File:Recording of discussion - WMF Annual Plan 2023-34 - regarding Commons workflows.webm

1 00:00:02.610 --> 00:00:08.620 Liam Wyatt: Okay, that is our recording. Hello and welcome. This is the

2 00:00:08.640 --> 00:00:12.960 Liam Wyatt: a conversation around the specific topic of the

3 00:00:13.010 --> 00:00:24.830 Liam Wyatt: Wikipedia Foundations Annual planning. Okay, ours, which is the objectives and key results, which is the the a term for the goals, for the gear.

4 00:00:25.540 --> 00:00:33.080 Liam Wyatt: for particularly for the community who work on Commons we can need a commons. and those who do high

5 00:00:33.160 --> 00:00:45.180 Liam Wyatt: or advanced editing workflows have advanced ending user rights or complex procedures because some of the okay, as some of the key results in the annual plan

6 00:00:45.380 --> 00:00:50.760 Liam Wyatt: for the product and technology department. the late specifically to

7 00:00:51.020 --> 00:00:56.730 Liam Wyatt: at how to improve or how to support the workflows of those people who have

8 00:00:57.240 --> 00:00:59.720 Liam Wyatt: advanced procedures they need to

9 00:00:59.830 --> 00:01:09.270 Liam Wyatt: to work through in content, creation, or content. Moderation we use the word moderation loosely. It does not necessarily mean administrator actions.

10 00:01:09.380 --> 00:01:13.840 Liam Wyatt: but it means the curation of content. And in some cases of people

11 00:01:13.880 --> 00:01:24.990 Liam Wyatt: on the sites that word is used. Moderation is used in the annual client for this broad meeting. My name is Liam Wyatt, representing the Wik Media Foundation. This

12 00:01:25.060 --> 00:01:43.770 Liam Wyatt: session is being recorded. and I will try and put the a machine generated, transcript up with it. As a result, we are also in compliance with the Wikimedia Foundation Friendly space policies which I have linked from the notes document that we are taking. I will share the notes document

13 00:01:44.060 --> 00:01:45.990 Liam Wyatt: and let it mix

14 00:01:46.180 --> 00:01:52.760 Liam Wyatt: in the minutes of the chat. When I publish this video on the Comedian Commons itself.

15 00:01:54.010 --> 00:01:59.000 Liam Wyatt: Thank you. Everyone for coming. I have linked into the notes document.

16 00:01:59.230 --> 00:02:03.350 Liam Wyatt: and I will link it here now into our video call

17 00:02:04.540 --> 00:02:16.460 Liam Wyatt: the okay as the objectives and key results. And I would like to draw your attention in particular to the one called Wiki experiences. One

18 00:02:16.800 --> 00:02:18.360 Liam Wyatt: point, 2

19 00:02:19.320 --> 00:02:22.210 Liam Wyatt: which refers to

20 00:02:22.730 --> 00:02:31.510 Liam Wyatt: the goal of complete improvements to 4 workflows that improve the experience of editors with extended rights

21 00:02:32.060 --> 00:02:36.120 Liam Wyatt: to extend their creativity. impacting at least 4 wikis

22 00:02:36.510 --> 00:02:40.880 Liam Wyatt: and improve collaboratively, could it collaboratively with the Volunteers?

23 00:02:41.270 --> 00:02:51.950 Liam Wyatt: This has not been specified yet which wikis many, which is better, it's better. and it has not been specified yet

24 00:02:52.120 --> 00:02:56.370 Liam Wyatt: which workflows. This is not a race. But

25 00:02:56.540 --> 00:03:03.110 Liam Wyatt: if there's a really obvious workflow that can help a lot of people do a thing that is high impact.

26 00:03:03.320 --> 00:03:10.220 Liam Wyatt: That is something that the staff of the Wikimedia foundation product and technology departments would really like to know about.

