'

Who needs another Windows 7 beta?

My ZDNet colleague Adrian Kingsley-Hughes is not too proud to beg to Microsoft: Can we have a Windows 7 beta 2 please?Once upon a time, I might have agreed with him. But now that I’ve had a chance to see how this beta cycle works, I think Sinofsky and Co. are making the right choice. One beta, one release candidate, and then ship it. Here's why.

My ZDNet colleague Adrian Kingsley-Hughes is not too proud to beg to Microsoft: Can we have a Windows 7 beta 2 please?

Once upon a time, I might have agreed with him. But now that I’ve had a chance to see how this beta cycle works, I think Sinofsky and Co. are making the right choice. One beta, one release candidate, and then ship it.

Why not have another beta release?

At the top of my list of reasons is the disruptive effect it has on the development process. Any build that is released to the public has to be locked down and then tested for several weeks. That’s a lot of dev and test hours spent on something that’s going to be thrown away.

But won’t an extra beta release turn up more bug reports? Yes, but more reports doesn’t mean more bugs. Many are duplicates of existing reports, others are disagreements over a design decision and will be resolved as “Won’t fix” or “by design.” The current beta release is essentially feature complete; a new build won’t expose any additional features.

The one example that Adrian uses is the recent dust-up over the implementation of UAC. In that case, the current beta did exactly what it was supposed to do, providing plenty of fertile testing ground and creating an opportunity for what diplomats like to call “a full and frank exchange of ideas.” Another beta won’t offer anything additional except perhaps an opportunity to have another debate over the implementation of that feature.

I’ve heard Adrian’s plea from other longtime beta testers, who are mourning the loss of an old Microsoft tradition, wherein a privileged few get access to interim builds not made generally available to the public. Sorry, but those days are gone. Thanks to the miracle of BitTorrent, any release that goes to more than a literal handful of trusted partners might as well be a general release.

The other big, perhaps inadvertent, benefit to the current development schedule is that it allows people like me time to actually use the product. In previous Windows test cycles, I spent half my time installing and configuring new builds, and after a week or two it was time to repeat the cycle. For Windows 7, I’ve had a chance to actually get to spend time with what’s changed (and to identify some annoyances that are still unfixed – more about that later this week).

Maybe, from a pure PR perspective, Microsoft would have been better off assigning the Beta 1 moniker to its PDC release last October and then calling the current release Beta 2. That would have been a pretty good mapping to the way things used to work, and it might have headed off some of the criticism.

With the current roadmap, Microsoft has set the bar for its release candidate very high indeed. If they blow it, there will be no shortage of critics ready to argue that another beta or two would have made things better.