Gentoo Archives: gentoo-devrel

From: Michael Tindal <urilith@g.o>
To: gentoo-devrel@l.g.o
Subject: [gentoo-devrel] Exchanges with Ciaranm
Date: Fri, 10 Jun 2005 06:29:26
Message-Id: 42A93379.7010904@gentoo.org
1 These are all of the emails that were exchanged between myself and
2 ciaranm. They're the raw form of the email, just pasted here. I do not
3 agree with everything he put, as he obviously does not agree with
4 everything in the proposal, however, he does bring up several points
5 worthy of discussion, and I do not want to paraphrase it and
6 unintentionally misinterpret what he meant. Also, do not take these as
7 me or him attempting to flame or undermine devrel. These reponses were
8 in reponse to the proposal and the problems the proposal intends to fix.
9
10 ---
11
12 First email (From me to ciaranm):
13
14 Hey Ciaran,
15
16 I wanted to write you to let you know some things that have happened
17 since you left, and should help ease some of your concerns. Your
18 observations were spot on, but I'd like to clear up some things. Devrel
19 did admit there were problems in the process, and I have written a
20 proposal which should be instituted within devrel on Tuesday, and
21 clarifies the process, makes the process more transparent, and ensures
22 that something like your situation happens again.
23
24 It ensures that any evidence used in a decision to take action, and the
25 details that determine that action are publicly viewable from the start.
26 It also takes the power to make those decisions from one or two people
27 and puts it into a group of people. It also includes some clarification
28 of who devrel answers to, and makes it so devrel isnt immune to being
29 changed or challenged.
30
31 You can view this proposal at
32 http://dev.gentoo.org/~urilith/devrel-proposal.txt. I'd definitely like
33 to hear your thoughts on it, whether you think its a good step forward
34 (its not trying to solve every problem, just giving us a base from which
35 to work forward), and what you would have changed.
36
37 Mike Tindal
38
39 ---
40
41 First reply from ciaranm:
42
43 On Sun, 05 Jun 2005 18:54:07 -0500 Michael Tindal <urilith@g.o>
44 wrote:
45 | You can view this proposal at
46 | http://dev.gentoo.org/~urilith/devrel-proposal.txt. I'd definitely
47 | like to hear your thoughts on it, whether you think its a good step
48 | forward (its not trying to solve every problem, just giving us a base
49 | from which to work forward), and what you would have changed.
50
51 Hm. It's a start, but there're things it doesn't address:
52
53 * The outcome can still be decided by a very small number of devrel
54 members. With the current situation, even various devrel members are
55 calling their own system a "kangaroo court". I was told by no less than
56 three different devrel members that dmwaters and jhuebel were dead set
57 on suspending me and that nothing said at the meeting that they were
58 planning would affect the outcome. With this proposal, a few highly
59 biased members on the Judiciary project can still skew the outcome for
60 personal vendettas.
61
62 * The people who will not be able to act in separate capacities during
63 the Investigation and Judiciary phases are the people who will claim
64 that they can. Being involved in one phase must utterly preclude being
65 involved in the other phase.
66
67 * The proposed system is still heavily skewed in favour of anyone who
68 happens to make a complaint. There is a huge 'Investigation'
69 (prosecution) group. We've already seen that devrel will attempt to
70 obfuscate any complaint by using pseudo-legal jargon (and getting it
71 wrong, no less). There should be an equally huge defence group.
72
73 * The proposed system is still heavily skewed towards a very small
74 number of people having all the power. Being on the Judiciary board
75 should utterly preclude also being a manager, foundation member or being
76 in any other position which has influence upon how devrel operate. The
77 final paragraph is insufficient.
78
79 * The proposal does not address devrel's lack of perspective on what the
80 real problems are. From experience, any complaint to devrel regarding
81 technical aspects, tree breakage or the like is met with "We are not
82 prepared to discuss this". However, devrel will be quite happy to
83 investigate claims from people who put out intentionally false press
84 releases when someone objects to us lying to our users, and they will
85 quite happily investigate claims from the ricer crowd that were clearly
86 just put together to cause trouble.
87
88 * The proposal does nothing to discourage people from using devrel as a
89 way of avoiding having sensible discussion. There is still no reason for
90 someone who thinks that they have a complaint to ask for clarification.