27 00:03:12.060 --> 00:03:19.560 Liam Wyatt: If you have come to this call it hopefully means you have some ideas in this direction. so

28 00:03:21.080 --> 00:03:29.020 Liam Wyatt: I would like to open the floor to anyone who might have particular suggestions or particular concerns that they have

29 00:03:29.070 --> 00:03:31.460 Liam Wyatt: coming into this call straight from the back.

30 00:03:32.840 --> 00:03:36.500 Liam Wyatt: Did anyone have anything they wanted to say from the very stop.

31 00:03:43.150 --> 00:03:44.290 Liam Wyatt: Susannah, please.

32 00:03:44.890 --> 00:03:45.600 Susanna Ånäs: Hmm.

33 00:03:46.420 --> 00:03:53.080 Susanna Ånäs: I come here as rather unprepared, but and also not.

34 00:03:53.240 --> 00:04:10.580 Susanna Ånäs: No, no, knowing thoroughly the the annual plan proposals. But I would like to emphasize that the institutional metadata input flows, and and at the the ease and the the like, the the

35 00:04:11.480 --> 00:04:20.829 Susanna Ånäs: the how how well they are preserved in the process, and how how which tools they go through, and and how they are then facilitated for further

36 00:04:20.850 --> 00:04:22.590 Susanna Ånäs: community curation.

37 00:04:23.490 --> 00:04:33.830 Susanna Ånäs: Could you give an example? I think I know what you mean coming from a glam background when you say institutional metadata and preserving the that we need to have a glam

38 00:04:33.840 --> 00:04:35.010 Susanna Ånäs: strategy.

39 00:04:35.140 --> 00:04:53.950 Susanna Ånäs: and then in that they are very often batch uploads, or it could be continual uploads or continual maintenance processes. They could be different. They could be a individual uploads through the the wizard or or any different kinds. I I personally

40 00:04:53.950 --> 00:05:11.920 Susanna Ånäs: have created this Wiki Documentaries project, for which we have a like applied for I don't know how it's going at a Google summer of code developer doing a an Sdc. Upload process minimum viable process from from from an institution to Stc.

41 00:05:12.050 --> 00:05:18.330 And I think that it's a little, quite a broken workflow, with a different

42 00:05:18.840 --> 00:05:23.060 Susanna Ånäs: well different obstacles. So I think they should be like a

43 00:05:23.450 --> 00:05:30.250 Susanna Ånäs: a a clear vision of how to facilitate that in order to better connect and engage

44 00:05:30.270 --> 00:05:32.690 Susanna Ånäs: the memory institutions.

45 00:05:32.970 --> 00:05:37.670 Susanna Ånäs: who would wish their collections to be

46 00:05:37.820 --> 00:05:40.890 Susanna Ånäs: let me federated through Wikimedia.

47 00:05:42.480 --> 00:05:46.080 Liam Wyatt: And you said so. That example.

48 00:05:46.240 --> 00:05:51.040 Liam Wyatt: speaks a lot about the the upload workflow, and particularly in batch

49 00:05:51.070 --> 00:05:52.310 Liam Wyatt: processing, which

50 00:05:52.600 --> 00:06:03.860 Liam Wyatt: you know I I've tried to work on that area as well, and and had some successes and some failures in in support for those kind of workflows. But you also said, in your introductory question or statement

51 00:06:03.990 --> 00:06:07.040 Liam Wyatt: about metadata, preservation.

52 00:06:07.080 --> 00:06:12.580 Liam Wyatt: institutional. And and yeah, I assume about you implied about round tripping.

53 00:06:13.050 --> 00:06:23.820 Susanna Ånäs: Well, my my personal, I have very much interest in a problem. And so when we get materials from

54 00:06:23.820 --> 00:06:42.460 Susanna Ånäs: different institutions, probably the same, maybe recognized through different intelligent algorithms that they are the same images. But the the metadata is different. So how do we? How do we go on about that? And

55 00:06:42.460 --> 00:06:44.210 Susanna Ånäs: so I I.

