Gentoo Archives: gentoo-dev

From: Daniel Campbell <zlg@g.o>
To: gentoo-dev@l.g.o
Subject: Re: That's all folks. (Re: OT Re: [gentoo-dev] Re: [gentoo-project] [RFC] Splitting developer-oriented and expert user mailing lists)
Date: Sun, 10 Dec 2017 04:31:30
Message-Id: 20171210043112.GA23693@clocktown
In Reply to: Re: That's all folks. (Re: OT Re: [gentoo-dev] Re: [gentoo-project] [RFC] Splitting developer-oriented and expert user mailing lists) by Rich Freeman
1 On Sat, Dec 09, 2017 at 08:13:18PM -0500, Rich Freeman wrote:
2 > On Sat, Dec 9, 2017 at 7:29 PM, Daniel Campbell <zlg@g.o> wrote:
3 > >
4 > > Other developers are required to subscribe to -dev, and are
5 > > expected to follow it so they stay informed.
6 >
7 > Developers are not required to subscribe to -dev.
8 >
9 > > If they missed something covered on the list, they are directed to the
10 > > archives and (usually) laughed at.
11 >
12 > Correct. While nobody is required to follow the lists, acting out of
13 > ignorance usually doesn't impress others. Devs are expected to be
14 > adults and figure out what they need to follow based on what they
15 > intend to contribute to. -core and -dev-announce are the only
16 > required subscriptions.
17 >
18 > >
19 > > Great things coming from Gentoo "leadership" here. What will you do when
20 > > mgorny starts targeting developers and pitching tantrums over them, too?
21 >
22 > You act as if this was the only reason that comrel took action. In
23 > the cases of appeals I've seen I've yet to see a case where there
24 > wasn't something else going on behind the scene that wasn't reasonably
25 > severe when they've taken action. I can't vouch for their reasons in
26 > this case as I'm not privy to them, and I imagine they're not going to
27 > be made public.
28
29 Well, let's consider the order of events here:
30
31 1. wltjr and others appear on the ML
32 2. Drama
33 3. mgorny suggests some change in structure to avoid dealing with said
34 people.
35 4. more drama
36 5. mgorny publicly insults comrel, accusing them of doing nothing
37 6. mgorny publishes formal plan to reform our mailing lists
38 7. more drama
39 8. comrel bans wltjr
40 9. mgorny's plan is put on Council agenda
41 10. comrel *mails to let everyone know wltjr was banned*, despite prior
42 claims of valuing privacy and secrecy
43 11. you are here
44
45 This looks awfully clear to me. I'm pointing out behavior that looks a
46 lot like one person twisting the social structure to suit their desires.
47 This concerns me because our community will be damaged by his plan, and
48 it is only the first step. In the second step, he will turn against
49 developers as well. But not you and his other buddies. Just the ones
50 *he* thinks are a problem.
51
52 >
53 > > This is precisely why we have unmotivated developers
54 > > and a bevy of unmaintained packages; nobody wants to contribute to a
55 > > distro that treats its users (and developers) so poorly.
56 >
57 > Go ahead and cite the list of people who have been banned in the last
58 > decade. You won't run out of fingers on one hand. Some might cry
59 > about it for months, but good luck finding another distro that hasn't
60 > banned twice as many in the same span of years.
61 >
62 > And keep in mind that failing to take action isn't without
63 > consequences. When somebody is harassing somebody else (and sometimes
64 > more than one other) you don't really get a choice about whether
65 > somebody is going to end up leaving, whether of their own accord or
66 > not. That is a situation I've seen happen more than once around here
67 > behind the scenes. Again, I have no specific knowledge about this
68 > particular comrel action - I'm talking about situations I've seen in
69 > the past.
70
71 I'm not focused on the ban, or whether it was deserved. That's a
72 separate subject. I'm pointing out behaviors that damage our image, our
73 credibility, and morale. I'm calling out unequal enforcement and
74 favoritism; these are things that you won't find in any records, because
75 the existence of such records would damn those culpable. The fact that
76 comrel has never acted against mgorny strongly indicates that they do
77 not care about the way he treats others. He is kept because of his
78 technical skill. Others do not get this convenience; we are accountable
79 for the code *and* the words that we write. You're blind if you don't
80 see this.
