When I was growing up, the family business was a Ford dealership. One of the bits of unwritten wisdom then was to never buy a car that was built on a Monday (or during labor negotiations), as they tended to suffer from more problems than cars built mid-week. (The image was of hung-over factory workers returning to work after weekend-long benders, picking up tools and then trying to assemble your car.)
I'm wondering if there's a similar effect with code. That is, do defects tend to correlate with the day of the week (or the day into the project) when code was written? I'll bet there are some interesting patterns there, should someone care to do the legwork. I've taken it as far as musing about whether there's enough information gleanable from CVS to timestamp each row of a file. At that point, this joined the list of "neat projects that'll never get off the ground in this lifetime."
Or is it just progressive slacking ? hmm...
Re:RT Stats
jdavidb on 2003-09-23T19:40:31
The bugfixes all introduced faults that were discovered on Friday.
:P ;) Unless, of course, you're properly testing your code and not releasing unless all tests pass.
:)
