Submitted by alastairs on 15 June 2011 - 10:00pm
Today I spent most of the day pairing on a task with another developer. This is unusual for our company (the usual arguments fears are raised) but we both found it a useful experience. Here's what I learned today.
As part of the pairing scenario, I suggested using a ping-pong test-first approach:
Submitted by alastairs on 22 November 2009 - 1:13pm
Submitted by alastairs on 24 January 2009 - 2:21pm
One of the reasons progress on my web frameworks' evaluation has been so slow of late is that my last check-in broke my CCNET automated build, and, as the good programmer I am, I felt that this needed fixing before I made any more progress.
I'd written some unit tests for the PostController (good) that were accessing the database directly (bad), and for whatever reason the build machine couldn't successfully connect to the database. A classic case of "works on my machine!", and, as is usually the case with such things, a good example of poor design.