1
00:00:01,030 --> 00:00:02,050
Good afternoon everyone

2
00:00:02,050 --> 00:00:06,610
So, Enrico is a guy with many hats in Debian and

3
00:00:07,740 --> 00:00:10,290
I didn't expect it but he will talk about a new hat:

4
00:00:10,290 --> 00:00:11,890
the "Debian contributor hat"

5
00:00:11,890 --> 00:00:13,840
So, please welcome Enrico.

6
00:00:14,160 --> 00:00:21,420
[applause]

7
00:00:21,420 --> 00:00:23,810
Nobody expects Debian contributors.

8
00:00:24,440 --> 00:00:26,750
Our chief weapons are Debian Developers.

9
00:00:26,750 --> 00:00:30,240
Debian Developers and Debian Maintainers are our two weapons

10
00:00:30,700 --> 00:00:34,070
Debian Developers and Debian Maintainers and Alioth accounts

11
00:00:34,570 --> 00:00:36,510
are our three weapons

12
00:00:36,510 --> 00:00:41,220
and Debian Developers, Debian Maintainers, Alioth accounts and bug reports

13
00:00:41,760 --> 00:00:45,350
are our four amongst our weapons

14
00:00:46,140 --> 00:00:47,580
amongst our weaponry

15
00:00:47,580 --> 00:00:52,190
are such elements as Debian Developers, Debian Maintainers are... I'll come in again.

16
00:00:54,650 --> 00:00:57,460
[comment about the Social Contract]

17
00:01:00,420 --> 00:01:03,510
Nobody expects Debian contributors

18
00:01:03,510 --> 00:01:05,930
amongst our weaponry

19
00:01:05,930 --> 00:01:08,650
are such diverse elements as

20
00:01:08,650 --> 00:01:10,650
Debian Developers, Debian Maintainers

21
00:01:10,650 --> 00:01:12,830
Alioth accounts, bug reporters

22
00:01:13,790 --> 00:01:15,240
and it miss contributors

23
00:01:17,740 --> 00:01:19,470
There's more possible

24
00:01:21,490 --> 00:01:24,780
We clearly need another hat

25
00:01:25,960 --> 00:01:29,620
to make sense of all this.

26
00:01:38,420 --> 00:01:38,980
Right?

27
00:01:42,030 --> 00:01:45,810
so let's make a new hat

28
00:01:47,080 --> 00:01:49,580
I would like to create a hat

29
00:01:49,580 --> 00:01:51,080
of Debian contributors

30
00:01:51,080 --> 00:01:52,910
so that we can say

31
00:01:53,410 --> 00:01:55,520
Nobody expects Debian

32
00:01:55,520 --> 00:01:59,350
our chief weapons are Debian contributors, full stop.

33
00:01:59,790 --> 00:02:04,290
A hat for everyone that contributes to Debian.

34
00:02:05,120 --> 00:02:07,840
Now, the details are here.

35
00:02:10,000 --> 00:02:12,330
This is the file

36
00:02:13,420 --> 00:02:15,110
with notes.

37
00:02:15,110 --> 00:02:17,110
Is it readable at the end?

38
00:02:18,190 --> 00:02:19,110
No?

39
00:02:19,110 --> 00:02:21,490
You have much better eyesight that

40
00:02:21,490 --> 00:02:25,650
No, sorry, I open in a different terminal.

41
00:02:31,570 --> 00:02:35,810
[laughter]

42
00:02:41,690 --> 00:02:42,530
Right.

43
00:02:42,920 --> 00:02:48,380
This is a text file with notes about what I plan to do

44
00:02:48,380 --> 00:02:50,910
I would like to go through it with you

45
00:02:50,910 --> 00:02:52,460
and then

46
00:02:52,460 --> 00:02:54,460
edit it together with you

47
00:02:54,460 --> 00:02:58,510
and at the end of these 45 minutes

48
00:02:58,510 --> 00:03:00,510
have a proposal that

49
00:03:01,970 --> 00:03:04,750
I'll try to implement during DebConf

50
00:03:05,090 --> 00:03:08,210
and fail unless somebody helps me, probably

51
00:03:08,390 --> 00:03:11,440
but what we see how much of this

52
00:03:11,440 --> 00:03:13,540
we want to do, how much of it

53
00:03:13,540 --> 00:03:15,540
can be implemented during DebConf and

54
00:03:15,540 --> 00:03:16,640
at the end of DebConf

55
00:03:17,400 --> 00:03:20,680
I'll make another talk about what we changed in Debian.

56
00:03:21,720 --> 00:03:23,170
So

57
00:03:28,950 --> 00:03:32,230
This is probably a good time to

58
00:03:34,880 --> 00:03:36,910
Ok, so

59
00:03:37,420 --> 00:03:40,080
the problem we want to solve

60
00:03:40,380 --> 00:03:42,570
the problem I want to solve

61
00:03:42,570 --> 00:03:44,570
and you probably also want to solve

62
00:03:44,570 --> 00:03:46,570
once you realise you have a problem

63
00:03:51,490 --> 00:03:54,230
is... words wrapping.

64
00:03:55,600 --> 00:03:58,070
[laughers] Now that we solved word wrapping.

65
00:03:59,430 --> 00:04:01,440
Ok, so

66
00:04:01,440 --> 00:04:04,070
in Debian, we have

67
00:04:04,490 --> 00:04:07,100
lots of contributions that

68
00:04:07,100 --> 00:04:10,240
we do not thank people for

69
00:04:10,490 --> 00:04:14,100
in the sense that we do not acknowledge

70
00:04:14,830 --> 00:04:17,540
that people that did

71
00:04:17,540 --> 00:04:18,830
translations,

72
00:04:18,970 --> 00:04:19,870
worked

73
00:04:20,320 --> 00:04:23,700
we don't acknowledge somewhere people who did translations

74
00:04:23,700 --> 00:04:25,700
or people who reported bugs

75
00:04:25,700 --> 00:04:29,030
There is lots of contributions outside of Debian

76
00:04:29,030 --> 00:04:30,220
but if we look

77
00:04:30,590 --> 00:04:33,440
around package tracking system

78
00:04:33,440 --> 00:04:36,190
Debian developer package overview

79
00:04:36,190 --> 00:04:37,860
it's all package-based

80
00:04:37,990 --> 00:04:40,690
because that's where we come from

81
00:04:41,070 --> 00:04:41,960
and

82
00:04:41,960 --> 00:04:45,740
and so we track uploads that are easy to track

83
00:04:45,740 --> 00:04:49,040
and we acknowledge uploads mostly

84
00:04:50,980 --> 00:04:52,060
but

85
00:04:53,810 --> 00:04:58,830
there's so much more that we do not acknowledge at the moment

86
00:04:58,830 --> 00:05:00,140
Also,

87
00:05:00,670 --> 00:05:02,250
There is a need

88
00:05:02,250 --> 00:05:07,240
Over the time, we created so many ways of contributing to Debian

89
00:05:07,520 --> 00:05:10,630
outside being a Debian Developer

90
00:05:10,770 --> 00:05:11,770
so

91
00:05:12,380 --> 00:05:13,540
the head

92
00:05:13,710 --> 00:05:17,540
of the Italian translations in Debian

93
00:05:17,540 --> 00:05:19,540
is not a Debian Developer

94
00:05:19,680 --> 00:05:21,090
Nobody knows

95
00:05:21,530 --> 00:05:23,090
who she is

96
00:05:24,750 --> 00:05:27,250
Well, I know who she is

97
00:05:27,250 --> 00:05:29,250
Francesca told me

98
00:05:29,250 --> 00:05:33,300
and at least two Debian Developers know who she is

99
00:05:33,550 --> 00:05:36,910
I've no idea who's the head of the

100
00:05:36,910 --> 00:05:39,020
Vietnamese translation

101
00:05:42,400 --> 00:05:47,510
Raise your head if you know who is the head of the Vietnamese translation

102
00:05:47,510 --> 00:05:48,560
in Debian.

103
00:05:48,560 --> 00:05:54,030
[chat in the public]

104
00:05:55,780 --> 00:05:57,550
[public] I know the French one

105
00:05:58,500 --> 00:06:01,770
yeah, we all know who the French one is.

106
00:06:01,770 --> 00:06:04,270
[laughter]

107
00:06:04,320 --> 00:06:09,010
[claps]

108
00:06:09,150 --> 00:06:14,040
But, we totally need a way to credit people

109
00:06:14,090 --> 00:06:15,920
properly.

