You can add location information to your Tweets, such as your city or precise location, from the web and via third-party applications. You always have the option to delete your Tweet location history. Learn more
I've wondered about this. Anecdotally, the place I've worked with the best quality didn't do code review (maybe three "serious" user-visible bugs during the 8 years I was there, one of which was a fab issue that couldn't have been caught with any amount of code review).https://twitter.com/skamille/status/1169765800829435904 …
I know of a few places where people say that code reviews are the reason they have such great code quality (I've even worked at one of those places), but AFAIK they all ship a lot more bugs while moving slower.
There is actually research about this. Steve McConnell reports modal defect removal efficiency of 35% for design reviews; 55% formal design inspections; 25% (i. e. LOW) code reviews; 60% formal code inspections; 65% modelling or prototyping; 30% unit test;
35% integration test; 25% regression test; 40% system test; 35% low-volume beta test; 75% high-volume beta-test. Reported in "Software Estimation" book (2006), sources: Jones 1986, Jones 1996, Shull et al 2002
I found they were mostly “useful” for nitpicking coding style and how the reviewer would solve the problem differently. Sometimes bugs or edge cases were thought up but it wasn’t super common. I think tooling with rules enforcement would be more valuable.
I think review is most effective either as a post-commit communication tool, or pre-commit when the author feels it makes sense. I also think a lot of folks at that CR place forget that their incoming reviews are blocking others, which sabotages the whole team's velocity.
how do you know how many bugs they would be shipping without code reviews?
Our team does not do review for "catching bugs." We have regression tests (unit, selenium, postman) for that. If cr catches a bug, it's an unexpected benefit. No, code review is about growing the team and situational awareness.
Since code reviews are a common mentorship technique, how did the place without code review grow junior engineers?
Reviewing code as a way of finding bugs is the same as writing code and calling it “implemented”. — Software quality is property of the deployment. Not a property of the source code.
And pair programming in your experience ?
What do you think is the reason of high quality at the place you worked earlier
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.
