BS-meter goes off scale, team brought in to rewrite 60% of BS-meter code.

March 25, 2006

I nearly lost my swill of coffee yesterday morning seeing how “news” outlets and pundits alike all just took a small website’s word for fact that Microsoft was rewriting 60% of Windows Vista. Now, now… anyone with a small amount of project management sense knows if you take a swag of the # of developers in Windows org and the # of lines of code that would need to be rewritten, then factor in the amount of time left to do this and QA work required, it’s a silly proposition. So, my BS-meter went way off scale and broke. This should be a reminder not to believe everything you read on the Internet.

Oh and I’m disappointed in a certain member of the digerati who appears to baiting Robert Scoble over the 60% rewrite (and others), calling into question Robert’s credibility and instincts (by the way, the saying is “reap what you sow”, not “sew”). I used to sing this person’s praises. Sour grapes because he’s upset he wasn’t invited to a meal with Bill Gates? Possibly. Trying to rattle cages? My read: Yes. Unsubscribed? Yes.

As for the annnouncement that consumer Windows Vista will ship in January, yes, I was disappointed too. But at the end of the day, I’m a cup-half-full kinda guy. What this tells me is that our management realizes just how important quality and security is in this release over near-term revenue and the pressures of the marketplace looking for a bump this holiday season. I know there are a lot of frustrated people right now, but I have to give some credit to those that are holding the line.

4 responses to BS-meter goes off scale, team brought in to rewrite 60% of BS-meter code.

  1. 

    "What this tells me is that our management realizes just how important quality and security is in this release over near-term revenue and the pressures of the marketplace looking for a bump this holiday season."

    Doesn’t it also tell you that though the Windows Vista team has project management issues? They have a timeline laid out for the entire project with milestone dates that they obviously intended to make. Now, the team has shown they’ve been unable to hit their goals by the scheduled dates. Either the original timeline was overly optimistic and someone screwed up there, or communication from the guys writing the code isn’t reaching the right people when they relay that the milestone dates are unachievable.

    I certainly don’t see where missed due-dates tell anyone that management has now become enlightened about quality and security. I would have thought those items were written into the schedule right from the start.

  2. 

    Ahh yes, this topic is covered in, "The Art of Project Management" by Scott Berkun. Net-net, schedules set at the onset, middle, and near the end of a project have wildly different levels of accuracy. If this were the first time a schedule was incorrect, I’d agree with you. I’ve seen projects get rushed out the door to hit a retail release window- not this one. The OS is arguably the most complex piece of software on any PC and sometimes, the schedule doesn’t work out. It’s happened to almost every OS (or game console) ever shipped.

    As for your last comment, the key point here isn’t some new enlightenment, it’s the reassurance and transparency that comes from the fact that they’re holding the line in the face of significant pressure to ship earlier.

  3. 

    This is a very good proposal!