110
00:06:15,920 --> 00:06:18,530
Debian is so much more than

111
00:06:19,840 --> 00:06:22,450
uploads and Cheese and Wine BoFs

112
00:06:25,660 --> 00:06:28,550
So, there's this idea of creating a hat

113
00:06:28,550 --> 00:06:30,550
of Debian contributor which is

114
00:06:30,550 --> 00:06:33,410
the ultimate do-ocracy hat

115
00:06:33,740 --> 00:06:36,240
There is no bureaucracy involved

116
00:06:36,240 --> 00:06:38,710
If you contribute to Debian

117
00:06:38,710 --> 00:06:41,260
you are a Debian contributor

118
00:06:41,680 --> 00:06:45,980
automatically: there is no new Debian contributor process

119
00:06:47,310 --> 00:06:52,670
If you stop contributing to Debian, you stop being a Debian contributor

120
00:06:52,830 --> 00:06:54,000
That's it.

121
00:06:55,720 --> 00:06:58,240
When you are a Debian contributor

122
00:06:58,240 --> 00:07:00,240
you get such

123
00:07:00,600 --> 00:07:03,150
diverse privileges as

124
00:07:03,150 --> 00:07:05,150
having your name on a list

125
00:07:05,150 --> 00:07:06,480
of Debian contributors

126
00:07:06,480 --> 00:07:08,090
[laughter]

127
00:07:11,590 --> 00:07:16,030
and possibly, if you're not a Debian contributor anymore

128
00:07:16,030 --> 00:07:20,390
then your name will be on a list of Debian contributors in 2012.

129
00:07:22,000 --> 00:07:25,800
If we want to be a bit fancy, when you click

130
00:07:26,020 --> 00:07:27,800
on a name on a list

131
00:07:27,800 --> 00:07:29,960
you get a page about that person

132
00:07:29,960 --> 00:07:31,960
that list what they've done

133
00:07:33,240 --> 00:07:34,900
And that's about it.

134
00:07:37,180 --> 00:07:39,340
That's all I had in mind

135
00:07:39,340 --> 00:07:41,890
You report a bug

136
00:07:41,890 --> 00:07:44,250
you become a Debian contributor

137
00:07:44,250 --> 00:07:47,940
Then maybe everything you do for Debian is reporting a bug

138
00:07:47,940 --> 00:07:51,610
on the 11th of August 2013

139
00:07:51,610 --> 00:07:54,080
then you will be a Debian contributor

140
00:07:54,080 --> 00:07:56,770
from the 11th of August 2013

141
00:07:56,770 --> 00:07:59,180
to the 11th of August 2013

142
00:07:59,180 --> 00:08:00,820
at least you're acknowledged

143
00:08:00,820 --> 00:08:03,090
as that you've helped Debian

144
00:08:23,690 --> 00:08:26,850
Mostly I've said all of this

145
00:08:33,290 --> 00:08:35,060
In order to do this

146
00:08:35,060 --> 00:08:38,450
there needs to be a way of collecting this information

147
00:08:40,280 --> 00:08:44,940
There are data sources we can easily tap into

148
00:08:44,940 --> 00:08:46,940
such as

149
00:08:47,300 --> 00:08:48,660
package uploads

150
00:08:53,040 --> 00:08:57,070
but we can see mailing list traffic, easily

151
00:08:57,260 --> 00:08:59,070
We can get

152
00:08:59,230 --> 00:09:02,480
bug tracking system activity hopefully easily

153
00:09:03,870 --> 00:09:06,920
We can hook into

154
00:09:06,920 --> 00:09:10,360
commit logs of Alioth

155
00:09:13,580 --> 00:09:15,470
version control systems

156
00:09:15,470 --> 00:09:17,580
[phone ring]

157
00:09:17,580 --> 00:09:18,490
[phone's owner] Sorry

158
00:09:18,490 --> 00:09:19,460
All right

159
00:09:19,460 --> 00:09:22,820
At least I don't have to answer

160
00:09:29,060 --> 00:09:32,980
And, this system, whenever

161
00:09:32,980 --> 00:09:36,280
at any point of its usage

162
00:09:36,280 --> 00:09:38,060
would be unfair

163
00:09:38,060 --> 00:09:41,770
because there will always be some bit of Debian that

164
00:09:41,770 --> 00:09:44,830
isn't yet providing data

165
00:09:44,830 --> 00:09:47,800
for the list.

166
00:09:48,490 --> 00:09:52,400
However, the point is that by having such a system in place

167
00:09:52,400 --> 00:09:53,820
we create

168
00:09:54,070 --> 00:09:57,200
a proper way of acknoledging people's work

169
00:09:57,480 --> 00:10:02,640
and we give motivations for teams to contribute

170
00:10:03,390 --> 00:10:06,300
contribution information to the list

171
00:10:10,630 --> 00:10:12,190
Questions so far?

172
00:10:28,200 --> 00:10:31,060
[Q] If you do this automatically

173
00:10:31,060 --> 00:10:33,560
which we can here assume you want to

174
00:10:33,560 --> 00:10:37,080
wouldn't that raise privacy concerns?

175
00:10:37,190 --> 00:10:38,970
Sorry, I did not understand the second part

176
00:10:39,600 --> 00:10:43,430
[Q] Wouldn't this raise privacy concerns?

177
00:10:43,550 --> 00:10:44,930
Good question.

178
00:10:44,930 --> 00:10:46,930
The question is,

179
00:10:46,930 --> 00:10:50,790
if we collect this data automatically, wouldn't there be privacy concerns?

180
00:10:50,790 --> 00:10:52,790
Yes, possibly

181
00:10:54,120 --> 00:10:55,760
To be discussed.

182
00:10:58,030 --> 00:11:01,030
All the information presented here

183
00:11:01,030 --> 00:11:04,110
is already public

184
00:11:05,750 --> 00:11:08,630
There is a public record of package uploads,

185
00:11:08,630 --> 00:11:11,100
mailing lists are publicly archived

186
00:11:11,680 --> 00:11:12,680
but

187
00:11:12,850 --> 00:11:15,710
the BTS is fully in the open

188
00:11:16,620 --> 00:11:18,900
commit logs on Alioth are public

189
00:11:18,900 --> 00:11:18,950
I would only mine
commit logs on Alioth are public

190
00:11:18,950 --> 00:11:20,560
I would only mine

191
00:11:20,560 --> 00:11:24,560
or get data from sources that are either public

192
00:11:24,560 --> 00:11:26,560
or that people agree

193
00:11:26,560 --> 00:11:28,000
to disclose

194
00:11:28,330 --> 00:11:32,110
However, it's the usual issue of

195
00:11:32,910 --> 00:11:36,440
the information is public but not presented in that way

196
00:11:36,440 --> 00:11:37,600
for example

197
00:11:38,430 --> 00:11:39,210
when

198
00:11:44,040 --> 00:11:45,840
when

199
00:11:45,840 --> 00:11:50,980
I collected information from Debian changelogs to

200
00:11:51,980 --> 00:11:56,030
see the history of people's contribution in the process

201
00:11:56,250 --> 00:12:00,410
I did not make that information publicly available

202
00:12:00,410 --> 00:12:02,720
because it's a bit

203
00:12:03,740 --> 00:12:05,990
I would like to have a discussion

204
00:12:05,990 --> 00:12:09,130
whether we want to present that information in

205
00:12:09,270 --> 00:12:11,370
collated in such a way

206
00:12:11,960 --> 00:12:13,340
So yes

207
00:12:14,430 --> 00:12:15,840
that is an issue

208
00:12:15,840 --> 00:12:19,890
My general idea would be that

209
00:12:22,890 --> 00:12:25,110
once the system goes public

210
00:12:25,110 --> 00:12:28,140
there should be a way for people to opt-out

211
00:12:28,140 --> 00:12:31,100
of having information about them displayed

212
00:12:31,910 --> 00:12:35,100
Most of the privacy scenario...

213
00:12:35,100 --> 00:12:37,490
so, generally people

214
00:12:37,790 --> 00:12:39,150
I would understand the general...

215
00:12:39,150 --> 00:12:41,210
The default would be that somebody likes

216
00:12:41,210 --> 00:12:43,210
their work to be credited

217
00:12:44,120 --> 00:12:46,450
However, there may be people

218
00:12:46,700 --> 00:12:47,810
that

219
00:12:49,060 --> 00:12:53,860
would have issues if some of their work is credited

220
00:12:57,690 --> 00:13:00,460
Most things revolve around

221
00:13:02,020 --> 00:13:05,850
job hunting like recruiters looking at the internet

222
00:13:06,790 --> 00:13:08,980
and you either want to be found

223
00:13:08,980 --> 00:13:11,180
or if you don't want to be found

224
00:13:11,180 --> 00:13:14,340
you already have a problem when contributing to Debian.

