NOTICE:
This content is in the "archives" of a retired blog called Gadgetopia. It has been moved to this subdomain as it is no longer considered relevant to the site. It is being hosted here for a indeterminate period of time. Its existence at this URL is not guaranteed, and it may be removed at any time. If you would like to refer to this content in the future, you are encouraged to save it to your local file system.

Some content from Gadgetopia was moved to the technical blog on deanebarker.net

The Perils of Developing Windows

Originally published by "dbarker" on 2006-06-15 14:34:00Z

The World As Best As I Remember It : Broken Windows Theory: This essay from a Windows developer (like, a guy who develops Windows, not develops on Windows) is a fairly telling look into the world of the Windows development group.

Some excerpts:

Vista has suffered a series of high-profile delays, including most recently the announcement that it would be delayed until 2007. The largest software project in mankind’s history now threatens to also be the longest.

[…] building Windows on a dual-proc dev box takes nearly 24 hours, and you’ll be slow enough to drive Miss Daisy.

[…] Deep in the bowels of Windows, there remains the whiff of a bygone culture of belittlement and aggression. Windows can be a scary place to tell the truth. When a vice president in Windows asks you whether your team will ship on time, they might well have asked you whether they look fat in their new Armani suit.

I don’t envy the people trying to manage that monster. And 24 hours to build? Holy cats.

It says something too that Microsoft allows this stuff to get posted. This essay is located under an MSDN domain name, so it’s eminating from somewhere within the Borg Cube. I never thought I’d see that.

Comment by "Chris" on 2006-06-15 16:31:00Z
Nice find, Deane. While not in a commercial programming field myself (aside from the occasional small SQL app to make my life easier), I've always been very interested in the inner workings of the industry. My mom was an OS/Unix programmer for Bell Labs back in the early 80's, and I remember her descriptions of how projects would slip and why. Things haven't really changed that much. It's funny how, given the tools we have today to simplify programming tasks, we just use them to make things more complicated. Here's a comment left on that blog page that I thought was very well stated: " Such enormous complexity will beget eternal fragility. It's good for Microsoft they've implemented a revenue stream that's based on continual, required security and quality improvements. It's in their best interests to ship a buggy Vista, and nobody has any doubt this next version of the same unreliable computing infrastructure will be foisted upon the world to deal with for the next decade." IMO, the Debian model for software/OS architecture is where the future of programming needs to go -- a single central kernel that interacts with the various components around it, and only the needed/wanted ones. The Microsoft model is like a manager who tries to accomplish every task himself without delegating anything, and in turn does only a mediocre job at best at getting everything done.
Comment by "Deane" on 2006-06-15 16:52:00Z
>a manager who tries to accomplish every task himself without delegating anything, and in turn does only a mediocre job at best at getting everything done. I am Windows.
Comment by "Mike" on 2006-06-15 06:57:00Z
The majority of the post was just removed by the author. >[I have removed the rest of this post of my own volition, without any external pressure whatsoever. What started as an opinion on the challenges of managing large software projects has turned out to be a rallying point for detractors, which isn't interesting or productive. - Ed.]
Comment by "Chris" on 2006-06-15 07:11:00Z
Bummer dude.