Hi,
Consider software applications in general. Any
application has
a) a number of bugs which exist in the
code.
b) a number of bugs which will actually be
encountered in the lifetime of the application. I.e., this is the number of
bugs which a user/customer actually sees and hence may report.
Any ideas/opinions/guesses on what the ratio of
these two bug counts might be, generally speaking ?
Any pointers to studies, theories, models, ...
?
I know very well that this is a vague, general
question, with no "right" answer. But I hope it's reasonably clear
what I'm shooting for anyway and I would appreciate your thoughts on this. The
question should be of real-world importance when trying to estimate the value of
a tool based on how many bugs it found in some application (since many of
the found bugs would never have shown up anyway in real use).
Thank you very much!
- Henrik
|