56 00:06:44.290 --> 00:06:52.970 Susanna Ånäs: If one uploads through the upload Wizard, there is a very, very like a thin

57 00:06:53.090 --> 00:06:54.260 Susanna Ånäs: Hmm.

58 00:06:54.500 --> 00:07:07.050 Susanna Ånäs: I can't remember proper word, but th it in a passage through which this all this data is squeezed, and none you can't really say that it would have been that the metadata would have been preserved.

59 00:07:07.120 --> 00:07:12.040 Susanna Ånäs: Now I kind of lost your question. So

60 00:07:12.450 --> 00:07:18.720 Liam Wyatt: what? Yeah, that's not I You did. You did. You did answer it.

61 00:07:19.180 --> 00:07:26.000 Liam Wyatt: but you extended it to another related suggestion, and I want to know if there's anyone else in this call who has

62 00:07:26.220 --> 00:07:27.500 Liam Wyatt: similar

63 00:07:27.510 --> 00:07:36.750 Liam Wyatt: request or similar concern, because this is not about Glam specifically one about. No, no; your example about

64 00:07:37.240 --> 00:07:40.790 Liam Wyatt: identifying matching images that already exist.

65 00:07:41.740 --> 00:07:49.800 Liam Wyatt: and and merging them, though those kinds of tools, whether for copyright or shel, merely for non duplicate work.

66 00:07:50.000 --> 00:07:56.500 Liam Wyatt: Was that something that is come up elsewhere? Perhaps a/C. And you had your hand up. Is that

67 00:07:56.650 --> 00:08:03.130 AntiCompositeNumber (he/him): so? Yeah, I I have some experience doing this both with my

68 00:08:03.210 --> 00:08:10.320 AntiCompositeNumber (he/him): license reviewing BoT for files for my naturalist. as well as generally finding duplicates.

69 00:08:11.770 --> 00:08:22.270 AntiCompositeNumber (he/him): My: from my perspective. The biggest problem that can be solved with technology is when our upload tools do not allow

70 00:08:22.970 --> 00:08:26.570 AntiCompositeNumber (he/him): users to specify enough information.

71 00:08:26.790 --> 00:08:33.740 So whether it's the Crosswiki upload tool that is only designed for use with own work, but only is used to upload, copy vio.

72 00:08:35.890 --> 00:08:49.100 AntiCompositeNumber (he/him): or it's the upload wizard not supporting derivative works well. or it's people not knowing how to specify source as well, and get things tagged for license, Review.

73 00:08:49.400 --> 00:08:53.060 AntiCompositeNumber (he/him): Those are all things that have to be handled manually and are

74 00:08:53.930 --> 00:08:55.880 AntiCompositeNumber (he/him): really caused by the tooling.

75 00:08:56.490 --> 00:08:57.450 Liam Wyatt: Okay.

76 00:08:59.140 --> 00:09:00.100 Liam Wyatt: could you?

77 00:09:00.620 --> 00:09:09.260 Liam Wyatt: Could you like this? Is this: this is the upload wizard that exists? That's the main method for people to upload who have not got special tools and so forth.

78 00:09:10.190 --> 00:09:12.210 Liam Wyatt: You're saying that the

79 00:09:13.060 --> 00:09:17.360 the edge cases are sufficiently large, but they are no longer edge.

80 00:09:17.920 --> 00:09:29.410 AntiCompositeNumber (he/him): So there there are 2 common ways for normal people to upload files. One of them is special upload wizard on commons, which is a well-designed tool

81 00:09:29.610 --> 00:09:37.640 AntiCompositeNumber (he/him): that tries to handle most basic use cases. It doesn't handle all basic use cases but it tries to handle most pretty well.

82 00:09:38.770 --> 00:09:42.010 AntiCompositeNumber (he/him): Then. alternately. There's the

83 00:09:42.070 --> 00:09:48.030 AntiCompositeNumber (he/him): upload button in visual editor and in the source editor on other Wikis. That tool is

84 00:09:48.050 --> 00:09:51.500 AntiCompositeNumber (he/him): designed to only work for own work.

