The article's title refers to "code quality", but the word "quality" never appears in the body. The article is about the number of bugs, so I think a more appro

The article's title refers to "code quality", but the word "quality" never appears in the body. The article is about the number of bugs, so I think a more appropriate term is "software quality" (more specifically "software functional quality", per Wikipedia; but of course it would be clearest if the title just said "number of bugs"). To me, "code quality" means something very different (well-written, readable, maintainable, etc.; Wikipedia calls this "software structural quality"). According to this definition, "code quality" may be correlated with the number of bugs (hopefully with a negative correlation!) but the two are very different things. And I'd like to boost @kragen's comment (https://news.ycombinator.com/item?id=33566329#33579042 https://news.ycombinator.com/item?id=33566329#33579042, toned down): > haldar's summary is ... incorrect ... > the paper finds that faults (and wmc (methods per class), cbo (coupling between object classes), response for a class (rfc), number of methods added (nma), etc.), correlate to class size (non-comment source lines of code), not program size