Gentoo Archives: gentoo-project

From: Tom Wijsman <TomWij@g.o>
To: gentoo-project@l.g.o
Cc: gnome@g.o, qa@g.o
Subject: [gentoo-project] Sticking to GNOME 2 complicated after regressive GNOME 3 stabilization. (was: Call for agenda items - Council meeting 2014-01-14)
Date: Wed, 08 Jan 2014 23:41:54
Message-Id: 20140109004039.08a7e814@TOMWIJ-GENTOO
In Reply to: [gentoo-project] Re: Call for agenda items - Council meeting 2014-01-14 by Ulrich Mueller
1 On Sat, 4 Jan 2014 20:46:05 +0100
2 Ulrich Mueller <ulm@g.o> wrote:
3
4 > In less than two weeks from now, the council will meet again. This is
5 > the time to raise and prepare items that the council should put on the
6 > agenda to discuss or vote on.
7
8 Hello everyone
9
10 There are reasons for an user to avoid upgrading to GNOME 3:
11
12 one reason is that the introduction of systemd is something the
13 user would not want to go through; whereas you can get a mostly
14 working installation without OpenRC, it is bound to get worse [1];
15
16 the other reason is that whereas GNOME 2 used to be light enough
17 to run on older hardware, GNOME 3 has became a much more heavy UI
18 both in terms of CPU, Memory and GPU; so, people with older
19 computers that upgrade are presented with a regressed broken system.
20
21 The former might be an option for the user; however, as seen in the
22 past as well as to this very day, quite some users would rather not
23 upgrade to it for various reasons [2]. The latter leaves no option for
24 the user, the user would love to continue using what the user already
25 has been using for quite some time; if it ain't broken, why change?
26
27 So, both of these reasons yield a lot of trouble [3] because GNOME 3
28 went stable in the same slot as GNOME 2; therefore, an user that wishes
29 to stick with GNOME 2 needs to assure he correctly puts a mask on any
30 GNOME 3 related ebuild. While this would be easy for the obvious main
31 GNOME packages with >=3, it's less obvious to do this on random
32 libraries which have other version numbers than 3.x.
33
34 This results in a set of mask files floating around in forum threads;
35 but as things are dynamic as well as the mask on some libraries
36 mismatches the actual range, it doesn't always work out well. Let alone
37 that people have to find this hidden in one or another forum thread.
38
39 From a pure QA perspective it appears even more problematic; because
40 another consequence of the upgrade to GNOME 3 is that maintainers have
41 the same problem when they are requested in a bug to fix a GNOME 2
42 version, they would need to get GNOME 2 to assure their fix works.
43
44 Well, GNOME 2 is so stable that it mostly works as well if you manage
45 to downgrade back to it or disallow any upgrade; but for how long will
46 that be? Time is ticking before GNOME 2 is no longer an option.
47
48 Since there is a need for this I would like to raise a discussion on
49 some questions here on project as well as in the Council meeting:
50
51 1. Will removing GNOME 2 when it becomes unmaintainable upset users?
52
53 (From what I've seen, yes, but what do you think?)
54
55 2. What can we do to satisfy users that want to stick to GNOME 2?
56
57 2.a. Bring a fork like MATE to the Portage tree?
58
59 2.b. Split off GNOME 2 to a different SLOT, category or name?
60 Can we still do this at this point in time?
61
62 3. How can we prevent this from happening in the future?
63
64 4. Is a news item needed for this situation? To suggest alternatives?
65
66 What do you think?
67
68 PS: For completeness, I should mention my interest with this mail is
69 to forward the feedback from users; I run GNOME 3.11 + systemd.
70
71 Rather than a tête–à–tête in #gentoo-desktop with one project
72 member or an user discussion on the forums, I am under the
73 impression that this needs much wider discussion than remaining
74 to be a hidden regression; this mail intents to help the Gentoo
75 GNOME project, it thus doesn't request a vote, just a discussion.
76
77 Thank you for reading.
78
79 References (there's more to be found, eg. on IRC; a small selection):
80
81 [1]: GNOME 3.11 - Features - Systemd for the user session
82 https://wiki.gnome.org/ThreePointEleven/Features/SystemdUserSession
83
84 [2]: A set of forum posts regarding avoiding systemd:
85
86 "Systemd headsup of L.P. - reads like at first of april"
87 http://forums.gentoo.org/viewtopic-t-962232.html
88
89 "Are we losing freedom of choice?"
90 http://forums.gentoo.org/viewtopic-t-974262.html
91
92 "Need help building a non-systemd modern system"
93 http://forums.gentoo.org/viewtopic-t-974500.html
94
95 "systemd"
96 http://forums.gentoo.org/viewtopic-t-975018.html
97
98 "Quick guide to preventing assimilation?"
99 http://forums.gentoo.org/viewtopic-t-977278.html
100
101 "please new forum _Advanced_Gentoo_without_KIT_"
102 http://forums.gentoo.org/viewtopic-t-977528.html
103
104 [3]: A set of forum posts regarding avoiding GNOME 3:
105
106 "Masking Gnome 3"
107 http://forums.gentoo.org/viewtopic-t-977288.html
108
109 "installing gnome2"
110 http://forums.gentoo.org/viewtopic-t-977638.html
111
112 "HowTo mask gnome-3.8"
113 http://forums.gentoo.org/viewtopic-t-977750.html
114
115 "emerging gdm pulls in gnome 3.8"
116 http://forums.gentoo.org/viewtopic-t-980528.html
117
118 "Help me block gnome-3.x?"
119 http://forums.gentoo.org/viewtopic-t-980866.html
120
121 "Masking Gnome 3 and installing MATE"
122 http://forums.gentoo.org/viewtopic-t-980922.html
123
124 --
125 With kind regards,
126
127 Tom Wijsman (TomWij)
128 Gentoo Developer
129
130 E-mail address : TomWij@g.o
131 GPG Public Key : 6D34E57D
132 GPG Fingerprint : C165 AF18 AB4C 400B C3D2 ABF0 95B2 1FCD 6D34 E57D

Attachments

File name MIME type
signature.asc application/pgp-signature

Replies