Where the bugs are

I've only skimmed the original paper, but this article is intriguing--"...in their study, the top 20 per cent of the largest programs contained over 60 per cent of the bugs."

I've felt like longer functions are more bug-prone per line of code than shorter ones, and this study seems to confirm that intuition.


  1. The result is only interesting if top 20% largest programs represent significantly fewer than 60% lines of code. I only skimmed ... did the authors address that point?

  2. My impression was that they *were* saying that bug densities in larger programs are higher. I'd have to read it more carefully, I guess.