Scalability is easy, provided you don't need it to work. Probably the number one failure of system scaling is when people dive right in and start building. No baselines, limited measurements, no analysis, just a hypothesis and a whole lot of late nights tweaking the system.

At my last job we had a non-functional attribute that another team used to decorate service methods that they consumed. The other team was working on an alternative client to our WCF services, and they weren't on the same release schedule they needed t…

One of the tricks I picked up from my last job (and our forum software, now that I think of it) is the idea of user emulation. I could log into the application, search for a user, and, at the push of a button, temporarily become that user. The only differences between emulating them and actually logging in as them were a black bar that indicated who I am (with a link to stop emulating), all audit records continued to reflect my own user id, and I didn't need to keep track of 30 different sample accounts and passwords.

Over my years in (and before) IT, I've seen long projects, failed projects, confused projects, wildly successful projects, and even fun projects that ended far differently than we expected. The consistent take-away for me is that I am a big picture type of person, and that understanding that big, abstract picture cuts out a lot of wasted time sprinting down the wrong paths.

For the last few months I've had the pleasure of working with NDepend version 3. Most of my development at home is on linux these days, so I haven't used it as much as I'd like, but I have been using it to poke around in various codebases and see what…