1 |
On Mon, Sep 30, 2013 at 8:54 AM, Daniel Campbell <lists@××××××××.us> wrote: |
2 |
> I'm not affected by anything regarding the /usr switch, but I'd like |
3 |
> to have a good talk with the first person who decided a |
4 |
> system-critical binary belonged in /usr instead of /bin or /sbin. |
5 |
> They've created a mess for every distro and any project that depends |
6 |
> on their work. |
7 |
|
8 |
(sorry for the previous post, accidentally clicked somewhere onscreen) |
9 |
|
10 |
As I've pointed out before: |
11 |
1) "system-critical" is actually dependent on the system. A system dependent |
12 |
on an smb share will find smbmount system critical. One dependent on |
13 |
zfs-fuse will find fuse system critical. With the advent of fuse, |
14 |
some filesystem |
15 |
that depends on an arbitrary user program will find that system-critical. |
16 |
While this works for for 99.(99?)% of user systems out there, FHS |
17 |
is supposed |
18 |
to be targetting all of them, and so it fails in principle in that respect. |
19 |
I remember making a lengthy thread on this mailing list challenging how FHS |
20 |
defined this and it appeared that nobody could make a defense. |
21 |
2) the reality is, it's not just binaries even. There are some things |
22 |
that binaries |
23 |
depend on, that in theory should be in /. For example, the hwid database, or |
24 |
libraries. Libraries make for a complex problem, because /usr is supposed to |
25 |
be network-sharable. Any libraries your programs depend on can't simply just |
26 |
be pushed to /, because then there'd be the chance that the |
27 |
programs and their |
28 |
libraries were not in sync. |
29 |
|
30 |
I made a handful of criticisms to FHS in that thread before, and nobody was |
31 |
able to mount a suitable defense. The point being, even in principle, separating |
32 |
/ and /usr is flaky design at best. That we just so happened to |
33 |
accumulate a number |
34 |
of packages that are historically installed to /usr is a consequence |
35 |
of that. It's not |
36 |
even necessarily the fault of the upstream developer, who's not |
37 |
supposed to care so |
38 |
much which PREFIX they install to, or the distro packager, who can't yet predict |
39 |
how the user will tailor their system. |
40 |
|
41 |
If you were in the shoes of the ebuild packagers, you would be hard-pressed to |
42 |
predict which packages belong in the / PREFIX and which ones in /usr PREFIX, |
43 |
100 times out of 100. But you need 100 times out of 100 or you'll get |
44 |
people whining |
45 |
that they can't boot or whining that they need to do some migration. That's |
46 |
why / and /usr separation is broken. |
47 |
-- |
48 |
This email is: [ ] actionable [x] fyi [x] social |
49 |
Response needed: [ ] yes [x] up to you [ ] no |
50 |
Time-sensitive: [ ] immediate [ ] soon [x] none |