1 |
On 12/16/2016 06:04 PM, Taiidan@×××.com wrote: |
2 |
> Thoughts: |
3 |
> Pulseaudio, ahavi, systemd, etc, any and all pottering products - it |
4 |
> really makes me wonder as to why the developers of every major distro |
5 |
> suddenly all at once forced them all a community that didn't want or |
6 |
> need them, with no easy way to disable them. |
7 |
> |
8 |
> Firefox is a browser made by those who are all too willing to compromise |
9 |
> their morals for profit (ads on newtab, google tracking software, pocket |
10 |
> and telefonia integration, not even as an addon just simply no way to |
11 |
> remove them), it is long overdue for a major fork. (gnu icecat exists, |
12 |
> but it needs more developers to be a real fork) |
13 |
> |
14 |
> It reminds me of facebook, with each and every update there is yet |
15 |
> another obscure security/privacy setting or change that a power-user |
16 |
> needs to fiddle with... |
17 |
> |
18 |
|
19 |
It generally comes down to convenience for a lot of maintainers. |
20 |
Apparently some maintainers (mostly Arch and Debian) had a lot of |
21 |
problems with the older software and would rather Embrace what was new |
22 |
than maintain what was old. Others choose to go with the new thing so |
23 |
they can close a bunch of old bugs or get rid of what they consider 'bad |
24 |
code'... by replacing it with newer and (at the time) untested code. |
25 |
Their time, their rules, I suppose. But I don't buy into that. |
26 |
|
27 |
Additionally, at least one of the projects you listed were actively |
28 |
suggested and peddled to the greater community. I won't go into |
29 |
conspiracy stuff since you really don't need to; the facts are damning |
30 |
enough: |
31 |
|
32 |
* Poettering and his followers have actively pushed for systemd to be |
33 |
default init on distributions. This is evidenced by Lennart's joining of |
34 |
the GNOME Foundation and suggesting that systemd be a dependency of |
35 |
GNOME back in 2011 [1], and followed of course by Fedora, Red Hat, and |
36 |
SuSE, given his ties to others in those communities. |
37 |
|
38 |
* One particular follower caused a lot of drama on Debian's mailing list |
39 |
a while back, [2] derailing it into systemd-pushing. Said conduct was |
40 |
*congratulated* and *encouraged* by Mr. Poettering on Google+. [3] |
41 |
|
42 |
* There was a culture change at Arch after Aaron Griffin took the helm. |
43 |
Whatever happened during the time between Judd's stepping down (Oct 1st, |
44 |
2007) and systemd being pushed onto user machines (August of 2012 iirc), |
45 |
it resulted in a few Arch developers gaining commit access to the main |
46 |
systemd repository [4] and the general treatment of Fedora and Red Hat |
47 |
as primary upstream by Arch. |
48 |
|
49 |
Sources: |
50 |
[1] |
51 |
https://mail.gnome.org/archives/desktop-devel-list/2011-May/msg00427.html |
52 |
[2] http://lists.debian.org/debian-devel/2012/11/msg00350.html |
53 |
[3] |
54 |
https://plus.google.com/+LennartPoetteringTheOneAndOnly/posts/jcCjMct3SJ3 |
55 |
(Ctrl+F "debian-devel"; from Dec 19th 2012) |
56 |
[4] |
57 |
"falconindy" -> Dave Reisner |
58 |
https://github.com/falconindy |
59 |
https://www.archlinux.org/people/developers/#dreisner |
60 |
|
61 |
"tomegun" -> Tom Gundersen |
62 |
https://github.com/teg |
63 |
https://www.archlinux.org/people/developers/#tomegun |
64 |
|
65 |
(cross-referenced from |
66 |
https://github.com/systemd/systemd/graphs/contributors . As such, it |
67 |
does not count merge-commits, which are irrelevant to this claim.) |
68 |
|
69 |
--- |
70 |
|
71 |
(Everything laid out above is factual, with credible sources. No libel |
72 |
or slander is intended in this e-mail. It is meant solely for |
73 |
informative purposes. My opinion is below, and does not reflect Gentoo |
74 |
policy or goals whatsoever.) |
75 |
|
76 |
And again, because someone will inevitably cry foul and not read: |
77 |
|
78 |
--- ABOVE ARE FACTS; BELOW IS MY OPINION, NOT GENTOO'S --- |
79 |
|
80 |
Some people don't think software evangelism or peddling is bad. They're |
81 |
entitled to that opinion. I personally find it horribly myopic, and |
82 |
would rather a piece of software stand on its own merits, without a |
83 |