225
00:13:14,450 --> 00:13:18,090
I need to be careful about contributing with a different name or something.

226
00:13:20,360 --> 00:13:21,660
But, yes

227
00:13:26,490 --> 00:13:28,320
my assumption is that

228
00:13:28,630 --> 00:13:32,070
It's not a big privacy issue to get it started

229
00:13:32,070 --> 00:13:33,540
as long as

230
00:13:34,650 --> 00:13:36,930
we give people the opportunity to opt-out

231
00:13:38,230 --> 00:13:41,620
Does it seem reasonable?

232
00:13:46,580 --> 00:13:48,800
[Q] Maybe with a opt-in

233
00:13:50,330 --> 00:13:52,520
Maybe with a opt-in

234
00:13:52,520 --> 00:13:55,080
By default it's not opt-in

235
00:13:55,270 --> 00:13:57,410
I would be fine with opt-in

236
00:14:04,120 --> 00:14:07,400
I wonder how many people would know

237
00:14:07,400 --> 00:14:10,980
that they can opt-in and how to opt-in

238
00:14:13,830 --> 00:14:15,110
At the moment

239
00:14:17,410 --> 00:14:21,720
the current Debian infrastructure is not even opt-out

240
00:14:21,720 --> 00:14:24,100
You can't say "Please don't show

241
00:14:24,100 --> 00:14:26,770
the messages that I sent to Debian mailing lists"

242
00:14:30,790 --> 00:14:35,310
I would see surprising that such a thing would be opt-in

243
00:14:35,730 --> 00:14:39,120
because anything is already out by default

244
00:14:40,200 --> 00:14:41,700
On the other hand

245
00:14:42,310 --> 00:14:44,250
I could be convinced

246
00:14:44,250 --> 00:14:46,250
about making it opt-in

247
00:14:47,440 --> 00:14:49,050
It's seems to be

248
00:14:49,050 --> 00:14:51,380
at the moment something that would

249
00:14:52,160 --> 00:14:53,820
impact

250
00:14:54,520 --> 00:14:56,710
the start of such project

251
00:14:56,710 --> 00:14:58,120
So, before

252
00:14:58,120 --> 00:15:00,070
somebody gets a useful list

253
00:15:00,070 --> 00:15:02,560
I would need to go and bother a thousand people

254
00:15:02,560 --> 00:15:05,670
to opt-in the system so that at least can get populated

255
00:15:06,870 --> 00:15:07,670
Yes?

256
00:15:13,330 --> 00:15:16,770
[Q] I think you should also look at legal

257
00:15:17,520 --> 00:15:21,350
aspects, you are collecting personal information

258
00:15:21,820 --> 00:15:25,430
as far... I don't know the law

259
00:15:25,760 --> 00:15:27,710
by heart

260
00:15:27,710 --> 00:15:29,710
but as far as I understand it

261
00:15:29,710 --> 00:15:30,920
it would be illegal

262
00:15:30,920 --> 00:15:33,230
in half of european countries

263
00:15:33,230 --> 00:15:35,230
Then Debian is already illegal

264
00:15:35,230 --> 00:15:37,230
No! Currently

265
00:15:37,230 --> 00:15:39,800
you are not collecting this information

266
00:15:39,800 --> 00:15:40,750
Yes, we are

267
00:15:40,750 --> 00:15:43,360
You have it in different pieces

268
00:15:43,360 --> 00:15:45,360
and at different places

269
00:15:51,930 --> 00:15:54,180
and it start

270
00:15:54,180 --> 00:15:56,620
The package tracking system would be illegal

271
00:15:56,620 --> 00:15:59,400
No it's not illegal

272
00:15:59,400 --> 00:16:02,340
You don't collect information there

273
00:16:03,670 --> 00:16:05,580
collated about a person

274
00:16:06,110 --> 00:16:08,110
There, you have it

275
00:16:08,110 --> 00:16:13,630
to follow specifics purpose

276
00:16:14,240 --> 00:16:18,150
and this is not illegal, but if you collect it

277
00:16:18,710 --> 00:16:22,150
and collect all information for a specific

278
00:16:22,150 --> 00:16:24,260
person or all persons

279
00:16:24,260 --> 00:16:25,650
and collate it

280
00:16:25,840 --> 00:16:29,620
this could possibly be illegal as far as I understand it

281
00:16:30,030 --> 00:16:31,620
I would like

282
00:16:32,640 --> 00:16:34,470
You said that you are not sure

283
00:16:34,470 --> 00:16:36,830
I would like to talk with somebody who is sure

284
00:16:37,050 --> 00:16:39,490
because we already have things

285
00:16:39,800 --> 00:16:40,600
that

286
00:16:40,910 --> 00:16:43,100
would potentially be illegal in that view

287
00:16:43,100 --> 00:16:44,910
like the package tracking system

288
00:16:44,910 --> 00:16:46,460
goes beyond

289
00:16:46,460 --> 00:16:48,650
listing of some of the uploaders

290
00:16:48,650 --> 00:16:50,840
because it collects information

291
00:16:50,840 --> 00:16:54,840
about everything visible about the person, bugs and so on

292
00:16:55,950 --> 00:16:57,390
[Q] Would it be

293
00:16:57,390 --> 00:16:59,390
a preset that is useful to

294
00:16:59,390 --> 00:17:01,720
structure this kind of things by, I mean

295
00:17:01,720 --> 00:17:02,780
some

296
00:17:02,780 --> 00:17:04,660
reasonable set of people

297
00:17:04,660 --> 00:17:06,770
listed as contributors

298
00:17:06,770 --> 00:17:10,550
You could start by assuming that

299
00:17:10,550 --> 00:17:12,550
people who have said

300
00:17:13,180 --> 00:17:14,900
"I would like to be a Debian Developer"

301
00:17:14,900 --> 00:17:16,900
or "I would like to be a Debian member"

302
00:17:17,120 --> 00:17:20,400
have implicitely given their authorisation

303
00:17:20,400 --> 00:17:21,980
to be listed as Debian contributors.

304
00:17:21,980 --> 00:17:23,980
It's a subset

305
00:17:25,170 --> 00:17:28,560
that we might have to run up by a lawyer

306
00:17:28,560 --> 00:17:30,140
but it seems a bit more reasonable than

307
00:17:30,140 --> 00:17:32,140
??????

308
00:17:33,770 --> 00:17:35,080
I was just gonna say I think

309
00:17:35,080 --> 00:17:38,130
the legal thing should probably be a separated BoF

310
00:17:38,130 --> 00:17:39,300
because

311
00:17:39,910 --> 00:17:41,850
in the practical terms if you're only worried

312
00:17:41,850 --> 00:17:43,240
from the legal point of view, you just need to

313
00:17:43,240 --> 00:17:45,400
get someone who is in Debian in the US

314
00:17:45,400 --> 00:17:47,400
to host a US server because the US

315
00:17:47,400 --> 00:17:49,760
doesn't basically have this kind of laws

316
00:17:51,230 --> 00:17:53,730
It is rather missing the point.

317
00:17:53,730 --> 00:17:55,730
Exactly, the relevant question is I think

318
00:17:55,730 --> 00:17:57,730
no about the legal thing it's a kind of ???

319
00:17:57,730 --> 00:18:01,160
but about what we want to do and what we think is fair and so on

320
00:18:01,160 --> 00:18:03,630
Well I keep thinking ??? something

321
00:18:03,630 --> 00:18:06,410
much more lightweight than launchpad

322
00:18:06,410 --> 00:18:08,100
is doing

323
00:18:12,100 --> 00:18:17,870
Another point of, another way to mailing list would be to

324
00:18:18,650 --> 00:18:20,370
do this automatism

325
00:18:20,370 --> 00:18:22,670
without publishing the data but

326
00:18:22,890 --> 00:18:25,750
with afterwards sending

327
00:18:25,920 --> 00:18:28,110
about once a year or

328
00:18:28,110 --> 00:18:30,190
even possibly less

329
00:18:31,130 --> 00:18:35,070
a mail to every contributor if he wants to be listed

330
00:18:35,070 --> 00:18:38,430
so we have an opt-in instead of an opt-out

331
00:18:38,540 --> 00:18:40,400
If we do it seldomly enough

332
00:18:40,400 --> 00:18:42,900
it would not even be so

333
00:18:42,900 --> 00:18:44,400
stressing people

334
00:18:44,400 --> 00:18:46,010
Yeah, that makes sense

335
00:18:46,230 --> 00:18:47,590
You don't need to wait a year,

336
00:18:47,590 --> 00:18:50,720
you can do it the first time you detect someone new

337
00:18:50,720 --> 00:18:52,440
someone starts contributing

338
00:18:52,440 --> 00:18:55,640
we can send "hello, you did something in Debian

339
00:18:55,640 --> 00:18:57,380
we would like to credit you"

340
00:18:57,500 --> 00:18:59,100
Wait for a month.

