1 |
On Tue, Feb 18, 2014 at 10:24 AM, Tanstaafl <tanstaafl@×××××××××××.org> wrote: |
2 |
> Hi all, |
3 |
> |
4 |
> Ok, before I go and open up a bug requesting this... |
5 |
> |
6 |
> I know there have to be a lot of people on this list who can answer this |
7 |
> question... |
8 |
> |
9 |
> Is making the use of systemd or not based on a selected Profile, as opposed |
10 |
> to manually trying to do it via USE flags etc, a practical request, or not? |
11 |
> |
12 |
> Meaning: |
13 |
> |
14 |
> 1. is it doable, |
15 |
|
16 |
Of course it is. |
17 |
|
18 |
> 2. how hard (on a scale of 1/easy to 10/hard to impossible) would it |
19 |
> be to do this, and |
20 |
|
21 |
Well, doing it I think it's easy. |
22 |
|
23 |
> 3. regardless of the difficulty in achieving it, how hard would it be |
24 |
> (on a scale of 1 (easy) to 10 (hard to impossible)) to maintain in |
25 |
> the long run? |
26 |
|
27 |
That's the pickle, isn't? Forget about how "hard" it is. |
28 |
|
29 |
The problem, as I see it, is that this systemd profile is being pushed |
30 |
by the people that *don't* want to use systemd, because they want it |
31 |
restricted to a separated profile so their machines are not "tainted" |
32 |
by systemd. |
33 |
|
34 |
If it was the other way around (systemd users wanting a systemd |
35 |
profile), there would be no problem, since they would happily do the |
36 |
job. But as it is it makes no sense; if you want to maintain a systemd |
37 |
profile, you need to *use* systemd to do the testing and supporting. |
38 |
Are the people that don't want to use systemd, will use systemd to |
39 |
test and support a profile so they don't need to use systemd? Really? |
40 |
|
41 |
(And never mind that more and more packages in the stack will want to |
42 |
use systemd's features, and some of them probably will not work, or |
43 |
work with limited functionality, without said features. And a profile |
44 |
is not going to change that.) |
45 |
|
46 |
That being said, the GNOME team introduced a gnome/systemd profile |
47 |
(and later the KDE team did the same). I've never saw the point in |
48 |
doing that, but that's what they decided and is fine. |
49 |
|
50 |
If the Gentoo developers that use and prefer systemd decide to create |
51 |
a systemd profile, then it will (easily, I think) happen. |
52 |
|
53 |
But it's the systemd users the ones you should convince to do it; not |
54 |
anyone *not* using it or avoiding it. |
55 |
|
56 |
> I'm hoping that even if the answer to #2 above is 7+, if the answer to #3 is |
57 |
> 'relatively easy' or better, then maybe it still stands a chance of getting |
58 |
> done. |
59 |
|
60 |
The possibility exists, sure. You can talk to Michał and the other |
61 |
members of the systemd project in Gentoo. If you convince them that it |
62 |
is a good idea (and good for systemd, obviously), they will do it. |
63 |
|
64 |
But anyone that wants to help this happen will need to use systemd to |
65 |
support a systemd profile. That's simple logic. |
66 |
|
67 |
> And incidentally, Canek, if I was a programmer, I'd be happy to volunteer to |
68 |
> take it on, but alas I'm just a lowly user... |
69 |
|
70 |
I would, but I don't think is worth it. It's not an unreasonable idea, |
71 |
but I think it will make thinks muddy when someone chooses the "not |
72 |
systemd" profile, and they will discover that anyhow they need systemd |
73 |
to run a lot of things. |
74 |
|
75 |
And, I repeat, usually when someone pushes for an X profile, it's |
76 |
because they want to use X. In this case you are asking for an X |
77 |
profile so you don't need to even see X; it makes no sense, since the |
78 |
X profiles needs the people maintaining it to use X. |
79 |
|
80 |
Regards. |
81 |
-- |
82 |
Canek Peláez Valdés |
83 |
Posgrado en Ciencia e Ingeniería de la Computación |
84 |
Universidad Nacional Autónoma de México |