| 1 |
On Tue, 21 Apr 2020 19:58:03 +0300 |
| 2 |
Consus <consus@××××.net> wrote: |
| 3 |
|
| 4 |
> Github bot warns |
| 5 |
> you that contributing new packages to the main repo is low priority and |
| 6 |
> probably no one will help you, |
| 7 |
|
| 8 |
Maybe that's a misinterpretation. |
| 9 |
|
| 10 |
Gentoo workflow isn't oriented around Pull requests, Pull requests are |
| 11 |
generally a mechanism used to allow users and other non-developers to |
| 12 |
"assist". |
| 13 |
|
| 14 |
People who are dev's generally don't use PR's for their work, although |
| 15 |
there are a few exceptions to this because the PR facility allows for |
| 16 |
various "pre-testing" and "I'm not the guy in charge of this, so I |
| 17 |
better let the guy who is look at it first" situations. |
| 18 |
|
| 19 |
And somewhat as a result, contributions via the PR mechanism *are* low |
| 20 |
priority. Not because nobody is around to do it, but because either we |
| 21 |
have more important things to do, *or*, sometimes there are things we |
| 22 |
must do *before* handling PR's. |
| 23 |
|
| 24 |
Sometimes even people filing bugs for "please update this" serves no |
| 25 |
purpose, because we're probably already aware there's an update, its |
| 26 |
just a matter of time constraints and priorities. |
| 27 |
|
| 28 |
I would argue we're actually more on top of, and aware of, things that |
| 29 |
need to be done than users might imagine. Its just largely invisible |
| 30 |
because you have to go *looking* for it, and you have to actually |
| 31 |
communicate with dev's to really see what's going on, because |
| 32 |
sometimes, its all in their head. |
| 33 |
|
| 34 |
Just the perception from a portage consumption point *may* look like a |
| 35 |
thing is stagnated "because no recent changes". |
| 36 |
|
| 37 |
If you were to look at portage, you might think no work is happening on |
| 38 |
the rust front, beyond the language itself being there. |
| 39 |
|
| 40 |
But lots of research and experimental work is happening on overlays, |
| 41 |
working out how to do a thing right, before burdening users with it. |
| 42 |
|
| 43 |
( And there's also the fact that properly reviewing a PR often requires |
| 44 |
the same work investment to merely verify it, as if you were to do the |
| 45 |
work yourself without a PR, which serves as a bit of a disincentive to |
| 46 |
care about PRs ) |
Archives