Engineering Practice

The root cause of most engineering disasters are information and incentive problems. These posts are about the techniques that hold up under that realization — how to debug methodically, how to ask questions that get useful answers, and how to think about the economics of how software gets built.

Read in this order

• Effective debugging: the observe-hypothesize-test cycle

  The methodical approach to non-trivial bugs, lifted from how researchers actually work.

• Combating information failure during software development

  The framing piece. Why so many "process" problems are really information-flow problems in disguise.

• 4 steps to effective development questions and bug reports

  Short and tactical. Do the homework that makes a useful answer possible.

• The MBA guide to software development

  How to think about the cost, value, and risk of software work in business terms.

Everything else

Full chronological list: Software Engineering category.