341
00:18:59,100 --> 00:19:01,660
Wait for a month after the first contribution

342
00:19:01,660 --> 00:19:04,160
if you get more, then

343
00:19:04,430 --> 00:19:07,490
say "Do you want to be thanked?"

344
00:19:07,490 --> 00:19:09,090
Makes sense

345
00:19:09,090 --> 00:19:12,730
[Q] Even without waiting. We can do as the Linux kernel

346
00:19:13,980 --> 00:19:16,390
you knows the stats about

347
00:19:16,390 --> 00:19:18,810
which companies contribute to the Linux kernel?

348
00:19:19,110 --> 00:19:22,140
It's like that, they check for new commits

349
00:19:22,140 --> 00:19:24,520
and just send an email to ask affiliation

350
00:19:24,660 --> 00:19:27,990
There's other systems that calculate karma

351
00:19:29,050 --> 00:19:30,880
Those are illegal because

352
00:19:30,880 --> 00:19:32,680
then they produce

353
00:19:32,680 --> 00:19:35,600
information that could be used to rank people

354
00:19:36,150 --> 00:19:38,290
from a job point of view

355
00:19:38,290 --> 00:19:39,950
[from audience] there is Google

356
00:19:40,700 --> 00:19:44,230
The other... yeah, well.

357
00:19:44,610 --> 00:19:47,670
I'm not that concerned with that regard at the moment

358
00:19:47,670 --> 00:19:49,860
[Q] Would that credit

359
00:19:50,330 --> 00:19:52,440
only be for

360
00:19:52,440 --> 00:19:56,410
persons, or also for organisations or companies?

361
00:19:56,410 --> 00:19:57,770
I'm

362
00:19:58,270 --> 00:20:00,740
interested about

363
00:20:01,850 --> 00:20:03,850
entities

364
00:20:07,040 --> 00:20:09,530
directly contributing to Debian

365
00:20:09,980 --> 00:20:12,170
So, if

366
00:20:12,950 --> 00:20:17,670
there's some email and GPG key

367
00:20:17,890 --> 00:20:21,970
which belongs to an entity called

368
00:20:24,300 --> 00:20:26,180
FooBar ltd

369
00:20:26,180 --> 00:20:29,380
then FooBar ltd will be credited

370
00:20:29,380 --> 00:20:31,230
I'm not interested

371
00:20:31,230 --> 00:20:33,730
that there is a real person behind this

372
00:20:33,730 --> 00:20:35,730
and not interested

373
00:20:35,730 --> 00:20:37,730
that's the real name of a person

374
00:20:38,030 --> 00:20:39,890
It's just whatever

375
00:20:40,590 --> 00:20:42,780
is chosen for contribution

376
00:20:45,550 --> 00:20:48,050
that we only check identities

377
00:20:48,050 --> 00:20:50,050
when we have to

378
00:20:50,050 --> 00:20:52,570
which is when we give people upload rights

379
00:20:52,570 --> 00:20:54,820
because we want to know where they live

380
00:20:54,820 --> 00:20:56,930
and go there to tickle them

381
00:20:57,070 --> 00:20:58,820
[laughter]

382
00:20:59,680 --> 00:21:01,250
I think that

383
00:21:01,250 --> 00:21:04,340
the main index would be the e-mail

384
00:21:05,700 --> 00:21:10,580
Even behind the scenes... But the main index would be the email?

385
00:21:10,580 --> 00:21:12,090
That is one issue

386
00:21:12,260 --> 00:21:13,980
That is an issue

387
00:21:14,670 --> 00:21:17,010
because depending on the data sources

388
00:21:17,010 --> 00:21:19,010
the index could be an email

389
00:21:19,290 --> 00:21:21,010
in the BTS for example

390
00:21:21,150 --> 00:21:22,330
an Alioth

391
00:21:22,690 --> 00:21:23,710
account

392
00:21:23,840 --> 00:21:25,920
for commits

393
00:21:25,920 --> 00:21:27,470
in Alioth

394
00:21:27,470 --> 00:21:28,650
or

395
00:21:28,650 --> 00:21:30,870
Debian Developer login name

396
00:21:30,870 --> 00:21:32,870
for sponsoring uploads

397
00:21:34,180 --> 00:21:34,870
so

398
00:21:35,210 --> 00:21:36,650
possibly, that

399
00:21:36,810 --> 00:21:38,510
that could be some

400
00:21:38,510 --> 00:21:41,220
need of mixing things at some point

401
00:21:41,430 --> 00:21:44,160
which is a bit of a separate problem

402
00:21:44,530 --> 00:21:46,460
I don't mind starting

403
00:21:46,460 --> 00:21:48,460
with that not being perfect

404
00:21:48,460 --> 00:21:50,340
or being fixed by hand

405
00:21:51,560 --> 00:21:53,670
And if we evolve at some point

406
00:21:53,670 --> 00:21:55,460
there may be a way for people

407
00:21:55,460 --> 00:21:57,630
to log in and prove that

408
00:21:57,630 --> 00:21:59,220
they control that GPG key

409
00:21:59,220 --> 00:22:02,020
or that e-mail address or that Alioth account

410
00:22:02,020 --> 00:22:04,470
you know, sending e-mail challenge saying "that's my e-mail"

411
00:22:04,620 --> 00:22:05,690
but ok, it is

412
00:22:05,690 --> 00:22:07,390
yes, ok, then I merge

413
00:22:07,560 --> 00:22:09,260
the two sets of contributions

414
00:22:10,810 --> 00:22:12,600
possibly for next year

415
00:22:12,600 --> 00:22:15,140
because it starts being a bit heavy weight

416
00:22:15,140 --> 00:22:18,020
but there's a problem of doing this collation

417
00:22:18,020 --> 00:22:19,520
it is a wider something that

418
00:22:19,520 --> 00:22:21,520
greatly help the MIA team

419
00:22:23,780 --> 00:22:25,180
So this

420
00:22:25,180 --> 00:22:27,180
in the end can be something

421
00:22:27,180 --> 00:22:29,180
used to detect when people

422
00:22:29,890 --> 00:22:31,760
are not contributing anymore

423
00:22:31,760 --> 00:22:33,830
I would also

424
00:22:33,830 --> 00:22:35,590
want to make it so that

425
00:22:35,590 --> 00:22:37,400
if you are a Debian Developer

426
00:22:37,400 --> 00:22:39,040
you are not automatically

427
00:22:39,040 --> 00:22:40,610
a Debian contributor

428
00:22:40,610 --> 00:22:46,610
[laughter]

429
00:22:46,870 --> 00:22:48,610
If you are not a Debian Developer

430
00:22:48,610 --> 00:22:50,610
you are not automatically a Debian contributor

431
00:22:50,610 --> 00:22:52,830
you are a Debian contributor only

432
00:22:52,930 --> 00:22:54,610
if you contribute to Debian

433
00:22:54,970 --> 00:22:56,740
If you are a Debian Developer that

434
00:22:56,740 --> 00:22:58,740
does not contribute to Debian

435
00:22:58,740 --> 00:23:00,290
then you are not on that list

436
00:23:00,290 --> 00:23:02,290
You keep being a Debian Developer

437
00:23:02,590 --> 00:23:03,840
There's the usual

438
00:23:03,840 --> 00:23:05,840
I'm not interested in changing the rules

439
00:23:05,970 --> 00:23:06,640
for

440
00:23:06,760 --> 00:23:08,740
getting removed from

441
00:23:08,740 --> 00:23:09,500
Debian keyring

442
00:23:09,500 --> 00:23:11,350
I think what we have is fine.

443
00:23:12,400 --> 00:23:13,690
But

444
00:23:14,230 --> 00:23:16,060
it's perfectly right if

445
00:23:16,060 --> 00:23:18,180
if I'm not active in Debian for a year

446
00:23:18,190 --> 00:23:19,740
but I still care about the project

447
00:23:19,740 --> 00:23:21,330
involved, and follow things

448
00:23:21,330 --> 00:23:22,730
but I don't contribute anymore

449
00:23:22,730 --> 00:23:24,980
then I'm perfectly happy that I'm not listed

450
00:23:24,980 --> 00:23:26,980
among the contributors for that year

451
00:23:29,870 --> 00:23:31,390
There was a hand?

