| 1 |
On 10/01/2017 08:12, Daniel Campbell wrote: |
| 2 |
> All goes well until "oops, this new guy pushed an ebuild without an |
| 3 |
> EAPI" or "the new guy didn't update the keywords for the version bump so |
| 4 |
> now it's considered stable and shouldn't", or "the new guy didn't run |
| 5 |
> repoman and now their metadata.xml doesn't adhere to the schema". |
| 6 |
|
| 7 |
I'd expect that to be covered by a number of non-trivial PRs. |
| 8 |
|
| 9 |
Similarly, this is something I'd be looking for when deciding whether to |
| 10 |
support someone's nomination for developer. If the evidence is not |
| 11 |
conclusive, I'd ask for more nontrivial changes, or maybe the quiz. |
| 12 |
|
| 13 |
> The quizzes serve as a) notes for future reference, and b) proves that |
| 14 |
> an individual is willing to research to solve their problem. You're not |
| 15 |
> expected to know the answers to the quizzes when you first take them. |
| 16 |
> The process of research exposes you to more information about Gentoo's |
| 17 |
> inner workings, including the PMS, which is a valuable resource whenever |
| 18 |
> you come across a weird aspect of ebuilds. |
| 19 |
|
| 20 |
Right, the quiz question often cover some corner/tricky cases which one |
| 21 |
might not otherwise encounter. |
| 22 |
|
| 23 |
Even devs after quiz make mistakes though, and it's not obvious when we |
| 24 |
reach a point of diminishing returns, where it'd be easier to quickly |
| 25 |
fix an occasional minor mistake, and have more contributions overall, |
| 26 |
than deal with under-staffing issues which also have consequences in |
| 27 |
some bugs not being solved at all. |
| 28 |
|
| 29 |
Paweł |
Archives