marketing team trying to get people to use it. The push for these pieces |
84 |
of software (all written by one particular guy...ain't that funny?) |
85 |
stands out from the typical libre software projects (Random examples: |
86 |
cgit, fluxbox, pelican, Tox), who at most get one large announcement and |
87 |
maybe a few reblogs. Some may argue it's because these projects have a |
88 |
good media team. I think evangelism taints software and detracts from |
89 |
its merit. There's a fine line between "awareness" and "propaganda", and |
90 |
it's important to walk it carefully. |
91 |
|
92 |
These days, systemd evangelism is somewhat dead, unless you're a new |
93 |
distro. Instead, the angle du jour is to court application makers and |
94 |
figure out what they want to see in Pulse or systemd or whatever, then |
95 |
put it in so the application developer will use it. Users are then |
96 |
forced to adopt the packages or go without the program that they need |
97 |
and/or use. This was attempted with kdbus (to force udev to run with |
98 |
systemd -- good thing we forked it!), which has gone back to the drawing |
99 |
board, and has been very effective with pulse (Steam, Skype, OBS Studio, |
100 |
now Firefox...) This is a very shrewd practice that works on the |
101 |
socially or politically unaware, and doubly so when combined with their |
102 |
"technical or nothing" outlook on community. |
103 |
|
104 |
Any person -- user or developer -- who chooses to ignore the world |
105 |
around them in favor of the computer screen in front of them is missing |
106 |
the rest of the libre software picture. It's all powered by people, and |
107 |
every project will unavoidably have both its technical and its social |
108 |
merit tested at some point or another. Ignoring it (or worse, forbidding |
109 |
it) shows a distinct lack of maturity and introspection, imo. This |
110 |
applies to both technical and social, though obviously I'm talking about |
111 |
social merit, which fewer projects get right. |
112 |
|
113 |
I currently run Pulse because OBS Studio requires it if you're going to |
114 |
stream output *and* input; a use case that's possible with ALSA but docs |
115 |
are scant on doing it right. A lot of people don't realize that Pulse is |
116 |
actually on top of -- and requires -- ALSA. Have you tried using |
117 |
media-sound/apulse with the new Firefox build(s)? I had a lot of success |
118 |
with it when I was still purely ALSA; I'd like to return to that, if |
119 |
only to send a message that homogenization agendas aren't welcome on my |
120 |
machines. Just keep in mind that it (apulse) was meant solely for Skype. |
121 |
The fact that it runs with other programs is just a bonus (or it |
122 |
demonstrates how easily Pulse can be replaced and shim'd... depending on |
123 |
your outlook). |
124 |
|
125 |
As for Firefox, I don't really need sound in it. I use (and maintain, |
126 |
full disclosure!) media-video/smtube and it works really well, though |
127 |
I've only tested it on youtube so far. There's also net-misc/youtube-dl |
128 |
if you want to hold onto videos for later. It supports more than |
129 |
youtube, too! Should there ever be a hard dependency on Pulse, I'll |
130 |
reject it and find some other browser project. I've been drifting away |
131 |
from the Web for a year or two now, so I won't personally miss out on |
132 |
much if I stop using Firefox (except the myriad security problems that |
133 |
come with a browser + JS). The browser space is long overdue for some |
134 |
competition. I'll gladly try out alternatives. |
135 |
|
136 |
Sorry for the essay. I've discussed the subject a lot in the past and |
137 |
hope that whatever others choose for their software, they choose *after* |
138 |
investigating the situation and learning more. That way, they're making |
139 |
the decision for themselves instead of following others. systemd is just |
140 |
one choice among many, and I'm proud to be part of a distribution that |
141 |
respects and enables user choice. I wouldn't have become a developer |
142 |
otherwise. |
143 |
-- |
144 |
Daniel Campbell - Gentoo Developer |
145 |
OpenPGP Key: 0x1EA055D6 @ hkp://keys.gnupg.net |
146 |
fpr: AE03 9064 AE00 053C 270C 1DE4 6F7A 9091 1EA0 55D6 |