452
00:23:31,390 --> 00:23:35,120
[Q] Yes, I'm relaying a comment from IRC

453
00:23:36,380 --> 00:23:37,350
Wouldn't

454
00:23:38,900 --> 00:23:41,650
recognising every small thing like

455
00:23:41,650 --> 00:23:44,060
just one bug report

456
00:23:44,280 --> 00:23:46,670
dillute the

457
00:23:48,260 --> 00:23:49,640
the weight

458
00:23:49,940 --> 00:23:52,190
of a Debian contributor

459
00:23:52,860 --> 00:23:54,020
that so

460
00:23:54,020 --> 00:23:55,460
are there Debian contributors like

461
00:23:55,460 --> 00:23:57,700
the head of a translation team

462
00:23:57,900 --> 00:23:59,700
be equal to

463
00:23:59,700 --> 00:24:00,990
one bug report

464
00:24:00,990 --> 00:24:03,510
That's the comment, I'm not saying that I agree

465
00:24:03,530 --> 00:24:04,820
The way I see it

466
00:24:05,010 --> 00:24:06,500
again, we can

467
00:24:06,500 --> 00:24:08,030
you can tell me if you don't like this

468
00:24:08,030 --> 00:24:10,310
but the way I see it, they are the same

469
00:24:13,710 --> 00:24:15,060
They are Debian contributors

470
00:24:15,060 --> 00:24:16,890
maybe one has been a Debian contributor

471
00:24:16,890 --> 00:24:18,890
for much longer than the other

472
00:24:19,170 --> 00:24:21,190
so the time span

473
00:24:21,190 --> 00:24:23,940
of the contribution is the only extra

474
00:24:24,350 --> 00:24:25,840
bit. Or you can click on a

475
00:24:25,840 --> 00:24:27,840
a person and list their contributions

476
00:24:30,270 --> 00:24:32,100
But, I don't want to

477
00:24:32,100 --> 00:24:33,970
calculate a number

478
00:24:33,970 --> 00:24:36,680
of how much one is a contributor

479
00:24:39,260 --> 00:24:42,040
I don't see

480
00:24:42,060 --> 00:24:44,210
why one would want to

481
00:24:44,570 --> 00:24:46,210
rank people

482
00:24:46,210 --> 00:24:47,840
by contribution

483
00:24:47,840 --> 00:24:49,110
I'm not interested in that

484
00:24:49,110 --> 00:24:50,850
I think it creates a kind of

485
00:24:50,850 --> 00:24:52,850
community where I don't want to be in

486
00:24:54,940 --> 00:24:58,000
personally, I mean, that would be my personal feeling

487
00:24:59,310 --> 00:25:00,980
But I would be able to

488
00:25:00,980 --> 00:25:02,980
say "thank you" even if

489
00:25:03,520 --> 00:25:05,160
a person reports a bug

490
00:25:05,160 --> 00:25:07,500
It takes time to report a bug

491
00:25:07,750 --> 00:25:09,500
There was an idea

492
00:25:10,350 --> 00:25:12,100
talking about this

493
00:25:12,900 --> 00:25:14,390
informally earlier

494
00:25:14,560 --> 00:25:15,560
of

495
00:25:15,930 --> 00:25:18,440
having some data sources used

496
00:25:18,440 --> 00:25:19,970
only to compute

497
00:25:19,970 --> 00:25:21,500
the time span but not

498
00:25:21,500 --> 00:25:23,260
the status of contributors

499
00:25:23,260 --> 00:25:25,750
for example mailing list traffic

500
00:25:27,580 --> 00:25:28,950
Maybe people don't like the idea

501
00:25:28,950 --> 00:25:30,810
that if you send a e-mail to

502
00:25:30,810 --> 00:25:32,810
a mailing list you are Debian contributors

503
00:25:33,350 --> 00:25:34,970
There's indeed many people

504
00:25:34,970 --> 00:25:37,060
who send lots of e-mails to Debian

505
00:25:37,060 --> 00:25:38,130
mailing lists

506
00:25:38,210 --> 00:25:39,870
but do not contribute

507
00:25:39,870 --> 00:25:42,650
[Laughter] to Debian

508
00:25:43,910 --> 00:25:45,090
at all

509
00:25:46,390 --> 00:25:47,980
However, if somebody

510
00:25:47,980 --> 00:25:50,690
is a contributor for some other reason

511
00:25:51,010 --> 00:25:52,320
then it makes sense to

512
00:25:52,320 --> 00:25:54,320
look at mailing list involvement

513
00:25:54,320 --> 00:25:56,520
say: "Well, you reported a bug today

514
00:25:56,520 --> 00:25:58,870
but you've been active in mailing lists

515
00:25:58,870 --> 00:25:59,780
for a year"

516
00:25:59,780 --> 00:26:01,780
so you are a Debian contributor because

517
00:26:02,310 --> 00:26:03,590
you reported a bug

518
00:26:03,590 --> 00:26:05,700
and you've been a Debian contributor for a year

519
00:26:05,700 --> 00:26:06,620
because

520
00:26:06,620 --> 00:26:09,070
you've been into Debian for a long time

521
00:26:10,100 --> 00:26:11,260
so splitting

522
00:26:12,150 --> 00:26:12,860
the...

523
00:26:12,860 --> 00:26:14,480
I think some data sources

524
00:26:14,480 --> 00:26:16,480
are only useful computing the time

525
00:26:16,480 --> 00:26:18,090
and not the status

526
00:26:20,740 --> 00:26:23,020
I would like to add something about

527
00:26:23,020 --> 00:26:25,110
the issue with detecting the people

528
00:26:25,320 --> 00:26:28,470
because we had some experiences in the team maintenance thing

529
00:26:28,630 --> 00:26:30,320
and, in UDD you have

530
00:26:30,320 --> 00:26:32,250
this Carnivore database

531
00:26:32,250 --> 00:26:33,760
which is based on the

532
00:26:34,130 --> 00:26:36,050
key fingerprint

533
00:26:36,050 --> 00:26:37,260
and we even have

534
00:26:37,260 --> 00:26:38,730
for these people

535
00:26:39,330 --> 00:26:40,900
somebody with five

536
00:26:40,900 --> 00:26:41,950
different names

537
00:26:41,950 --> 00:26:44,170
names spellings and so on

538
00:26:44,440 --> 00:26:46,960
people are using different e-mail addresses so

539
00:26:46,960 --> 00:26:48,790
I really really really doubt

540
00:26:48,850 --> 00:26:50,860
you can manage this for

541
00:26:51,340 --> 00:26:54,070
say, more than 500 people

542
00:26:56,060 --> 00:26:57,270
because it's

543
00:26:57,270 --> 00:26:59,970
manual work and we tried it

544
00:26:59,980 --> 00:27:01,410
it's hard to cope with this

545
00:27:01,410 --> 00:27:04,110
No no, I don't want to do manual work

546
00:27:05,170 --> 00:27:07,160
Yeah but, ok

547
00:27:07,160 --> 00:27:08,850
Good luck with automatic detection

548
00:27:08,850 --> 00:27:10,230
No, I don't want to do

549
00:27:10,230 --> 00:27:12,920
that much automatic detection either

550
00:27:14,700 --> 00:27:15,690
I would

551
00:27:15,850 --> 00:27:16,950
like

552
00:27:18,260 --> 00:27:20,990
things to be fixed as much as possible

553
00:27:20,990 --> 00:27:22,260
and

554
00:27:22,630 --> 00:27:24,960
at lower level of data sources

555
00:27:25,230 --> 00:27:26,860
When that is not possible

556
00:27:26,860 --> 00:27:28,860
I would like to offer people a way to

557
00:27:28,860 --> 00:27:30,550
fix the data for themselves

558
00:27:32,690 --> 00:27:34,010
There's that

559
00:27:34,270 --> 00:27:35,750
unpronounceable and

560
00:27:35,750 --> 00:27:37,500
evil web site that tracks

561
00:27:37,500 --> 00:27:39,000
free software developers

562
00:27:39,000 --> 00:27:41,590
that in my opinion should be illegal but isn't

563
00:27:42,090 --> 00:27:44,110
it starts with "o", ohlo?