91 For example, a forums mod could still quite happily take some quote
92 completely out of context, misunderstand it thanks to not speaking
93 english english, use it as a complaint and not bother to find out what
94 the actual issues involved are. Sure, it *might* be caught later on if
95 people are actually ever told what the charges against them are, but
96 that's a hell of a lot of mess.
97
98 Incidentally, it may amuse you to find out that devrel still haven't
99 told me what the actual charges against me were, and the only reason I
100 know about them is because someone was kind enough to give me an HTML
101 dump of a restricted access bug. What's worse is that of all the things
102 that devrel *could* have legitimately gotten me for, not one of them is
103 in that bug.
104
105 Anyway, have fun. By all means use my observations thingie as a
106 justification for your proposal if you haven't done so already.
107 Something else that may help you -- I've received quite a few emails
108 from people saying they're seriously considering resigning because of
109 the behaviour from a few people in devrel. Might add some clout...
110
111 *shrug* I think it's a nice start. I don't think it's enough, but
112 equally I don't think anything that would really fix things would be
113 accepted.
114
115 Regards,
116 -- Ciaran McCreesh Mail: ciaranm at firedrop.org.uk
117
118 ---
119
120 Second email from me to Ciaranm:
121
122 Ciaran McCreesh wrote:
123
124 >> On Sun, 05 Jun 2005 18:54:07 -0500 Michael Tindal <urilith@g.o>
125 >> wrote:
126 >> | You can view this proposal at
127 >> | http://dev.gentoo.org/~urilith/devrel-proposal.txt. I'd definitely
128 >> | like to hear your thoughts on it, whether you think its a good step
129 >> | forward (its not trying to solve every problem, just giving us a base
130 >> | from which to work forward), and what you would have changed.
131 >>
132 >> Hm. It's a start, but there're things it doesn't address:
133 >>
134 >> * The outcome can still be decided by a very small number of devrel
135 >> members. With the current situation, even various devrel members are
136 >> calling their own system a "kangaroo court". I was told by no less than
137 >> three different devrel members that dmwaters and jhuebel were dead set
138 >> on suspending me and that nothing said at the meeting that they were
139 >> planning would affect the outcome. With this proposal, a few highly
140 >> biased members on the Judiciary project can still skew the outcome for
141 >> personal vendettas.
142
143
144
145 Thanks for your quick reply. I will definitely reword some of this
146 then, since the entire proposal was brought about by my utter disgust in
147 how they treated your situation.
148
149 The proposal isn't supposed to be a means by which to solve all
150 problems, I would expect that to take more time, however it does attempt
151 to address several issues that exist within devrel at this time. Under
152 this proposal, any evidence used to reach a decision (which was not
153 presented to you, for example) must be available to the public at the
154 time that decision is reached. If for some reason the members are
155 biased, the evidence available will point to that, and the decision
156 overturned. Having too many people sit on the panel just increases the
157 chances for bias.
158
159
160 >> * The people who will not be able to act in separate capacities during
161 >> the Investigation and Judiciary phases are the people who will claim
162 >> that they can. Being involved in one phase must utterly preclude being
163 >> involved in the other phase.
164
165
166 I will add that clarification then. I didn't get enough comments from
167 -core to justify changing the wording, but since the proposal was for
168 your treatment, I value your opinion above others at this point.
169
170
171 >> * The proposed system is still heavily skewed in favour of anyone who
172 >> happens to make a complaint. There is a huge 'Investigation'
173 >> (prosecution) group. We've already seen that devrel will attempt to
174 >> obfuscate any complaint by using pseudo-legal jargon (and getting it
175 >> wrong, no less). There should be an equally huge defence group.
176
177
178 Hmm, I do not believe I limited the size of the defense group. The
179 system within the investigative subgroup still favors complaintants, I
180 agree, however, the requirement of being publicly available should keep
181 the judicial subgroup from favoring any party. I will however add
182 specific wording to the effect that a defendant may have any number of
183 peers dispute evidence or claims in their defense. Would this be
184 sufficient to avoid bias at the judicial level?
