1 |
On Fri, Dec 28, 2012 at 12:28 AM, Kevin Chadwick <ma1l1ists@××××××××.uk> wrote: |
2 |
> |
3 |
> Again you don't break the spec unless you have to and you don't change |
4 |
> the spec unless it is an improvement or you have no choice. Non of |
5 |
> which is the case. Just like you do not mould a mail RFC to a |
6 |
> widely used technically inferior hotmail implementation. |
7 |
|
8 |
The spec - or implementation - of / and /usr separation is broken and |
9 |
has been for quite a while now. Nobody here's even bothered answering |
10 |
how the modern Gentoo distro / sysad would survive /usr being out of |
11 |
sync with /, for instance, or the fact that some udev programs tend to |
12 |
be located in /usr, or even just a solid detailed specification on the |
13 |
precise criteria for inclusion into /. Even the FHS is mum on all the |
14 |
extra crap we randomly decide between / and /usr to land in. You'd |
15 |
think, for instance, something as clear cut as filesystem manipulation |
16 |
tools, e.g., xfs_admin, would belong in /sbin rather than /usr/sbin. |
17 |
But no it's not. Or - for crying out loud, at least a text editor that |
18 |
isn't ed. |
19 |
|
20 |
Again, the broken state of the / and /usr split is a different thing |
21 |
from the usefulness state of your own already installed distro. |
22 |
|
23 |
TLDR: The spec is broken. |
24 |
|
25 |
> |
26 |
>> He's like DJB on crack. |
27 |
> |
28 |
> Except DJB made every Linux system on this planet more reliable simple |
29 |
> and secure through better coding practices and pointing out how buggy |
30 |
> sendmail was. Lennart if anything will accomplish the exact opposite |
31 |
> where systemd is used. |
32 |
|
33 |
If you have something more than FUD to back up your technical claims, |
34 |
go ahead. You're directly claiming that wherever systemd is used, the |
35 |
system will be less reliable and secure, and that Lennart isn't |
36 |
pointing out buggy behaviors in - what's the analogue for sendmail? oh |
37 |
yeah - SysVInit scripts. |
38 |
|
39 |
To carry the analogy, DJB's main point was that the size of the code |
40 |
was one of - if not the - most important factors in increasing code |
41 |
quality and security, and worked to make qmail and its configuration |
42 |
about as spartan as you can get. That's kind of the point of systemd |
43 |
unit files, trimming the boilerplate size to reduce gotchas like init |
44 |
scripts failing to detect whether a service is running or not, or if |
45 |
its dependencies have been started. |
46 |
-- |
47 |
This email is: [ ] actionable [ ] fyi [x] social |
48 |
Response needed: [ ] yes [x] up to you [ ] no |
49 |
Time-sensitive: [ ] immediate [ ] soon [x] none |