1 |
Hola Ryan, |
2 |
|
3 |
On Fri, 2006-04-28 at 10:14 -0700, Ryan Phillips wrote: |
4 |
> This is a follow up to Mark's (halcy0n's) thread regarding QA Policies and |
5 |
> seemant's letter on herds, teams, and projects. |
6 |
> |
7 |
> I believe the way Gentoo is doing things is broken. There I have said it. The |
8 |
> entire project has reached a level of being too political and trying to solve |
9 |
> certain problems in the wrong way. |
10 |
|
11 |
There certainly is scope for doing some things better, but this is of no |
12 |
surprise to any of us. Gentoo has grown and matured quite a lot over the |
13 |
past few years and what worked with 15 developers doesn't work all that |
14 |
well with 300+, in the process of finding out what works best we have |
15 |
(and will continue to) encounter difficulties and situations that |
16 |
doesn't please everyone. However, I wouldn't say that it's all bad. |
17 |
|
18 |
> Some of these problems are intermixed. Please consider them starting points |
19 |
> for discussion. |
20 |
> |
21 |
> __Problem: Developer Growth__ |
22 |
> |
23 |
> I find that developer growth as being a problem. Adding a developer to gentoo |
24 |
> should be as easy as 1. has the user contributed numerous (~5+) patches that |
25 |
> helps the project move forward. If yes, then commit access should be given. |
26 |
> Adding a developer is usually quite a chore. There are numerous reasons why |
27 |
> this is a problem: having a live tree, taking a test, and not defining within |
28 |
> policy when a person could possibly get commit access. |
29 |
|
30 |
Having just completed my ebuild and end quiz last week (having been a |
31 |
'fake developer/staffer' for a little while longer) I must admit that I |
32 |
didn't in any way find the quizzes to be at all difficult or hard, and |
33 |
certainly not at all something that made it more difficult to become a |
34 |
developer. If anything I found taking the quizzes to be educational, |
35 |
now, I may be blessed by having Seemant as my mentor; he did ensure that |
36 |
he took the time to review my quiz properly and that he also took the |
37 |
time to expand on the questions already asked in the quiz (We played 20 |
38 |
questions and he made me answer a bunch of semi related stuff which in |
39 |
turn helped me ellaborate on my answers as well as getting a better |
40 |
understanding of the various bits mentioned in the quiz). And while I |
41 |
have been around for a bit now, I certainly value the work my mentor did |
42 |
in ensuring that I felt comfortable and confident before taking the |
43 |
plunge and submitting. |
44 |
|
45 |
As for the work involved from recruiters, they certainly didn't leave me |
46 |
hanging around and when my quizzes were submitted it went pretty fast |
47 |
ahead, they were marked, passed and my access was changed. |
48 |
|
49 |
I personally believe that the quizzes could be more difficult. And I |
50 |
hope that the current revamp of the recruitment process will be one that |
51 |
will benefit potential developers as well as current developers, Gentoo |
52 |
as a whole and of course our user base. |
53 |
|
54 |
> All these reasons leave the project stagnant and lacking developers. |
55 |
|
56 |
On the contrary, now while I agree that some projects may be lacking in |
57 |
man power I think overall there is more likely to be some deadweight. I |
58 |
know Bryan (kloeri) has done quite a job out of clearing out and |
59 |
retiring inactive devs, but I still believe that we may be overstaffed |
60 |
in some areas. Maybe those areas that are lacking need to look at why |
61 |
they are not attracting people? Or why people who showed interest didn't |
62 |
stick around for the entire recruitment process? I believe that we need |
63 |
to ensure that we have top notch, high quality devs rather than aiming |
64 |
for quantity. |
65 |
|
66 |
And of course, we don't want a situation where it's too easy to become a |
67 |
dev only to have people yo-yo in/out of the project as and when the |
68 |
fancy takes them. |
