| 1 |
On Sun, 21 Sep 2014 20:45:17 -0400 |
| 2 |
Rich Freeman <rich0@g.o> wrote: |
| 3 |
|
| 4 |
> |
| 5 |
> You can create device nodes using mknod, and I'd be |
| 6 |
> shocked if that ever went away. |
| 7 |
> |
| 8 |
|
| 9 |
But now certain static USB nodes, in particular those for |
| 10 |
scanners, have been removed in favor of dynamic allocation |
| 11 |
using udev or its equivalents. When this happened I was |
| 12 |
certainly shocked, and it could be the beginning of a trend. |
| 13 |
|
| 14 |
> |
| 15 |
> Just what is it that you actually |
| 16 |
> need the kernel to do for you that you don't think will still be |
| 17 |
> around in 20 years? Linus is VERY conservative about removing system |
| 18 |
> calls. |
| 19 |
> |
| 20 |
|
| 21 |
There are things which are not system calls that could easily be |
| 22 |
changed. It is not too far fetched to consider a time if and when |
| 23 |
systemd became so popular and entrenched that the kernel would be |
| 24 |
hard-coded to pass control only to systemd and nothing else. |
| 25 |
|
| 26 |
> |
| 27 |
> If the whole world moves to systemd the biggest problem you'll have is |
| 28 |
> that you'll have to write your own service startup scripts, but from |
| 29 |
> the sound of things you're doing that anyway. Most of the services |
| 30 |
> you probably run aren't linux-exclusive either, so while it seems |
| 31 |
> likely that many will start reporting their status to systemd it seems |
| 32 |
> unlikely that they will refuse to work without it. |
| 33 |
> |
| 34 |
|
| 35 |
There are a growing number of applications that will no longer compile |
| 36 |
without either dbus or udev. In fact, even though I don't use them, |
| 37 |
I had to install both eudev and dbus in order to be able to use certain |
| 38 |
applications (I just substituted a symlink to /bin/true in place of |
| 39 |
dbus-launch to keep that unnecessary daemon from starting). |
| 40 |
|
| 41 |
I am not that familiar with systemd components, but it is not too |
| 42 |
unrealistic to consider many more applications in the future making |
| 43 |
at least some components mandatory. |
| 44 |
|
| 45 |
It is obvious that the Linux of 10 years ago is no longer appealing |
| 46 |
to many people and there will be mounting pressure to introduce changes |
| 47 |
just for the sake of having changes. |
| 48 |
|
| 49 |
If I have to adapt then I will certainly adapt, but it would be better |
| 50 |
to keep current options. |
Archives