564
00:27:45,620 --> 00:27:48,120
"O-l-o-h-o" something like that

565
00:27:48,960 --> 00:27:50,470
I actually had an argument with

566
00:27:50,470 --> 00:27:51,870
their CEO, saying

567
00:27:51,870 --> 00:27:53,610
I've asked to opt out

568
00:27:53,610 --> 00:27:55,890
of their system and they told me to fuck off

569
00:27:57,500 --> 00:27:58,350
So

570
00:27:59,280 --> 00:28:01,380
talking about privacy of these things

571
00:28:01,390 --> 00:28:03,280
I guess if they can do what they do

572
00:28:03,280 --> 00:28:05,480
I think we don't have a problem to say "thank you"

573
00:28:05,480 --> 00:28:06,850
to a bunch of people

574
00:28:10,650 --> 00:28:12,030
They do of

575
00:28:12,030 --> 00:28:14,020
they have a mess, their dataset is a

576
00:28:14,030 --> 00:28:14,870
a mess

577
00:28:14,940 --> 00:28:16,350
but people still like them

578
00:28:16,680 --> 00:28:17,880
I'm listed about

579
00:28:17,880 --> 00:28:19,880
twenty times in their system

580
00:28:20,260 --> 00:28:22,540
and if I really care about

581
00:28:22,830 --> 00:28:25,460
contributing them my identity for free

582
00:28:25,880 --> 00:28:27,390
for them to send it out

583
00:28:27,390 --> 00:28:28,850
then I

584
00:28:29,570 --> 00:28:30,940
can log in and

585
00:28:30,940 --> 00:28:32,940
merge these identities for them

586
00:28:33,760 --> 00:28:34,650
I guess

587
00:28:34,650 --> 00:28:36,100
that's something we can offer

588
00:28:37,890 --> 00:28:39,010
and

589
00:28:39,160 --> 00:28:41,010
and that can improve things

590
00:28:41,010 --> 00:28:43,010
like carnivore and so on

591
00:28:43,010 --> 00:28:45,320
and possibly we can have

592
00:28:45,940 --> 00:28:48,120
personal homepage...

593
00:28:48,710 --> 00:28:50,750
for contributors

594
00:28:52,270 --> 00:28:53,820
[Q] Where they actively

595
00:28:53,960 --> 00:28:56,440
contribute to their identities?

596
00:28:57,640 --> 00:28:58,910
At least we've done

597
00:28:58,910 --> 00:29:00,910
all we can to thank them

598
00:29:01,070 --> 00:29:01,750
but

599
00:29:04,900 --> 00:29:07,210
Well, you know, if I tell you "thank you"

600
00:29:07,780 --> 00:29:09,890
I don't expect you to say

601
00:29:10,930 --> 00:29:14,030
"You didn't thank me in a an appropriate way"

602
00:29:14,440 --> 00:29:15,210
Right?

603
00:29:17,230 --> 00:29:18,350
On the other hand

604
00:29:20,490 --> 00:29:22,120
I'm ok if people say

605
00:29:22,120 --> 00:29:24,390
"Well, I'm also that person

606
00:29:24,390 --> 00:29:27,070
you don't need to thank me so many times"

607
00:29:29,430 --> 00:29:30,210
and that

608
00:29:30,210 --> 00:29:33,080
that's fine, but, on the other hand

609
00:29:33,200 --> 00:29:35,180
I'm perfectly happy if

610
00:29:35,180 --> 00:29:36,530
people want

611
00:29:36,700 --> 00:29:38,500
to use multiple identities

612
00:29:38,500 --> 00:29:40,170
for contributing to Debian

613
00:29:40,330 --> 00:29:42,610
Somebody in Debian science

614
00:29:42,780 --> 00:29:46,130
may want to contribute to Debian games under a different name

615
00:29:49,790 --> 00:29:52,480
For similar reasons, I don't want to trunk

616
00:29:53,550 --> 00:29:56,110
the time frame of contributions

617
00:29:56,110 --> 00:29:58,750
below the month level

618
00:29:59,780 --> 00:30:00,790
I don't want

619
00:30:00,960 --> 00:30:01,830
to say

620
00:30:01,980 --> 00:30:05,000
"You have contributed to Debian games

621
00:30:05,100 --> 00:30:06,100
between

622
00:30:07,530 --> 00:30:12,050
10am to 11:30am on a Thursday

623
00:30:12,050 --> 00:30:13,450
when you were at a meeting

624
00:30:13,450 --> 00:30:14,660
[laughter]

625
00:30:14,660 --> 00:30:15,020
Right?

626
00:30:17,020 --> 00:30:18,460
So, it's

627
00:30:18,460 --> 00:30:20,460
I want it to be coarse

628
00:30:20,820 --> 00:30:23,860
It really needs to be mostly about thank you

629
00:30:23,860 --> 00:30:25,860
and about building reputation

630
00:30:28,100 --> 00:30:31,210
Reputation is also what we can reward people with

631
00:30:31,210 --> 00:30:33,210
which is kind of the point of the exercise

632
00:30:33,640 --> 00:30:36,200
and that reputation is nice to acknowledge

633
00:30:36,200 --> 00:30:37,920
and at the same time it's nice

634
00:30:38,330 --> 00:30:39,480
that we can

635
00:30:39,580 --> 00:30:41,510
look up someone's reputation

636
00:30:41,510 --> 00:30:42,170
at least

637
00:30:42,670 --> 00:30:45,590
as a Debian account manager and Front Desk member

638
00:30:45,680 --> 00:30:49,640
if I can see what's somebody's reputation and have a

639
00:30:50,110 --> 00:30:51,390
reasonable look

640
00:30:51,510 --> 00:30:54,930
then I can make the process much swifter for them

641
00:30:55,580 --> 00:30:58,510
and especially for non uploading DDs

642
00:30:58,510 --> 00:31:00,320
it's currently very hard

643
00:31:00,600 --> 00:31:01,570
because

644
00:31:03,670 --> 00:31:06,130
I can go and look at package changelogs

645
00:31:06,130 --> 00:31:08,690
to see that they have been active for ten years in the project

646
00:31:10,250 --> 00:31:11,100
and

647
00:31:11,100 --> 00:31:13,510
if somebody is a non uploading DD

648
00:31:13,510 --> 00:31:15,980
that has done translations at the moment

649
00:31:16,690 --> 00:31:19,000
I have very hard time seeing

650
00:31:19,000 --> 00:31:20,280
what they've done

651
00:31:20,450 --> 00:31:22,280
possibly I don't understand the language

652
00:31:26,690 --> 00:31:28,130
So, well

653
00:31:28,600 --> 00:31:30,050
and that makes

654
00:31:30,310 --> 00:31:31,910
makes it easier and

655
00:31:32,200 --> 00:31:33,910
that another

656
00:31:34,330 --> 00:31:36,070
outcome that I would really want to see

657
00:31:36,070 --> 00:31:37,430
out of this is that

658
00:31:37,430 --> 00:31:39,670
we start to actually see

659
00:31:40,050 --> 00:31:41,370
in front of us

660
00:31:41,370 --> 00:31:43,210
that Debian is not just

661
00:31:43,210 --> 00:31:44,870
about technical development

662
00:31:46,460 --> 00:31:47,900
There's a lot more

663
00:31:48,140 --> 00:31:50,520
but if people still perceive

664
00:31:50,520 --> 00:31:52,170
"Oh, I'd like to contribute to Debian but

665
00:31:52,170 --> 00:31:54,170
but I'm not a technical person."

666
00:31:54,590 --> 00:31:55,480
and

667
00:31:56,030 --> 00:31:58,590
many of us may have

668
00:31:58,710 --> 00:32:01,380
difficulty in pointing out places

669
00:32:01,380 --> 00:32:03,380
but one can go and have a look at

670
00:32:03,380 --> 00:32:05,480
such a list of people and say: "hey, look

671
00:32:06,790 --> 00:32:09,390
there's people over there that do also other stuff

672
00:32:10,030 --> 00:32:11,880
it could be indexed by

673
00:32:11,880 --> 00:32:13,450
topic of contribution

674
00:32:13,450 --> 00:32:14,320
as well

675
00:32:18,020 --> 00:32:19,000
Again

676
00:32:20,600 --> 00:32:23,390
uploading, these are contributors that do uploading

677
00:32:23,390 --> 00:32:25,520
these are contributors that do translations

678
00:32:26,460 --> 00:32:28,510
That may be introduced at some point

679
00:32:30,420 --> 00:32:31,940
so that we

680
00:32:31,940 --> 00:32:34,420
we can turn on a spotlight

681
00:32:34,420 --> 00:32:36,690
in several aspects of

682
00:32:36,690 --> 00:32:38,400
Debian that we usually don't look into

683
00:32:38,400 --> 00:32:40,400
we just take it for granted

684
00:32:41,450 --> 00:32:43,730
[Q] I have a lot of questions

685
00:32:43,730 --> 00:32:44,230
Yeah?

686
00:32:44,230 --> 00:32:47,260
I'm more concerned, more than anything

687
00:32:47,260 --> 00:32:50,660
If this...