85 00:09:51.670 --> 00:09:55.650 AntiCompositeNumber (he/him): However, it's used to upload all sorts of various

86 00:09:55.990 --> 00:10:04.730 AntiCompositeNumber (he/him): works, whether it's public domain works that are clearly public domain, but because it only works for own work. We do not have a source, and then they get deleted.

87 00:10:05.690 --> 00:10:14.190 AntiCompositeNumber (he/him): and all of that takes man over view of someone has to go to try and find a source, and then they can't find a source, and then someone else has to review it and eventually delete it

88 00:10:15.800 --> 00:10:25.000 AntiCompositeNumber (he/him): that that's probably the biggest problem. But then recurring problems are. People try to upload derivative works through the upload Wizard

89 00:10:26.440 --> 00:10:30.270 AntiCompositeNumber (he/him): and just don't know how to specify the

90 00:10:30.570 --> 00:10:33.240 AntiCompositeNumber (he/him): copyright metadata successfully

91 00:10:33.880 --> 00:10:36.330 AntiCompositeNumber (he/him): before we even get to things like

92 00:10:38.790 --> 00:10:40.880 AntiCompositeNumber (he/him): original sources and things

93 00:10:43.220 --> 00:10:51.220 Liam Wyatt: right. So i'm. I'm trying to take notes as we go. As to make sure i'm capturing people's points your your point with the upload wizard

94 00:10:51.520 --> 00:10:56.300 Liam Wyatt: on Commons is that it? It misses the ability to express a lot of

95 00:10:56.500 --> 00:10:57.660 Liam Wyatt: metadata

96 00:10:58.070 --> 00:11:03.400 Liam Wyatt: variants, notably intuitive work, and as a result, things get reviewed, which

97 00:11:03.610 --> 00:11:12.500 Liam Wyatt: creates workflow which needn't exist because that information could have been added. If it was, it was easy to do so. In the first place.

98 00:11:12.990 --> 00:11:13.890 AntiCompositeNumber (he/him): right.

99 00:11:14.340 --> 00:11:18.340 Liam Wyatt: I find that I get that a lot with people leading to

100 00:11:18.960 --> 00:11:29.870 Liam Wyatt: submit tickets afterwards saying, I give permission for this thing, but they have to do that only after they've uploaded the file, which means to to get the URL saying this thing here.

101 00:11:30.090 --> 00:11:35.870 Liam Wyatt: But that means they're under the clock to to make sure they send that email in before they

102 00:11:36.080 --> 00:11:41.250 Liam Wyatt: before the file gets delivered, because it didn't have matching permission email. So it's a bit of a catch 22 there.

103 00:11:42.750 --> 00:11:52.220 Liam Wyatt: but your your Your other point was about the upload button on other wiki's being underpowered for the variety of uses that it's put towards.

104 00:11:53.470 --> 00:12:05.000 AntiCompositeNumber (he/him): Yes, the there has been a persistent request ever since that existed from the Commons community to just shut the thing off, or place it with a link to special upload wizard, and doing that alone

105 00:12:05.080 --> 00:12:07.240 AntiCompositeNumber (he/him): would like

106 00:12:07.530 --> 00:12:14.470 AntiCompositeNumber (he/him): significantly reduce the amount of workload that it takes to review those files.

107 00:12:15.750 --> 00:12:18.300 Liam Wyatt: I think that would definitely fall under the

108 00:12:18.940 --> 00:12:29.220 Liam Wyatt: her view of each Wikipedia, or which Wiki's local decision making, I doubt that we can do foundation would want to make turning off local upload.

109 00:12:29.470 --> 00:12:40.360 AntiCompositeNumber (he/him): not local upload cross wiki upload. That is an important distinction. We don't care what they do with their local files.

