dld@F.GP.CS.CMU.EDU (David Detlefs) (02/09/89)
I am trying to find references to empirical work that classifies the sources of errors in programs. I would like some evidence on just how much program reliability is enhanced by automatic storage management. I can find a lot of anecdotal/wisdom-of-the-elder evidence saying that improper explicit storage management is the source of many nasty bugs, but I'd rather find something more scientific. Ideally, the abstract of such a paper would say something like: "We instituted strict bug reporting and tracking procedures on the HOOFDEWHO project, written in the C programming language. An analysis of this data indicates that x% of all bugs are caused by improperly free'd storage. Finding these bugs took y% of all debugging time, and z% of bug-correction time." Any leads to anything remotely like this would be appreciated. Thanks! -- Dave Detlefs Any correlation between my employer's opinion Carnegie-Mellon CS and my own is statistical rather than causal, dld@cs.cmu.edu except in those cases where I have helped to form my employer's opinion. (Null disclaimer.) --