1 |
On Mon, Jan 3, 2022 at 7:39 PM Sam James <sam@g.o> wrote: |
2 |
> |
3 |
> |
4 |
> |
5 |
> On 3 Jan 2022, at 17:16, Alec Warner <antarus@g.o> wrote: |
6 |
> |
7 |
> [snip] |
8 |
> |
9 |
> |
10 |
> I'm trying to understand your principles here. Like on what basis do |
11 |
> you remove or add flags (in general). |
12 |
> |
13 |
> I want to remove: |
14 |
> - bash-completion |
15 |
> |
16 |
> |
17 |
> FWIW, I've managed to remove basically all instances of {bash,zsh}-completion |
18 |
> and made upstream PRs (all of one of which have been merged!) for fixing |
19 |
> `./configure` behaviour accordingly. |
20 |
> |
21 |
> This is in align with our small files policy: https://projects.gentoo.org/qa/policy-guide/installed-files.html#pg0301. |
22 |
> |
23 |
> - acl |
24 |
> - ldap |
25 |
> |
26 |
> |
27 |
> ACL is kind of similar to what Ionen said for PAM, i.e. sometimes |
28 |
> people may want to turn it off and it makes sense to expose |
29 |
> this option for those who do, but we don't need to try support it. |
30 |
|
31 |
I feel like 'acl' or 'pam' or 'policykit' are not really USE flags in |
32 |
the general sense. They are more like profile settings. You don't want |
33 |
to toggle these flags, you want to switch profiles. I think |
34 |
conceptually profile migrations are larger changes that require a |
35 |
rebuild of a bunch of stuff, and typically have downtime (e.g. where |
36 |
your system isn't expected to work the entire time.) |
37 |
|
38 |
There are practical problems with profile proliferation, but I think |
39 |
it is closer to what these flags represent, if that makes sense. |
40 |
|
41 |
-A |
42 |
|
43 |
|
44 |
> |
45 |
> I think some of this kind of comes back to "how do we better |
46 |
> make clear what is/isn't okay (supported)to customise?" |
47 |
> |
48 |
> LDAP is a fun one because IIRC we had it enabled by default |
49 |
> for too long and it didn't really break anything when we turned |
50 |
> It off. |
51 |
> |
52 |
> Overall, I think we kind of come back to this idea of |
53 |
> trying to just set better IUSE defaults rather than |
54 |
> in profiles, so that it's per-package where possible. |
55 |
> |
56 |
> - policykit |
57 |
> |
58 |
> |
59 |
> This is a reasonable flag to keep given the heavy polkit |
60 |
> dependency of spidermonkey (for now) and it's somewhat |
61 |
> heavy-handed as a concept anyway. |
62 |
> |
63 |
> - readline |
64 |
> |
65 |
> |
66 |
> readline is a tricky one because of its relation with libedit too. |
67 |
> |
68 |
> - sound |
69 |
> |
70 |
> (Part of this is just to have a meta discussion so we settle on some |
71 |
> driving principles on why we keep one flag over the other.) |
72 |
> |
73 |
> I can easily craft a narrative for getting rid of ipv6, for example, |
74 |
> but I cannot really craft a good narrative for getting rid of pam, or |
75 |
> policykit, or ldap as flags. So why do we keep some and remove others? |
76 |
> |
77 |
> |
78 |
> |
79 |
> It's very hard to quantify :( |
80 |
> |
81 |
> |
82 |
> Best, |
83 |
> sam |
84 |
> |