185
186
187 >> * The proposed system is still heavily skewed towards a very small
188 >> number of people having all the power. Being on the Judiciary board
189 >> should utterly preclude also being a manager, foundation member or being
190 >> in any other position which has influence upon how devrel operate. The
191 >> final paragraph is insufficient.
192
193
194 I will add wording to that effect then. Like I mentioned previously, I
195 did not receive strong enough opinions.
196
197
198 >> * The proposal does not address devrel's lack of perspective on what the
199 >> real problems are. From experience, any complaint to devrel regarding
200 >> technical aspects, tree breakage or the like is met with "We are not
201 >> prepared to discuss this". However, devrel will be quite happy to
202 >> investigate claims from people who put out intentionally false press
203 >> releases when someone objects to us lying to our users, and they will
204 >> quite happily investigate claims from the ricer crowd that were clearly
205 >> just put together to cause trouble.
206
207
208 Unfortunately that is beyond the scope of this proposal. If you would
209 like, I can write up another proposal to discuss with devrel that would
210 add specific policies dicussing the differences between these
211 complaints. This proposal is just trying to open the process up so
212 someone isnt unjustly suspended like you were.
213
214
215 >> * The proposal does nothing to discourage people from using devrel as a
216 >> way of avoiding having sensible discussion. There is still no reason for
217 >> someone who thinks that they have a complaint to ask for clarification.
218 >> For example, a forums mod could still quite happily take some quote
219 >> completely out of context, misunderstand it thanks to not speaking
220 >> english english, use it as a complaint and not bother to find out what
221 >> the actual issues involved are. Sure, it *might* be caught later on if
222 >> people are actually ever told what the charges against them are, but
223 >> that's a hell of a lot of mess.
224
225
226 Under this proposal people must be told what the charges are, or no
227 decision can be brought against them. If you would like, I can add
228 specific wording to say that evidence must include the complete context
229 of the quote (perhaps the whole page, relevant posts, etc), or the
230 evidence cannot be used (much like the proposal states private evidence
231 cannot be used).
232
233
234 >> Incidentally, it may amuse you to find out that devrel still haven't
235 >> told me what the actual charges against me were, and the only reason I
236 >> know about them is because someone was kind enough to give me an HTML
237 >> dump of a restricted access bug. What's worse is that of all the things
238 >> that devrel *could* have legitimately gotten me for, not one of them is
239 >> in that bug.
240
241
242 It does not amuse me, since that was the entire reason I wrote this
243 proposal. This proposal requires charges to be publicly documented.
244 See http://dev.gentoo.org/~urilith/warning-suspension-draft.txt for some
245 drafts of the forms. In your case, your charges would have been in the
246 first URL, the evidence in the second. And I agree the bug is bullshit.
247 As do many people. It might also make you feel better to know
248 g2boojum agrees the bug is insufficient. You might wish to appeal the
249 decision following Tuesday's meeting, stating that there is not enough
250 publicly accessible evidence to justify your suspension. There's quite
251 a few people in the project who would agree with me (myself included).
252
253
254 >> Anyway, have fun. By all means use my observations thingie as a
255 >> justification for your proposal if you haven't done so already.
256 >> Something else that may help you -- I've received quite a few emails
257 >> from people saying they're seriously considering resigning because of
258 >> the behaviour from a few people in devrel. Might add some clout...
259 >>
260 >> *shrug* I think it's a nice start. I don't think it's enough, but
261 >> equally I don't think anything that would really fix things would be
262 >> accepted.
263
264
265 I don't think its enough, either, incidentally. However, my main
266 concern was trying to get something in to make it more transparent, and
267 hopefully convince you to come back, either when you appeal or when your
268 60 days is up. Theres many people in the project who miss your
269 technical knowledge and the things you have done. If theres anything
270 else you would like me to bring up to devrel, I would be more than happy
271 to. Also, would you mind if I shared these three emails with devrel to
272 show that you do indeed have legitimate, reponsible claims? It would
273 definitely provide proof that you're not just trying to turn the issues
274 away from yourself.
