1 |
On Fri, Dec 13, 2013 at 8:24 AM, Alan McKinnon <alan.mckinnon@×××××.com> wrote: |
2 |
> On 13/12/2013 14:57, Canek Peláez Valdés wrote: |
3 |
>> On Fri, Dec 13, 2013 at 1:59 AM, Alan McKinnon <alan.mckinnon@×××××.com> wrote: |
4 |
>>> On 13/12/2013 00:47, Canek Peláez Valdés wrote: |
5 |
>>>> On Wed, Dec 11, 2013 at 4:52 PM, <gottlieb@×××.edu> wrote: |
6 |
>>>>> At home I use a wired connection so did notice the following problem |
7 |
>>>>> until I traveled and tried to connect wirelessly. |
8 |
>>>>> The problem must have started sometime within the past month. |
9 |
>>>>> |
10 |
>>>>> If I have wicd started by systemd, i.e. |
11 |
>>>>> systemctl enable wicd |
12 |
>>>>> The wired network is started fine but not the wireless. Instead, I see |
13 |
>>>>> in the systemd journal |
14 |
>>>>> |
15 |
>>>>> wicd[290]: Failed to connect to non-global ctrl_ifname: wired error: No |
16 |
>>>>> such file or directory |
17 |
>>>>> wicd[290]: Failed to connect to non-global ctrl_ifname: wireless error: No |
18 |
>>>>> such file or directory |
19 |
>>>>> |
20 |
>>>>> If I instead systemctl disable wicd, reboot, and then manually type |
21 |
>>>>> wpa_supplicant -i wireless -c /etc/wpa_supplicant/wpa_supplicant.conf -B |
22 |
>>>>> it works. |
23 |
>>>>> |
24 |
>>>>> Indeed after I have booted I can start wicd and cannot get the error |
25 |
>>>>> above, but the actual behavior is not consistent. |
26 |
>>>>> |
27 |
>>>>> My system is ~amd64, profile gnome/systemd |
28 |
>>>>> |
29 |
>>>>> My wireless driver is from the package broadcom-sta (wl) |
30 |
>>>> |
31 |
>>>> I have never used wicd, so I can't say exactly what it's the problem; |
32 |
>>>> but I was under the impression that wicd is basically dead. Its last |
33 |
>>>> release was more than a year and a half ago. |
34 |
>>>> |
35 |
>>>> Regards. |
36 |
>>>> |
37 |
>>> |
38 |
>>> release more than a year and a half ago != dead |
39 |
>> |
40 |
>> In this particular case I think it is. |
41 |
>> |
42 |
>>> the code the user has still works whether the devs adds upstream commits |
43 |
>>> or not. |
44 |
>> |
45 |
>> Well, apparently not [1]. |
46 |
>> |
47 |
>>> It hasn't bit-rooted, is not incompatible with everything else and |
48 |
>>> doesn't have outstanding security bugs with little chance of being fixed. |
49 |
>> |
50 |
>> Checking [1] and [2], I would think that wicd satisfies (or *at least* |
51 |
>> starts to satisfy) the very definition of bitrot. |
52 |
>> |
53 |
>>> So what's the problem? |
54 |
>> |
55 |
>> If the code worked perfectly, none. But apparently it doesn't; I don't |
56 |
>> know, I don't use it myself. The usual signs of bitrot are there, |
57 |
>> though. |
58 |
>> |
59 |
>>> By that logic, zenity needs to have died 5 years ago but it's still around |
60 |
>> |
61 |
>> That's a really bad example. Zenity didn't had a 3.10 release, but it |
62 |
>> had a 3.8 [3] in march, so it's 9 months since the last release, not |
63 |
>> 18. Also, now zenity has a 3_10 tag in git [4]. And lastly, its lats |
64 |
>> commit was 6 days ago, and it had several bugfixes committed not three |
65 |
>> weeks ago [5]. On the other hand, wicd only has had translations |
66 |
>> committed in the last 6 *months* [6], and the "development" branch for |
67 |
>> 2.0 hasn't been touched in *3 years* [7]. |
68 |
>> |
69 |
>> This is only after a quick search through wicd and zenity repositories |
70 |
>> (and Gentoo bugzilla). Perhaps wicd has reached perfection and it |
71 |
>> doesn't need an upstream since everything simply works and there is |
72 |
>> nothing else to do with it. That would be a first in software history, |
73 |
>> though. |
74 |
>> |
75 |
>> I would simply not use it, and I will recommend any of its users to |
76 |
>> change to either NetworkManager [8] or connman [9], like pronto. |
77 |
>> |
78 |
>> [1] https://bugs.launchpad.net/wicd/+bugs |
79 |
>> [2] https://bugs.gentoo.org/show_bug.cgi?id=486440 |
80 |
>> [3] ftp://ftp.gnome.org/pub/gnome/sources/zenity/3.8/ |
81 |
>> [4] https://git.gnome.org/browse/zenity/tag/?id=ZENITY_3_10_0 |
82 |
>> [5] https://git.gnome.org/browse/zenity/log/ |
83 |
>> [6] https://code.launchpad.net/~wicd-devel/wicd/experimental |
84 |
>> [7] https://code.launchpad.net/~wicd-devel/wicd/aqua |
85 |
>> [8] http://projects.gnome.org/NetworkManager/ |
86 |
>> [9] https://connman.net/ |
87 |
>> |
88 |
>> Regards. |
89 |
|
90 |
(Sorry it took me so long to answer, I was away from my computers). |
91 |
|
92 |
> I'm not convinced, your evidence is rather under-whelming. |
93 |
|
94 |
It seems you are the only one saying that. |
95 |
|
96 |
> Yes, wicd is currently in need of a maintainer and some simple fixes per |
97 |
> your [1] are not being applied. But on the whole the code works for the |
98 |
> majority of folks, I can't find any outstanding CVEs and I don't see how |
99 |
> you can qualify this as needing to not be used. YMMV, yes, but don't use |
100 |
