1 |
On Aug 14, 2005, at 9:34 PM, Holly Bostick wrote: |
2 |
|
3 |
> Paul Hoy schreef: |
4 |
> |
5 |
>> See inline |
6 |
>> |
7 |
>> |
8 |
>> On Aug 14, 2005, at 5:51 PM, Holly Bostick wrote: |
9 |
>> |
10 |
>> |
11 |
>>> Nick Rout schreef: |
12 |
>>> |
13 |
>>> |
14 |
>>>> On Sun, 14 Aug 2005 13:12:31 -0700 |
15 |
>>>> Zac Medico wrote: |
16 |
>>>> |
17 |
>>>> |
18 |
>>>> |
19 |
>>>> |
20 |
>>>>> Hi Paul, |
21 |
>>>>> |
22 |
>>>>> Are we really far behind? That's difficult to believe. For what |
23 |
>>>>> packages specifically? Do |
24 |
>>>>> |
25 |
>>>>> |
26 |
>>>> |
27 |
>>>> you know how to unmask unstable packages (marked M or M~ at |
28 |
>>>> packages.gentoo.org)? |
29 |
>>>> |
30 |
>>>> Unstable does not really cut it IMHO. I am a gentoo enthusiast |
31 |
>>>> through |
32 |
>>>> and through, but plonking something in portage with a ~ beside |
33 |
>>>> it does |
34 |
>>>> not constitute a release of a recent version IMHO. |
35 |
>>>> |
36 |
>>>> |
37 |
>>>> |
38 |
>>> OK, I'll bite. What then do you consider "a release of a recent |
39 |
>>> version" |
40 |
>>> to be constituted from? |
41 |
>>> |
42 |
>>> |
43 |
>> |
44 |
>> I don't really understand your question. The most recent version |
45 |
>> to me |
46 |
>> coincides to a release date closest to whatever today is. |
47 |
>> |
48 |
> |
49 |
> OK, so what you're saying is that an application's entry into Portage |
50 |
> unstable does not constitute a 'release' of the package in Gentoo |
51 |
> terms, |
52 |
> as far as you're concerned? So until Firefox 1.0.6 and KDE 3.4.2 |
53 |
> propagate down to stable (which could take time, admittedly), it's not |
54 |
> actually released? Well, to each his or her own, I guess. |
55 |
> |
56 |
> |
57 |
>> |
58 |
>> |
59 |
>>> If it's been released upstream, and it's in Portage a couple of |
60 |
>>> hours |
61 |
>>> later, so I can install it, I don't know what more you could |
62 |
>>> want.... |
63 |
>>> what, you want a Mandrake- (or worse, still, Debian) -style wait of |
64 |
>>> months before you can use the upstream version? |
65 |
>>> |
66 |
>>> |
67 |
>> |
68 |
>> I don't agree with you. There are many examples where a file that has |
69 |
>> been released upstream has not found its way into Portage. I've |
70 |
>> provided examples elsewhere in this thread. You can also compare with |
71 |
>> the Fedora feedlist. |
72 |
>> |
73 |
> |
74 |
> Yes, I know. I'm creating a list of interesting programs I've |
75 |
> discovered |
76 |
> that aren't in Portage or b.g.o, to practice my ebuild writing skills. |
77 |
> |
78 |
> But you know, I don't give the first hairy hoot about the Fedora |
79 |
> feedlist. This idea that 'marking' a package 'stable' is some kind of |
80 |
> magic bullet that actually *makes* the package stable is starting |
81 |
> to get |
82 |
> on my nerves a bit. |
83 |
|
84 |
It appears I may be contradicting myself, but I agree with you here. |
85 |
Fedora releases something as stable, but in some cases, it's far from |
86 |
it. NetworkManager is my favourite example. |
87 |
|
88 |
|
89 |
> What Gentoo marks or doesn't mark the package, or in |
90 |
> fact whether or not it's in Portage, generally has nothing to do with |
91 |
> the status of the package itself. There are plenty of perfectly stable |
92 |
> packages in Gentoo unstable, plenty of stable ebuilds (meaning that |
93 |
> they |
94 |
> compile the application correctly, and beyond that point it depends on |
95 |
> the upstream stability) in b.g.o, and even a few on breakmygentoo.org. |
96 |
> And plenty of 'stable' packages that just act wonky in various ways as |
97 |
> upstream manages the changes in whatever they're doing (migrating |
98 |
> to the |
99 |
> freedesktop standard, implementing DirectX 9 support, working around |
100 |
> video driver bugs, kernel bugs, scheduler changes, you name it). |
101 |
> |
102 |
> I use what I need, and I get what I need from wherever it may |
103 |
> happen to |
104 |
> be. Most of it comes from Portage, of course, but I've got some |
105 |
> ebuilds |
106 |
> in my overlay from b.g.o, a couple from Project Utopia, and some perl |
107 |
|
108 |
Yes, I've scanned over the instructions for creating your own ebuilds |
109 |
and I've experimented with the Gnome 2.12 beta ebuild put out by |
110 |
someone. |
111 |
|
112 |
> modules from cpan. It all works pretty well, and when it doesn't, I |
113 |
> either ditch the package until it works a bit better, or fix it myself |
114 |
> (and report what I had to do up the chain, if appropriate). It all |
115 |
> looks |
116 |
> a bit patchwork I suppose, but it's my patchwork, and so I know what |
117 |
> sticky-out-bit goes where... most of the time. And I decide if there's |
118 |
> going to be sticky-out-bits at all...there's no way, with an ATI card, |
119 |
> that I'm going anywhere near the new modular X for quite a while, for |
120 |
|
121 |
Yes, that is one of my great joys - having an ATI card on my Notebook. |
122 |
|
123 |
> example. But not because of Gentoo... because there's way too many |
124 |
> upstream cooks for me to think they're going to concoct a 'stable' |
125 |
> brew, |
126 |
> *for me*, anytime soon. I said before and I do believe that the Gentoo |
127 |
> dev team will do their very best (and that's damn good) to provide |
128 |
> stable ebuilds that work as well as possible, but there's way too much |
129 |
> whitewater flowing down the channel for me to believe that even |
130 |
> they can |
131 |
> successfully guide me through these difficult transitions. |
132 |
> |
133 |
> It just seems to me that if you want or expect a team of well-paid |
134 |
> experts monitoring all possible inconveniences and smoothing them over |
135 |
> before you even see them... well, then Fedora would be the place to |
136 |
> be. |
137 |
> Or SuSE. Gentoo or Ubuntu, on the other hand.... |
138 |
> |
139 |
|
140 |
Again, I don't think Fedora removes all the defects at all. SuSE |
141 |
doesn't either, at least for the Gnome desktop. And, believe it not, |
142 |
neither does Ubuntu, notably with packaging. |
143 |
|
144 |
> Holly |
145 |
> -- |
146 |
> gentoo-user@g.o mailing list |
147 |
> |
148 |
> |
149 |
|
150 |
-- |
151 |
gentoo-user@g.o mailing list |