| 1 |
On Mon, 22 Sep 2014 13:22:28 -0500 |
| 2 |
Canek Peláez Valdés <caneko@×××××.com> wrote: |
| 3 |
|
| 4 |
> |
| 5 |
> As a professional programmer, I completely disagree with any dogma |
| 6 |
> based on "philosophy" rather than technical merits. |
| 7 |
> |
| 8 |
> "So I think many of the "original ideals" of UNIX are these days more |
| 9 |
> of a mindset issue than necessarily reflecting reality of the |
| 10 |
> situation. |
| 11 |
> |
| 12 |
|
| 13 |
That's not the issue here at all. The issue is the possible |
| 14 |
hegemony being imposed by systemd. Whether or not this is true |
| 15 |
this issue at least deserves some attention. |
| 16 |
|
| 17 |
Regarding the "Unix philosophy," I doubt that anyone still considers |
| 18 |
that a "pipeline of simple tools" is the ideal approach. Software |
| 19 |
has assumed gigantic proportions (to match hardware capabilities) of |
| 20 |
late and that traditional Unix model certainly would never fit. |
| 21 |
|
| 22 |
But not all software is gigantic. I would venture a guess that |
| 23 |
a large majority of programs are simple one-off concoctions designed |
| 24 |
to meet some simple individual need. In these cases it sure is nice |
| 25 |
to have the standard Unix tools available. I use them frequently |
| 26 |
for various simple purposes. |
| 27 |
|
| 28 |
Regarding the booting and configuring of a Linux system, the job can be |
| 29 |
either very complex or very simple. For the simple case, is there |
| 30 |
technical merit in having to use systemd? I would claim that there |
| 31 |
is not. For complex scenarios, by all means utilize systemd. |
| 32 |
But let's keep the appropriate tools available for the appropriate job. |
Archives