assumptions I won’t miss..."

    Those two comments (out of three) that you quote are things that I think USED to be true and aren't any more. So I DON'T think users need more time to give feedback, and I DON'T think long beta cycles make a better product.

    By the way, the big cuts in features were done well before the public ever saw a beta, in 2004, when it became obvious to management that the development effort was completely off track.

  2. @Ed Bott: I don't know how I managed to misread the post but it certainly appears that I did. I actually agree completely with what you're saying and it is good to see the back of these old assumptions. Suppose we all make mistakes.