688
00:32:51,130 --> 00:32:53,140
I mean, some

689
00:32:53,140 --> 00:32:56,140
One thing that you said at the beginning is that

690
00:32:56,140 --> 00:32:58,280
maybe some people would be quite interested

691
00:32:58,280 --> 00:33:00,940
in this for collaborate research, for work

692
00:33:00,940 --> 00:33:02,570
for creating a curriculum

693
00:33:02,570 --> 00:33:03,080
[A] Yes

694
00:33:03,080 --> 00:33:03,940
[Q] and

695
00:33:04,300 --> 00:33:07,980
if it's as easy as to filing a bug which

696
00:33:07,980 --> 00:33:09,890
can be very hard to do

697
00:33:09,890 --> 00:33:12,040
depending on the bug or very easy to do

698
00:33:12,040 --> 00:33:14,040
depending on how much you put in it,

699
00:33:14,040 --> 00:33:16,840
wouldn't be able to

700
00:33:16,840 --> 00:33:19,640
to calling people to put this

701
00:33:19,970 --> 00:33:22,090
random stuff or easy stuff there

702
00:33:22,090 --> 00:33:23,680
for just being in the list

703
00:33:23,680 --> 00:33:24,630
Yes, well

704
00:33:24,630 --> 00:33:27,380
at that point they're really bad job recruiters

705
00:33:28,990 --> 00:33:31,230
Say that if there's a recruiter that just

706
00:33:31,230 --> 00:33:32,710
does random google search

707
00:33:32,710 --> 00:33:34,820
and contacts whoever comes out

708
00:33:34,930 --> 00:33:36,260
like Google recruiters

709
00:33:36,930 --> 00:33:42,270
I keep being contacted for things like managing clusters

710
00:33:42,780 --> 00:33:44,620
If anyone's had a look at what I do

711
00:33:44,620 --> 00:33:47,090
I don't manage clusters, right?

712
00:33:47,260 --> 00:33:50,180
So, if there's recruiters like that

713
00:33:50,180 --> 00:33:51,720
there's nothing you can do

714
00:33:51,970 --> 00:33:55,320
If people care to actually go in and to look

715
00:33:55,320 --> 00:33:57,320
at what a person has done

716
00:33:57,690 --> 00:33:58,620
then

717
00:34:00,100 --> 00:34:02,110
there's a way to find out

718
00:34:02,110 --> 00:34:04,200
what contributions actually were

719
00:34:04,200 --> 00:34:06,460
from just linking to

720
00:34:06,460 --> 00:34:08,260
the usual Debian mailing list

721
00:34:08,260 --> 00:34:10,570
archives or uploads or

722
00:34:10,570 --> 00:34:12,570
commit logs and so on

723
00:34:15,700 --> 00:34:18,880
[Q] So, if I understand correctly, you are ready to

724
00:34:18,880 --> 00:34:20,780
use the BTS as a source

725
00:34:20,780 --> 00:34:23,230
for the contributors, and you could have

726
00:34:23,230 --> 00:34:25,600
a lot of nicknames there

727
00:34:25,750 --> 00:34:28,590
It could be "the Pope" or "Bill Gates", or

728
00:34:28,590 --> 00:34:31,100
the name of a serial killer or whatever

729
00:34:31,100 --> 00:34:33,450
Are you ready to use that?

730
00:34:34,780 --> 00:34:37,490
Maybe there could be some issue with

731
00:34:37,490 --> 00:34:40,090
people that would be falsely

732
00:34:40,090 --> 00:34:41,340
credited

733
00:34:44,760 --> 00:34:46,730
[A] We try and see what happens

734
00:34:48,190 --> 00:34:50,840
It could be that for the bug tracking system

735
00:34:50,840 --> 00:34:52,400
we need to have

736
00:34:52,910 --> 00:34:56,230
some extra intelligence built in to avoid crediting

737
00:34:56,230 --> 00:34:58,720
China manufacturing corporations

738
00:34:59,720 --> 00:35:02,950
or viagra supplies for cheap

739
00:35:04,880 --> 00:35:07,500
mailing list archives are being spammed

740
00:35:07,860 --> 00:35:09,500
somehow

741
00:35:09,500 --> 00:35:11,240
but again

742
00:35:12,890 --> 00:35:14,630
then, maybe

743
00:35:14,630 --> 00:35:17,360
maybe there's lots of noise

744
00:35:17,360 --> 00:35:20,660
and we want to set a threshold that one

745
00:35:21,820 --> 00:35:24,970
contribution to the bug tracking system is not good to be enough

746
00:35:25,520 --> 00:35:29,860
or maybe we can look if that bug was closed

747
00:35:32,180 --> 00:35:34,640
because we close spam bugs

748
00:35:34,640 --> 00:35:36,260
anyway

749
00:35:36,820 --> 00:35:39,230
If there's cruft, if there's noise

750
00:35:39,230 --> 00:35:41,040
that we can filter somehow

751
00:35:41,070 --> 00:35:43,230
maybe we'll loose some contributions

752
00:35:43,730 --> 00:35:46,990
but at least it wasn't for lack of trying

753
00:35:47,710 --> 00:35:48,430
I mean

754
00:35:48,430 --> 00:35:51,300
I don't want to go out of my way to thank someone

755
00:35:51,720 --> 00:35:53,300
If I need to

756
00:35:53,800 --> 00:35:55,320
find out where you live

757
00:35:55,320 --> 00:35:57,050
to come to your home and say thank you

758
00:35:57,060 --> 00:35:59,180
because I don't see you anymore

759
00:36:00,040 --> 00:36:02,940
or because I met you in a crowded

760
00:36:02,940 --> 00:36:04,840
square and you gave me directions

761
00:36:04,840 --> 00:36:07,250
but I didn't take down your phone number

762
00:36:07,250 --> 00:36:09,360
then, it's unfair

763
00:36:09,360 --> 00:36:10,360
you won't be thanked

764
00:36:10,360 --> 00:36:12,810
when I finally reach my destination, but

765
00:36:12,810 --> 00:36:14,810
that's life

766
00:36:16,900 --> 00:36:19,200
So

767
00:36:19,200 --> 00:36:21,010
then maybe it takes

768
00:36:22,730 --> 00:36:25,110
ten e-mails to the BTS

769
00:36:25,110 --> 00:36:27,110
with different content

770
00:36:30,000 --> 00:36:32,440
to be acknowledged

771
00:36:32,940 --> 00:36:36,460
It's possibly something to be

772
00:36:38,160 --> 00:36:42,350
I would leave freedom to the people doing an import

773
00:36:42,600 --> 00:36:46,840
thing, something that gets data out of the bug tracking system

774
00:36:46,840 --> 00:36:50,780
give that area freedom to work it out

775
00:36:52,000 --> 00:36:55,580
and I don't want anyone to be perfect

776
00:36:55,580 --> 00:36:58,080
The system is not going to be perfect

777
00:36:58,330 --> 00:37:00,080
if

778
00:37:00,080 --> 00:37:03,820
somebody would really like to be credited but

779
00:37:03,820 --> 00:37:05,820
isn't in the list

780
00:37:05,820 --> 00:37:09,350
and what they do is report bugs

781
00:37:09,600 --> 00:37:12,900
I would be surprised because if

782
00:37:13,650 --> 00:37:16,510
if somebody wants to be credited on reporting one bug

783
00:37:16,510 --> 00:37:18,730
I'm going to say " Well, just upload some more"

784
00:37:20,200 --> 00:37:22,970
I'm not sure I want to have people

785
00:37:22,970 --> 00:37:24,500
in Debian

786
00:37:24,500 --> 00:37:26,500
that contribute

787
00:37:26,720 --> 00:37:28,720
in order to be credited

788
00:37:31,080 --> 00:37:32,050
If

789
00:37:32,050 --> 00:37:34,050
If contributing some more is

790
00:37:34,050 --> 00:37:36,600
a way to get credited, then

791
00:37:36,760 --> 00:37:38,600
yeah, we don't need to be perfect

792
00:37:38,600 --> 00:37:40,790
just ask people to contribute some more

793
00:37:40,950 --> 00:37:42,760
or if people contribute

794
00:37:42,760 --> 00:37:44,760
the problem is when people contribute a lot

795
00:37:44,760 --> 00:37:46,310
but they're not credited

796
00:37:46,310 --> 00:37:48,750
then maybe we need to figure out how to import

797
00:37:48,750 --> 00:37:51,530
data from another part of Debian that

798
00:37:51,530 --> 00:37:54,140
we currently don't track so well

799
00:37:57,800 --> 00:38:01,020
I'm totally not looking for perfection here

800
00:38:04,040 --> 00:38:07,370
[Q] Have you thought about other ways to use mailing lists

801
00:38:07,370 --> 00:38:09,950
archive because you said

802
00:38:09,950 --> 00:38:12,560
we don't always want to use it as

803
00:38:12,560 --> 00:38:14,560
source of contribution but more

804
00:38:14,560 --> 00:38:16,640
to track the length of contribution

805
00:38:16,640 --> 00:38:19,830
but there are some lists, for example

806
00:38:19,830 --> 00:38:22,770
l10n translation lists

807
00:38:22,770 --> 00:38:27,190
where I guess most people who contribute are real contributors

808
00:38:27,190 --> 00:38:29,190
and sometimes they

809
00:38:29,190 --> 00:38:31,190
also use

810
00:38:31,570 --> 00:38:34,480
some code

811
00:38:34,480 --> 00:38:37,090
I mean, the e-mail could be detecting

812
00:38:37,090 --> 00:38:39,810
like "Request for review"

813
00:38:39,810 --> 00:38:42,500
and stuff like this so you can add some hints on the

814
00:38:42,500 --> 00:38:44,500
type of contribution more precise that

815
00:38:44,500 --> 00:38:46,970
just "sent a mail to the mailing list"

816
00:38:46,970 --> 00:38:49,800
and maybe also for a user mailing list

817
00:38:49,800 --> 00:38:53,550
not tracking the question but only answers

818
00:38:54,100 --> 00:38:56,180
or sort like that...

