1 |
>>>>> On Sat, 13 Oct 2018, desultory wrote: |
2 |
> On 10/11/18 13:35, Ulrich Mueller wrote: |
3 |
>>>>>>> On Thu, 11 Oct 2018, Alec Warner wrote: |
4 |
|
5 |
>>> My reading of ulm's proposal is that it is allowed. |
6 |
|
7 |
>>> Ebuilds "shall" use the simple attribution, not that they "must" use it. |
8 |
|
9 |
>>> To me that implies the simple attribution should be the default, but the |
10 |
>>> complex attribution is acceptable in the ::gentoo repo. |
11 |
|
12 |
>>> Maybe I'm misunderstanding the proposal? |
13 |
|
14 |
>> No, you've understood it exactly how it was meant. |
15 |
|
16 |
> Especially given the audience, "should" would convey that intent more |
17 |
> clearly than "shall". [1] |
18 |
|
19 |
> [1] https://www.ietf.org/rfc/rfc2119.txt |
20 |
|
21 |
| 3. SHOULD This word, or the adjective "RECOMMENDED", mean that there |
22 |
| may exist valid reasons in particular circumstances to ignore a |
23 |
| particular item, but the full implications must be understood and |
24 |
| carefully weighed before choosing a different course. |
25 |
|
26 |
LGTM, especially the part that the implications must be carefully |
27 |
weighed before ignoring the policy. |
28 |
|
29 |
Ulrich |