I had an interesting conversation today at a party. A neighbor is a technologist responsible for change control for IT projects in the conservative oil and gas industry. Though our backgrounds are dramatically different, we found significant common ground:
- Productivity rules. If you want to build effective software, you’ve got to build it quickly.
- Fixed bid, fixed date, fixed scope contracts rarely work. He didn’t call it agility, but he was fundamentally using an agile process.
- The only successful measure of a project is the value you deliver to the business. Migrations for the sake of migrations, and big-bang projects (like massive Web Services projects, SAP installations spanning years, and the like rarely ever work.
- Small teams rule. If you want to build great software, find great talent, or build teams from within. Overpay if you have to. You’re not really overpaying…you’ll never pay a top programmer 20 times what you’d pay a poor one, yet the top programmer is easily 20 times as productive. (I’d even be willing to bet that poor programmers on many projects are a net negative.)
- Leverage rules. It’s not nearly as important to build with popular technologies anymore. You need to build with the most productive technology that will work for your project. Sometimes, as with my practice, the leverage comes from the languages and frameworks you use. Sometimes, as with his practice, the leverage comes from technology developed for a business application.
It was a fascinating conversation. I never expected to find so much in common.