81
82 >
83 > > A distro should never bend its entire social structure to protect the
84 > > feelings of one surly developer (or his/her entourage),
85 >
86 > Certainly, and that works both ways.
87 >
88 > > but naturally
89 > > since every council member is friends with mgorny and comrel is afraid
90 > > to take any action against him, they'll make exceptions to established
91 > > procedures and ignore any complaints about the way he treats others.
92 >
93 > Considering that he won a significantly contested election to Council,
94 > I suspect that more people around here approve of mgorny than just the
95 > members of the council. And I can certainly vouch that not all
96 > council members are necessarily fans of some of his actions, though I
97 > suspect that his technical contributions are praised by just about all
98 > (rightly, IMO).
99 >
100 > I've yet to see a discussion between Council members where people were
101 > strongly playing favorites the way you imply.
102
103 I'm not criticizing any code he's written. I do not have the same
104 background, nor the same open schedule needed to reach that level of
105 skill yet. This isn't a thread about code review.
106
107 The fact you're trying to change the subject isn't helping you. Can we
108 suddenly ignore it when someone's an asshole as long as they commit and
109 push good code? That's wonderful for a pure meritocracy (which like all
110 "pure"isms, is impractical and non-existent), but it's not going to help
111 this distribution because it's not how humans work.
112
113 >
114 > > Unfortunately, GLEP 39 does not have a section on recalling or
115 > > impeachment...
116 >
117 > This whole debate has been going on for over a year, and there has
118 > been an election in the interim. Do you really think that a majority
119 > of developers somehow missed the hundreds of posts on -dev the last
120 > time this debate happened? I'm not sure why you think a recall would
121 > succeed even if one were possible. Besides, the council hasn't even
122 > made any decisions here. This matter was never appealed to the
123 > council, so it seems a bit silly to hold them accountable.
124
125 More demoralizing. "Don't try, you won't be able to do anything about
126 it." You're proving what I wrote, man.
127
128 >
129 > > This whole situation highlights why the Council has no
130 > > business sticking its head into non-technical matters. It's clearly not
131 > > up to the task. It's no surprise, since technical skill does not
132 > > guarantee or even imply social skill. (or vice-versa)
133 >
134 > Dealing with social issues is a major part of the Council's purpose,
135 > per GLEP 39. I don't think the developers were blind to this in the
136 > last election, especially considering all the fiasco this was causing
137 > in the months leading up to the election.
138 >
139 > And again, this particular issue was never appealed to the Council.
140 >
141 > I'm not sure where else you would see something like this appealed.
142 > The Trustees have struggled with simply filing the tax returns. If
143 > you don't think that somebody can have both technical and social
144 > skills, I'm not sure why you think that somebody could have both
145 > financial/legal and social skills.
146
147 I invite you to navigate the Foundation's records and pull together what
148 we do have the way robbat2, quantumsummers, et al have worked to
149 recover. If it was that easy, don't you think it would've been done? The
150 unfortunate truth is exactly as wltjr said; the records fell into
151 disarray and it's taken years to get anywhere close to "ready for the
152 IRS". It's been a work-in-progress for quite a while, well before I
153 became a Trustee. New Councils do not inherit the same "work debt" that
154 Trustees do, so the comparison isn't equal.
155
156 I did not say that someone could not have both technical and social
157 skills. I said the Council isn't fit to decide on social issues. The
158 Council's purpose was to create hierarchy within the distribution
159 through which developers could seek inter-project conflict resolution
160 and final *technical* decisions that affected the distribution, like
161 what gets into EAPIs, whether eclasses get deprecated and/or removed,
162 user and group management policy, etc. The theme here is that it ties
163 into the main tree, the wiki, or other practical endeavors of
164 development. Splitting a mailing list and locking down another, while a
165 technical job in the end that Infra will do, is something with
166 far-reaching *social* consequences, that have a great chance of damaging
167 our community by creating a "chosen users" group that are worthy of
168 communicating directly with us. Other users WILL notice this, and act
169 accordingly.
170
171 I do not believe that this is brought on by anything but mgorny's
172 attempt to suit the entire ML to himself. That is selfish and
173 short-sighted. Social problems are rectified through communication, not
174 cutting out an entire subset of the community. The approach he is taking
175 to perceived slights shows that he does not care about the effects of
176 his plan, as long as he and other developers he likes can discuss
177 things.