275
276 Thanks,
277
278 Mike Tindal
279
280 ---
281
282 Second reply from ciaranm:
283
284 On Mon, 06 Jun 2005 17:40:00 -0500 Michael Tindal <urilith@g.o>
285 wrote:
286 | The proposal isn't supposed to be a means by which to solve all
287 | problems, I would expect that to take more time, however it does
288 | attempt to address several issues that exist within devrel at this
289 | time. Under this proposal, any evidence used to reach a decision
290 | (which was not presented to you, for example) must be available to the
291 | public at the time that decision is reached. If for some reason the
292 | members are biased, the evidence available will point to that, and the
293 | decision overturned. Having too many people sit on the panel just
294 | increases the chances for bias.
295
296 Problem is, this still leaves judgement in the hands of a very small
297 number of people. If these were people like g2boojum or fmccor, who
298 really do seem to be capable of acting without bias, it wouldn't be a
299 problem. On the other hand, give it to the current crowd and we're back
300 at the kangaroo court. At the very least, the committee would need to
301 include people from a variety of roles -- ensure that there are some
302 real 'tree' people in there as well as the current PR and relations
303 shower.
304
305 Whilst we're at it, there's another related issue which I'm probably
306 not going to explain very clearly.
307
308 We're dealing with a bunch of developers who, by and large, use words to
309 mean exactly what they're intended to mean. Equally, we've already seen
310 that devrel love incorrectly using various pseudo-legal terms because
311 they think that it tries to add weight to their arguments -- the
312 warning I was sent was very much a Humpty Dumpty ("When I use a word, it
313 means just what I choose it to mean -- neither more nor less.") use of
314 English. So if, say, someone is accused of "slander", chances are
315 they'll respond to the allegation of slander, and not to the allegation
316 of "not being sufficiently tactful when making truthful statements",
317 which is what devrel actually mean.
318
319 If the judiciary committee happens to consist of (as it probably will)
320 the types who are currently highly placed within devrel, they'll
321 interpret the incorrect accusations as what they were trying to say, not
322 what they actually meant. This is an immediate and extremely large bias
323 against the accused, who will be busy trying to defend himself against
324 what he was actually accused of.
325
326 There are ways around this. I'd suggest requiring that all warnings are
327 given in plain, straight forward english and that they must state what
328 is meant.
329
330 | > * The proposed system is still heavily skewed in favour of anyone
331 | > who happens to make a complaint. There is a huge 'Investigation'
332 | > (prosecution) group. We've already seen that devrel will attempt to
333 | > obfuscate any complaint by using pseudo-legal jargon (and getting it
334 | > wrong, no less). There should be an equally huge defence group.
335 |
336 | Hmm, I do not believe I limited the size of the defense group. The
337 | system within the investigative subgroup still favors complaintants, I
338 | agree, however, the requirement of being publicly available should
339 | keep the judicial subgroup from favoring any party. I will however
340 | add specific wording to the effect that a defendant may have any
341 | number of peers dispute evidence or claims in their defense. Would
342 | this be sufficient to avoid bias at the judicial level?
343
344 I'd rather see something along the lines of a dedicated defence group,
345 *on top of* the extra peers. It's unreasonable to expect some people to
346 be able to compete fairly with the hard-core witch hunters.
347
348 | > * The proposal does not address devrel's lack of perspective on what
349 | > the real problems are. From experience, any complaint to devrel
350 | > regarding technical aspects, tree breakage or the like is met with
351 | > "We are not prepared to discuss this". However, devrel will be quite
352 | > happy to investigate claims from people who put out intentionally
353 | > false press releases when someone objects to us lying to our users,
354 | > and they will quite happily investigate claims from the ricer crowd
355 | > that were clearly just put together to cause trouble.
356 |
357 | Unfortunately that is beyond the scope of this proposal. If you would
358 | like, I can write up another proposal to discuss with devrel that
359 | would add specific policies dicussing the differences between these
360 | complaints. This proposal is just trying to open the process up so
361 | someone isnt unjustly suspended like you were.
362
363 I'm not sure that separating it is a good idea. Devrel's current "stuff
364 we can enforce" rules are limited to the self-written never agreed upon
365 "Etiquette Policy". Trying to get any reasonable sense of justice out of
366 this would be a lot easier if the scope of devrel's investigations were
367 properly set out. Currently devrel's response to any technical
368 complaints is "we are not prepared to discuss this" or "isn't that QA's
369 job, not ours?".
