| 1 |
On Fri, 06 Jan 2012 10:31:14 -0500 |
| 2 |
Ian Stakenvicius <axs@g.o> wrote: |
| 3 |
|
| 4 |
> -----BEGIN PGP SIGNED MESSAGE----- |
| 5 |
> Hash: SHA256 |
| 6 |
> |
| 7 |
> On 06/01/12 04:28 AM, Duncan wrote: |
| 8 |
> > In it he said he wished he'd introduced systemd with everything on |
| 9 |
> > /usr in the first place. He sees the mistake now, and prefers to |
| 10 |
> > correct it while the fewest gentoo users possible are yet using |
| 11 |
> > systemd. As few are using it at this time and that number is |
| 12 |
> > likely to grow over time, only making the transition harder, he's |
| 13 |
> > moving much faster with the change than he'd ordinarily move. |
| 14 |
> |
| 15 |
> If that's the case, why not force the breakage now and not install a |
| 16 |
> symlink?? |
| 17 |
> |
| 18 |
> Maybe make this particular systemd update an interactive ebuild if |
| 19 |
> it's replacing an older version of itself, so it forces users to |
| 20 |
> actually stop, read, and take action when they do the upgrade? |
| 21 |
|
| 22 |
What if user thinks he/she's got the init= update right while, say, |
| 23 |
forgets to call 'lilo'? Forcing the breakage would mean systemd will |
| 24 |
fail to boot. Delaying it will just make user see 'you didn't update |
| 25 |
init=' one more time. |
| 26 |
|
| 27 |
-- |
| 28 |
Best regards, |
| 29 |
Michał Górny |
Archives