819
00:38:56,180 --> 00:38:58,180
because, well, asking a question is not a real

820
00:38:58,180 --> 00:39:01,040
contribution to Debian, but helping someone else to

821
00:39:01,040 --> 00:39:03,040
use the Debian system is

822
00:39:03,040 --> 00:39:05,040
a Debian contribution

823
00:39:05,760 --> 00:39:10,000
and one of the most important one to start with I guess

824
00:39:12,670 --> 00:39:16,530
[A] Decoding mailing lists patterns

825
00:39:16,530 --> 00:39:18,530
fairly is not

826
00:39:18,530 --> 00:39:20,530
easy as far as I understand

827
00:39:20,530 --> 00:39:22,960
but I guess it changes from list to list

828
00:39:23,460 --> 00:39:24,530
but

829
00:39:24,740 --> 00:39:27,010
that could be solved by

830
00:39:27,760 --> 00:39:30,070
adding a general mailing list

831
00:39:30,070 --> 00:39:32,070
source that only looks at time frame

832
00:39:32,290 --> 00:39:35,340
and having the translation team saying

833
00:39:35,340 --> 00:39:37,340
"In our mailing list

834
00:39:37,340 --> 00:39:39,340
there's more semantic information that can be

835
00:39:39,340 --> 00:39:40,310
extracted"

836
00:39:40,310 --> 00:39:42,310
so make

837
00:39:42,310 --> 00:39:44,310
an extra import if

838
00:39:44,310 --> 00:39:46,990
importer for my team that decodes

839
00:39:47,360 --> 00:39:50,240
things and put there properly

840
00:39:50,240 --> 00:39:52,240
as long as you can have

841
00:39:52,240 --> 00:39:55,740
many things dumping data into a bucket

842
00:39:55,740 --> 00:39:57,740
then that works

843
00:39:58,120 --> 00:40:01,840
I currently have no idea

844
00:40:01,840 --> 00:40:04,090
well, only vague ideas on

845
00:40:04,090 --> 00:40:05,060
how...

846
00:40:05,670 --> 00:40:08,330
what's the protocol to collect this data

847
00:40:11,140 --> 00:40:13,500
I would

848
00:40:13,860 --> 00:40:17,460
from the central bit

849
00:40:17,460 --> 00:40:20,210
I wouldn't go much further than

850
00:40:20,210 --> 00:40:24,680
having a list of urls from where I download something

851
00:40:25,590 --> 00:40:27,570
on a regular basis

852
00:40:27,570 --> 00:40:29,570
because that's the simplest

853
00:40:30,510 --> 00:40:33,340
protocol to identify instead of

854
00:40:33,340 --> 00:40:36,580
submits and whatnot

855
00:40:37,170 --> 00:40:41,410
and what's in the file that gets downloaded

856
00:40:44,130 --> 00:40:47,600
could be identity, initial time and final time

857
00:40:50,880 --> 00:40:53,400
and then each datasource

858
00:40:53,400 --> 00:40:56,010
makes a file available somewhere

859
00:40:56,450 --> 00:40:58,840
I was thinking of something

860
00:40:58,840 --> 00:41:01,730
super simple minded like that but then it's not enough

861
00:41:01,730 --> 00:41:03,970
because then if you want to look

862
00:41:03,970 --> 00:41:06,140
to make a link

863
00:41:06,140 --> 00:41:09,000
showing the contributions then

864
00:41:09,690 --> 00:41:11,490
well I need something

865
00:41:12,210 --> 00:41:15,490
about how to

866
00:41:17,270 --> 00:41:22,680
how to make this data available for collection

867
00:41:23,340 --> 00:41:26,450
at the central point I'm only

868
00:41:26,450 --> 00:41:28,450
I want it as simple minded

869
00:41:28,450 --> 00:41:30,860
as possible, I don't want

870
00:41:32,060 --> 00:41:35,030
the core of this to be about data collection

871
00:41:37,330 --> 00:41:39,020
about mining

872
00:41:39,020 --> 00:41:43,050
it should just be about collecting what as been mied

873
00:41:46,930 --> 00:41:48,760
One last question?

874
00:41:55,090 --> 00:41:57,500
[Q] Just a thought on what you said there is a

875
00:41:57,500 --> 00:41:59,500
Google Summer of Code project

876
00:41:59,500 --> 00:42:02,280
on Fedmsg

877
00:42:02,280 --> 00:42:04,830
which is a ??? maybe to

878
00:42:04,830 --> 00:42:06,830
have a bus

879
00:42:06,830 --> 00:42:08,830
collecting lots of information from

880
00:42:08,830 --> 00:42:10,830
various sources of information

881
00:42:10,830 --> 00:42:12,960
among those, there's already

882
00:42:12,960 --> 00:42:14,960
mentors.debian.net

883
00:42:14,960 --> 00:42:17,210
and I guess the BTS would be

884
00:42:17,210 --> 00:42:19,620
[A] main source? [Q] it's a source

885
00:42:19,620 --> 00:42:21,200
???

886
00:42:21,200 --> 00:42:26,640
yeah but I guess it could have enough metadata

887
00:42:26,640 --> 00:42:30,220
like ???

888
00:42:30,220 --> 00:42:34,190
the software is fedmsg

889
00:42:34,190 --> 00:42:36,910
F-E-D-M-S-G

890
00:42:36,910 --> 00:42:39,660
it's from Fedora and

891
00:42:39,660 --> 00:42:43,350
the summer of code student is trying to adapt it for Debian

892
00:42:43,350 --> 00:42:46,010
just a software bus

893
00:42:46,010 --> 00:42:50,510
where you get lots of events from various sources

894
00:42:51,620 --> 00:42:54,420
maybe it could be

895
00:42:54,420 --> 00:42:58,610
assuming it goes further than this summer

896
00:42:58,610 --> 00:43:03,660
it could be an interesting use of this project

897
00:43:04,440 --> 00:43:05,490
Ok

898
00:43:05,490 --> 00:43:08,710
[Chairman] As the mentor of this project, I agree

899
00:43:08,710 --> 00:43:11,260
[laughter]

900
00:43:11,260 --> 00:43:13,650
Ah, let's talk

901
00:43:21,530 --> 00:43:25,660
and I think the student is arriving today, so

902
00:43:29,080 --> 00:43:35,130
[laughter]

903
00:43:37,430 --> 00:43:39,840
So, we're out of time

904
00:43:39,840 --> 00:43:41,230
so, thank you Enrico

905
00:43:41,230 --> 00:43:46,870
[Enrico] We can talk about this, I plan to work on it during the DebConf

906
00:43:47,810 --> 00:43:53,190
shortly, that should be Martin Ferrari coming who was also

907
00:43:53,190 --> 00:43:57,550
intending to work on a site, similar to this

908
00:43:57,550 --> 00:43:59,770
and I get we should all get together

909
00:43:59,770 --> 00:44:01,430
I can think of something

910
00:44:01,430 --> 00:44:04,650
and feel free to stop me to talk about this and

911
00:44:04,650 --> 00:44:06,650
offer help or

912
00:44:07,710 --> 00:44:11,510
that's the main thing I want to do a DebConf

913
00:44:11,510 --> 00:44:13,510
and I want to get something land before

914
00:44:13,510 --> 00:44:15,510
the end of DebConf so we can

915
00:44:15,510 --> 00:44:17,510
present it properly

916
00:44:18,110 --> 00:44:29,630
[applause]