370
371 | > * The proposal does nothing to discourage people from using devrel
372 | > as a way of avoiding having sensible discussion. There is still no
373 | > reason for someone who thinks that they have a complaint to ask for
374 | > clarification. For example, a forums mod could still quite happily
375 | > take some quote completely out of context, misunderstand it thanks
376 | > to not speaking english english, use it as a complaint and not
377 | > bother to find out what the actual issues involved are. Sure, it
378 | > *might* be caught later on if people are actually ever told what the
379 | > charges against them are, but that's a hell of a lot of mess.
380 |
381 | Under this proposal people must be told what the charges are, or no
382 | decision can be brought against them. If you would like, I can add
383 | specific wording to say that evidence must include the complete
384 | context of the quote (perhaps the whole page, relevant posts, etc), or
385 | the evidence cannot be used (much like the proposal states private
386 | evidence cannot be used).
387
388 Guess I didn't explain that point too well. Any issues should be
389 discussed with the relevant parties before they get to the devrel stage.
390 At least one of the allegations on the bug I'm not supposed to have seen
391 is a result of someone not understanding what I was actually saying.
392
393 | It does not amuse me, since that was the entire reason I wrote this
394 | proposal. This proposal requires charges to be publicly documented.
395 | See http://dev.gentoo.org/~urilith/warning-suspension-draft.txt for
396 | some drafts of the forms. In your case, your charges would have been
397 | in the first URL, the evidence in the second. And I agree the bug is
398 | bullshit.
399
400 Bleh. That reads like "we are going to open a can of devrel-style
401 whoopass on you", not "there is an issue which we would like to have
402 resolved sanely".
403
404 | As do many people. It might also make you feel better to know
405 | g2boojum agrees the bug is insufficient. You might wish to appeal the
406 | decision following Tuesday's meeting, stating that there is not enough
407 | publicly accessible evidence to justify your suspension. There's
408 | quite a few people in the project who would agree with me (myself
409 | included).
410
411 Bleh, I've had enough of arguing with devrel. Their "no discussion"
412 policy precludes any reasonable dialogue. There're far less futile
413 things upon which I could waste my time.
414
415 | > Anyway, have fun. By all means use my observations thingie as a
416 | > justification for your proposal if you haven't done so already.
417 | > Something else that may help you -- I've received quite a few emails
418 | > from people saying they're seriously considering resigning because
419 | > of the behaviour from a few people in devrel. Might add some
420 | > clout...
421 | >
422 | > *shrug* I think it's a nice start. I don't think it's enough, but
423 | > equally I don't think anything that would really fix things would be
424 | > accepted.
425 |
426 | I don't think its enough, either, incidentally. However, my main
427 | concern was trying to get something in to make it more transparent,
428 | and hopefully convince you to come back, either when you appeal or
429 | when your 60 days is up.
430
431 If me coming back is going to involve me having to watch out for
432 vindictive trouble-makers, I don't see any point. We all know fine well
433 that there're some people out there who will use any excuse to try to
434 get me kicked off again.
435
436 | Theres many people in the project who miss
437 | your technical knowledge and the things you have done.
438
439 What, my "net negative contribution" to Gentoo?
440
441 | If theres anything else you would like me to bring up to devrel, I
442 | would be more than happy to.
443
444 Heh, I think they already know my thoughts. Or at least, they know where
445 to find them should they ever decide they want to read what I actually
446 wrote.
447
448 | Also, would you mind if I shared these three emails with devrel to
449 | show that you do indeed have legitimate, reponsible claims? It would
450 | definitely provide proof that you're not just trying to turn the
451 | issues away from yourself.
452
453 Feel free. Heck, print them out and nail them to a church door if you
454 think it'll make a difference.
455
456 -- Ciaran McCreesh Mail: ciaranm at firedrop.org.uk
457
458 --------------
459
460 Since my internet access went out early Monday afternoon, that was the
461 last correspondence I had with him. Any opinions expressed here by me
462 are my personal opinions based on the evidence presented in the bug and
463 elsewhere, and do not reflect upon my opinions of devrel or devrel's
464 members. That said, I would appreciate your comments and suggestions so
465 I can update the proposal and respond to Ciaran accordingly.
466
467 Mike Tindal

Attachments

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