1 |
On Sat, Feb 13, 2016 at 4:47 PM, Raymond Jennings <shentino@×××××.com> wrote: |
2 |
> The guys making the API change bear the burden of fixing anything it breaks, |
3 |
> however, if something gets officially deprecated, don't go out of your way |
4 |
> to support continued use. |
5 |
|
6 |
We tend to do this already for things like PMS, which is as close as |
7 |
Gentoo gets to something like the kernel API. |
8 |
|
9 |
However, sometimes a gradual transition doesn't always make as much |
10 |
sense, and Gentoo doesn't always have the manpower to make every |
11 |
change a pretty one. |
12 |
|
13 |
And there is a cost to maintaining that kind of backwards |
14 |
compatibility. For example, debian chose to keep its LSB init scripts |
15 |
and write systemd unit wrappers around them. If they had chosen |
16 |
openrc instead I wouldn't be surprised if they kept the LSB init |
17 |
scripts and wrote an openrc compatibility layer around that. While |
18 |
this does provide a more stable experience, it also leaves around a |
19 |
ton of cruft. |
20 |
|
21 |
In general I tend to favor a balance. Trying to get everything just |
22 |
right was why the git migration literally took years, and even that in |
23 |
the end had a few bumps. Gentoo users need to be willing to deal with |
24 |
the occasionally bump in the road - we try to provide a fairly cutting |
25 |
edge experience, with minimal layers in integration. |
26 |
|
27 |
But, there is nothing really wrong with your suggestion, and we try to |
28 |
accommodate that approach when we can. |
29 |
|
30 |
> And yes, the personal attacks probably should die down. |
31 |
|
32 |
++ |
33 |
|
34 |
-- |
35 |
Rich |