> it? Nah, I can't see a legitimate case. |
101 |
|
102 |
You say "it works for the majority of folks". Where do you get that? |
103 |
NM is the default network manager for GNOME, and I think at least the |
104 |
most popular with KDE (if not the default), and therefore it's the one |
105 |
used by the "majority" of folks. Search for bugs related to |
106 |
netwokrmanager *recently*; it just works for the "majority" of folks. |
107 |
|
108 |
Recommending wicd (which is not maintained for all practical purposes) |
109 |
instead of NM doesn't seem to be a very smart thing to do. I will |
110 |
always recommend people that they should use maintained software; and |
111 |
if they really want to keep an old piece of code working, that they |
112 |
should step up to the plate and take over maintaining it. |
113 |
, |
114 |
> As for zenity, it appears someone has stepped up to the plate in recent |
115 |
> months, but I clearly recall it being mostly abandoned for years. I |
116 |
> needed it for winetricks as the alternative kdialog is just ... poor. |
117 |
> But zenity couldn't be gotten to work at all. If we'd applied your POV |
118 |
> towards wicd to zenity, see where I'm going? |
119 |
|
120 |
Either I'm not understanding you, or you are not making any sense: |
121 |
zenity is maintained. It has no trivial and not translation commits |
122 |
going in in the last few days. wicd has no such things in *months*. |
123 |
|
124 |
I don't know what you "recall"; fact is zenity was never abandoned. It |
125 |
had tarballs for basically all 2.x GNOME releases, and it missed 3.10, |
126 |
but it has recent commits. Don't trust your memory (I know I don't |
127 |
trust mine); check the repositories: I just cloned zenity's, and it |
128 |
has tags for almost every GNOME release, it only misses 1.2, 1.7, |
129 |
2.25, 2.29, 3.3 and 3.5. Also, it has a gap between 1.8 and 2.5, but |
130 |
if you check the commit logs, it was a bump from 1.8 to 2.5 so it was |
131 |
in sync with GNOME version numbers. |
132 |
|
133 |
So you memory is completely wrong: zenity was *NEVER* "abandoned for |
134 |
years". Historical records preserved with strong cryptography trumps |
135 |
any fuzzy memory. |
136 |
|
137 |
> As for network-manager, we have years of history on this very mailing |
138 |
> list of people reporting problems getting it to work in anything but |
139 |
> simple straightforward cases. In so many of these cases, switching to |
140 |
> wicd fixed the issue. In all that time, you are the only person that |
141 |
> comes to mind often claiming that nm works great for them. Based on that |
142 |
> alone, I classify nm as "meh software" which might work but all too |
143 |
> often doesn't. |
144 |
|
145 |
We have talked about this before; that *some* people had troubles with |
146 |
NM years ago means *nothing* now. Fact is, NM is the most used network |
147 |
manager in Linux for desktop and laptop users (embedded and servers |
148 |
are another beast), and the *majority* of people doesn't have any |
149 |
problems with it. Just as with PulseAudio, NM uncovered some bugs and |
150 |
inconsistencies in network drivers on the Linux stack (most of the |
151 |
time not even related to NM at all), and at the beginning some people |
152 |
erroneously attributed the problems to NM. That's all. |
153 |
|
154 |
Instead wicd is bitrotten software, and it seems that no one thinks |
155 |
it's even worth to try to salvage it. |
156 |
|
157 |
> As for connman, it works great on my phone but my limited experience |
158 |
> with it on desktops was similar to nm. |
159 |
|
160 |
I haven't ever used connman; I mentioned it because at least is maintained it. |
161 |
|
162 |
> Admittedly both nm and connman might have improved by leaps and bounds |
163 |
> in recent months and perhaps they are now awesome, but I don't see it. |
164 |
|
165 |
In recent years, Alan, not months: your memory is wrong again. Once |
166 |
more, when was the last problem related to NM reported in this list? I |
167 |
saw a few days ago that the notification icon in GNOME was not |
168 |
updated, but that's a problem with nm-applet, not NM directly. And |
169 |
it's a cosmetic problem, besides. |
170 |
|
171 |
> Just call wicd for what it really is at this point: ymmv |
172 |
|
173 |
I will call wicd for that it is, now that I did even more research: |
174 |
wicd is bitrotten software that no one wants to bring back to life |
175 |
since we have so much better alternatives, primary among them |
176 |
NetworkManager. I don't think anybody should use it, much less |
177 |
recommend it. I think desktop and laptop users should stop using it, |
178 |
and switch to NetworkManager, or perhaps connman, since that one is at |
179 |
least maintained it. |
180 |
|
181 |
Regards. |
182 |
-- |
183 |
Canek Peláez Valdés |
184 |
Posgrado en Ciencia e Ingeniería de la Computación |
185 |
Universidad Nacional Autónoma de México |