178
179 Where are the users' opinions in all of this? Did anyone bother seeing
180 what our userbase thought? I didn't see any mail about it, nothing
181 recent on the forums (I found a single sticky from 2008, last post in
182 July of this year [1])... where exactly did we research our userbase?
183 ...Right. Nowhere. We're acting without considering our users. I'd
184 like to remind you that most council proposals get passed. Perhaps we
185 should count how many mgorny wrote to prove a point.
186
187 [1]: https://forums.gentoo.org/viewtopic-t-702248.html
188
189 >
190 > > Would you have done anything different if it were me or some
191 > > other developer who was proposing this change?
192 >
193 > What change are you proposing?
194
195 I thought I made it clear, but I'll try again: If it was me or some
196 other developer who wrote up mgorny's proposal, would it have been
197 received the same way? I posit that it wouldn't, due to a bias within
198 the Council. Or more to the point, the relationships had between Council
199 members or other long-term developers has created a sort of "good old
200 boys" club where people within the circle are valued more than the
201 "outsiders". This is known as gatekeeping, especially with the proposal
202 to partition the club from everyone else. It's plain as day.
203
204 >
205 > > It wouldn't have made it to the Council agenda if he didn't write it,
206 > > period. Everyone else would've been told to suck it up and deal with it.
207 >
208 > This is silly. Go ahead and find a single example of ANYTHING
209 > submitted by ANYBODY for the Council agenda which didn't make it onto
210 > the agenda in the last five years. I can't vouch for how things
211 > worked a decade ago but the process is basically that if somebody
212 > replies to the call for agenda items, it goes on the agenda. That
213 > doesn't guarantee the outcome that the submitter desired, but I've yet
214 > to see anything come up and be dismissed without so much as a reason
215 > unless it was retracted by the submitter.
216 >
217 > And the only item recently submitted that is relevant is the item for
218 > the splitting of the mailing list, and the Council hasn't even met to
219 > make any decisions one way or another. You're exasperated over
220 > something the Council hasn't even done.
221
222 Again, the Council approves most things that come down the line,
223 especially if it's from mgorny.
224
225 >
226 > > And knowing how the Council is, in a few days we'll all get to deal with
227 > > the churn of mailing lists to protect one person's ego. Sad.
228 >
229 > Well, if you have such a problem with the Council why don't you consider:
230 >
231 > 1. Running for the council and convincing a majority of your peers to
232 > elect you.
233 >
234 > 2. Submit whatever issues you're concerned about to the council to be
235 > discussed on the agenda instead of just whining about it on the
236 > mailing lists.
237
238 A Trustee should not be a Council member at the same time. I do not have
239 the background or work experience to lead technical decisions of a
240 distribution. I already vote in elections.
241
242 You told me a few paragraphs ago that it wouldn't be worth it anyway,
243 and now you're telling me to run for the Council? Which is it? Given the
244 council's history, the mail outlining the agenda is little more than a
245 warning of things to come. It would be futile to show up to the meeting
246 and say something.
247
248 >
249 > IMO the reason these discussions never seem to end is because opinions
250 > like yours are held by a very tiny minority of developers who assume
251 > that they're the opinion of some kind of majority that can't figure
252 > out how to vote for the right council members. All they can do is
253 > talk endlessly because the governance structure of Gentoo, by design,
254 > is intended to prevent the "special treatment to certain members of
255 > this community" that you are in fact the one who is seeking.
256
257 What special treatment am I seeking? You'll need to try harder than that
258 to deflect.
259
260 >
261 > A majority of devs selected a Council to represent their concerns and
262 > govern the distro. If you don't like the job they're doing, then
263 > don't vote for them. If they're elected anyway, consider that perhaps
264 > others are just fine with what is going on...
265 >
266 > --
267 > Rich
268 >
269
270 Ah, the populist argument. "If more people disagree with you, you are
271 wrong and should shut up."
272
273 This discussion has nowhere left to go if that's how it is.
274 --
275 Daniel Campbell - Gentoo Developer, Trustee, Treasurer
276 OpenPGP Key: 0x1EA055D6 @ hkp://keys.gnupg.net
277 fpr: AE03 9064 AE00 053C 270C 1DE4 6F7A 9091 1EA0 55D6

Attachments

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

Replies