Re:The Epistemology of Software Quality
by hn_user82179 on Friday April 28, 2023 @12:35AM
(#35737508)
Huh, how was the reputation determined? Was it like a quantified & automated thing, or more casual like, a senior developer decides you need to do more testing since you’ve broken too much stuff.
Re:The Epistemology of Software Quality
by giovannibonetti on Friday April 28, 2023 @08:48AM
(#35740649)
Given the thousands of developers there, I would assume it was automated. FB is known for having built great tooling for this kind of thing, like measuring how flaky each test is.
Re:The Epistemology of Software Quality
by disgruntledphd2 on Friday April 28, 2023 @08:51AM
(#35740669)
I think that it was originally up to the release engineers, but they probably automated it after they moved to continuous deployments. (chuckr, where are you now?).
I learned about this from outside, but my understanding was that negative reputation was earned when someone from the deployment or development team pointed to your code as being the reason a patch had to be rolled back. Presumably because it broke during CI on the master/release branch. Not sure if there were other was to earn/lose reputation points.
That said, processes and technologies don’t have to treat each employee as a newbie. Example: I remember Facebook had a deployment process which gave each developer a reputation. Getting a reputation for deploying code that requires hot fixes to revert broken code meant penalties. Penalties required additional pre-merge / pre-deploy reviews and proof of testing. I remember that as a good combination of treating developers as adults but also providing targeted procedures when they are needed.