69 |
|
70 |
<snip> |
71 |
|
72 |
> __Problem: QA Policies__ |
73 |
> |
74 |
> http://article.gmane.org/gmane.linux.gentoo.devel/37544 |
75 |
> |
76 |
> It seems that the QA Policies are a product of a Live Tree, and going partially |
77 |
> non-live would solve the problems listed. |
78 |
> |
79 |
> Everyone here is on the same team. There will be some breakages in the tree |
80 |
> and those can be dealt with. Like Seemant [1] said, herds are just groups of |
81 |
> like *packages*. The QA Policy is wrong when it says cross-team assistance; we |
82 |
> are all on the *same* team. The tree should naturally work. If it doesn't |
83 |
> then that is a bug for all of us. |
84 |
|
85 |
Although we are all here to work on Gentoo and hopefully have a unision |
86 |
goal, the motivation and the ideas for implementation and of course what |
87 |
we find important differ. For the most part we are a group of pretty |
88 |
decent people, but we don't always agree. Which is perfectly fine. |
89 |
|
90 |
> Conflict resolution should not be a subproject. It should *not* exist at all. |
91 |
> Rules need to be in place to avoid conflict. Having some sort of voting |
92 |
> structure for all the developers (this doesn't mean requiring everyone to vote) |
93 |
> and not just the council or devrel makes a lot of sense for most things. If I |
94 |
> don't like how someone is acting within the project there should be a vote and |
95 |
> then see if that person is kicked out. No trial, no anything besides a vote. |
96 |
> And if I lose I have to deal with it. Either stay with the project, or find |
97 |
> something else. This solution just works. |
98 |
|
99 |
I am not entirely certain what your definition of conflict is, but in a |
100 |
group as large as ours there will no doubt be a conflict of interests |
101 |
and beliefs. People can't be expected to agree 100% all the time, and |
102 |
lets be realistic, if we did we would be going absolutely nowhere and |
103 |
this would be terribly boring. |
104 |
|
105 |
Now, I believe that having a votes only system in place could be quite |
106 |
dangerous and could easily be abused. Say Developer X annoyed me, so I |
107 |
decided to ensure that there was a vote, I could easily fabricate some |
108 |
reason for why I thought he was bad for the project and I could |
109 |
certainly win people over on my side to ensure that he was voted out. |
110 |
|
111 |
Now, admittedly the current policy for dealing with conflicts of the |
112 |
sort of nature where action needs to be taken against a developer has |
113 |
been proven to give more headache than what it's worth and therefor |
114 |
Devrel are attempting to work on changing the way they deal with this. |
115 |
Now, the discussion about this has been open on the gentoo-devrel@ ML |
116 |
for a few days and it appears that no-one has any input on the proposed |
117 |
new policy at all. |
118 |
|
119 |
Personally I would like to see a change, I would like us to be able to |
120 |
avoid having to go through things such as a hearing process. I would |
121 |
like for trolling/flaming and personal attacks to be discouraged and |
122 |
stomped down on as and when they occur rather than when they have become |
123 |
the sort of problem that affects the morale and motivation for all of |
124 |
us. But that's a different discussion for a different time and a |
125 |
different mailing list. |
126 |
|
127 |
|
128 |
> Gentoo should be a fun environment. The previous paragraph should be taken as |
129 |
> a last resort. |
130 |
|
131 |
Quite. "It's supposed to be fun, too." Now, personally I find Gentoo to |
132 |
be primarily fun. I am involved with some incredibly awesome teams and I |
133 |
deal with people who just plain kick arse. And I know that for many they |
134 |
find the same within their projects/teams/whatevers... It's just a shame |
135 |
we can't make it global yet ;) |
136 |
|
137 |
Cheers, |
138 |
Christel Dahlskjaer |
139 |
|
140 |
|
141 |
|
142 |
|
143 |
|
144 |
-- |
145 |
gentoo-dev@g.o mailing list |