Gentoo Archives: gentoo-dev

From: Chris Johnson <cmjohn@×××××××××××.edu>
To: gentoo-dev@g.o
Subject: Re: [gentoo-dev] Re: Unstable branch proposal - second round
Date: Sat, 30 Mar 2002 03:03:19
Message-Id: 1017478988.10692.62.camel@mule.relentless.org
In Reply to: Re: [gentoo-dev] Re: Unstable branch proposal - second round by Troy Dack
1 <special note to new readers>
2
3 I wonder if anyone else on this list cares to give input. (Hint:
4 maintainers who really know what this is all about, drobbins, etc.).
5
6 </special note>
7
8 On Sat, 2002-03-30 at 00:02, Troy Dack wrote:
9 > On Sat, 30 Mar 2002 10:40, Chris Johnson got a bunch of monkeys together
10 > and come up with:
11 >
12 > http://relentless.org:8000/gentoo/forum/message?message_id=6584&forum_id=6581
13
14 <snip>
15
16 > Fair enough... I realise that the -rx designations are there, however I
17 > have had -rx .ebuilds fail on numerous occassions because there was simply
18 > not enough testing before the ebuild was submitted to CVS.
19 >
20 > This is fine if you already have a package installed ... simply file a bug,
21 > or slap the ebuild maintainer on IRC and in a few hours (a day or two at
22 > most) the ebuild is fixed and away you go.
23 >
24
25 See, this whole "waiting on the maintainer" bit is what my proposal
26 avoids. In general, the idea is whenever *someone* gets it to work
27 right, their ebuild rises up into cvs to be tested by anyone running in
28 "10 stable installs or less" mode, which will be a sizeable portion of
29 the group. (My already stated threshold is in this category, e.g. if I
30 saw a single successful install of OpenOffice, I'd try it!--already
31 have, unsuccessfully.)
32
33 > The problem comes when a new user is trying to build their system and they
34 > get all these errors. We don't want to discourage newcomers by having a
35 > tree of ebuilds that is not 100% stable for their first installation.
36
37 In this case, the default will be to install in "more than {50|choose a
38 big number} successful installs" mode, i.e. very tested versions of each
39 package. Then, in the "Desktop guide" or other documents, it can be made
40 clear how to get the *latest* builds by changing one's make.conf to set
41 the "{10|20|50} successful installs or less" mode.
42
43 > That was my main reason for suggesting seperate CVS branch(es).
44
45 I'm trying to avoid branching, because that ultimately brings duplicated
46 effort and communication problems, not to mention contributes to the
47 slowness already alluded to (i.e "slap the maintainer on IRC to fix"
48 becomes "slap the maintainer on IRC to test it and bump it to
49 'stable'").
50
51 On George's scale of terms, I'm arguing for an _active system, passive
52 users_ as much as possible. I don't enjoy spending my whole day
53 bothering 50 maintainers to fix their packages, esp. when it's *easier*
54 to fix them myself. It's just that I'm frustrated when I make a fix--and
55 it goes nowhere, helps no one, and then I have to slog through web pages
56 to submit a bug that gets looked at a week later... yawn.
57
58 > I agree that Gentoo is not targeted at the "I've never seen linux before
59 > and thought I'd give it a go" type of user (that's what RH & MDK do), but I
60 > don't think we should make new users jump through too many hoops simply
61 > because an ebuild maintainer has hastily submitted an ebuild --
62 > particularly for core packages (baselayout is one that comes to mind).
63
64 Agreed. Baselayout/bootstrap + core systems are "maintained"; the rest
65 are "free float". Besides, if what I'm saying could be implemented,
66 then ebuilds with high success rates tend to stick in cvs, and ones that
67 are not successful in the field get demoted and replaced. Automatically.
68
69 > Perhaps a comprimise ....
70 >
71 > A stable/install CVS branch that is only used during the initial
72 > bootstrap/build process and afterwards portage defaults to using the
73 > regular CVS tree?
74
75 Why a need for seperate trees: only allow the "freefloat" system to
76 affect non-core packages, and when doing an install only use packages
77 that have ratings of "100 successful installs or greater|'STABLE'
78 blessing". This seems to solve the problem.
79
80 > Still it is a refreshing way to get my linux "fix"!
81 >
82 > --
83 > Troy Dack
84 > http://linuxserver.tkdack.com
85 >
86
87 Me too! Thanks for your response and participation! How can we measure
88 if this conversation will be fruitfull?
89
90 Chris