[Nix-dev] On nixpkgs and reasonable code size

[Vladimír Čunát]

On 02/22/2016 10:35 AM, Adrien Devresse wrote:
> The way the other distributions solved this issue is by distributing the
> responsabilities. A model where a little minority has to approve /
> validate the merge each package modification can not scale to a huge
> number of packages.

IMHO the discussion has shifted mainly to things that IMO don't pose
much of a problem in nixpkgs. It already is the case that each
contributor focuses on some subset of nixpkgs, in some cases only a
handful packages. It's just too large not to have any focus in there. We
trust each other that we shan't vandalize the rest (and we have all the
history and tools for emergency cases).

The control isn't enforced, but if we tried that, the whole thing shifts
to actually deciding who has push access to which parts, which is a far
more time-consuming task... all this trying to prevent misconduct that
I'm not aware of ever happening yet.

Some (soft) responsibility is what `meta.maintainers` is for (partly).

The number of people with push access isn't that small - 70 ATM. Many of
them aren't active (anymore), but on the other hand we've got quite a
few active meta.maintainers that take care of their packages through PRs.

--Vladimir


-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 3771 bytes
Desc: S/MIME Cryptographic Signature
Url : http://lists.science.uu.nl/pipermail/nix-dev/attachments/20160223/28721c73/attachment.bin