110 00:12:40.500 --> 00:12:46.040 AntiCompositeNumber (he/him): We care when there's a button on another wiki that says, create a problem on commons.

111 00:12:46.100 --> 00:12:50.450 AntiCompositeNumber (he/him): and people click it all the time every day and create problems and comments.

112 00:12:51.020 --> 00:12:56.140 DerHexer/Martin: Be a mean and visual additive find out in through which I adjust, and such

113 00:12:57.800 --> 00:12:58.600 AntiCompositeNumber (he/him): Yes.

114 00:13:02.170 --> 00:13:06.790 Liam Wyatt: Sam. or perhaps look up. I'm i'm.

115 00:13:06.810 --> 00:13:19.130 Liam Wyatt: Calling upon you as, or Benedict as the Wikimedia Foundation staff in the room here has this question of crosswiki upload being on the radar with

116 00:13:19.180 --> 00:13:20.600 Liam Wyatt: work before.

117 00:13:26.360 --> 00:13:34.120 Sam Walton: From From my perspective I I can't recall us talking about it, though I have personally not been in the loop on Commons conversations very much recently.

118 00:13:38.230 --> 00:13:42.540 Benedict Udeh: Yeah, Same with me. I the camera call that's coming up.

119 00:13:42.620 --> 00:13:43.490 Benedict Udeh: Also.

120 00:13:44.840 --> 00:13:47.300 Liam Wyatt: it's an interesting one, because it's the first one that

121 00:13:47.360 --> 00:13:52.140 Liam Wyatt: that I've come across. In this conversation about improving workforce the Commons. That is

122 00:13:52.230 --> 00:14:07.830 Liam Wyatt: not a uniquely common topic. It's a multimedia topic that affects across 2 wikis which might result in the workflow falling, the workflow problem falling between the cracks of different people's responsibilities

123 00:14:07.980 --> 00:14:10.620 Liam Wyatt: and and purview. Not

124 00:14:13.550 --> 00:14:22.130 DerHexer/Martin: Yeah. I a bit uncertain if. although it might fix the Commons process.

125 00:14:23.180 --> 00:14:24.980 DerHexer/Martin: whether it's really

126 00:14:26.960 --> 00:14:29.440 DerHexer/Martin: looked at from the right perspective.

127 00:14:29.460 --> 00:14:34.220 DerHexer/Martin: So it flow for people who want to

128 00:14:34.360 --> 00:14:36.300 DerHexer/Martin: upload files.

129 00:14:37.110 --> 00:14:46.620 DerHexer/Martin: They usually don't start on commons with uploading files, but Wikipedia and other Wikipedia projects, so it

130 00:14:46.630 --> 00:14:48.480 DerHexer/Martin: intentionally

131 00:14:49.920 --> 00:14:57.150 DerHexer/Martin: makes most sense to me if piles will be uploaded. Maybe it's pressed out on the local wikis.

132 00:14:57.270 --> 00:14:58.940 DerHexer/Martin: but then

133 00:14:59.430 --> 00:15:06.420 DerHexer/Martin: going to, we can get comments, and what I understood from a cn is that

134 00:15:06.690 --> 00:15:12.130 DerHexer/Martin: the the current upload form on the local Wikis

135 00:15:12.230 --> 00:15:19.730 DerHexer/Martin: Us. This is special, local, very variation. Instead of the upload visit which exists and comes

136 00:15:19.820 --> 00:15:21.840 DerHexer/Martin: so instead of

137 00:15:22.220 --> 00:15:28.610 DerHexer/Martin: moving everyone from the local Vicki to come and to upload the special uplet with it.

138 00:15:30.540 --> 00:15:37.790 DerHexer/Martin: I think it makes more sense to move, upload with it 2 commons to the local wikis

139 00:15:38.590 --> 00:15:45.570 DerHexer/Martin: which you disagree on. This is, and what what? What was your suggestion? I'm, facing the

140 00:15:45.620 --> 00:15:56.040 AntiCompositeNumber (he/him): so from my perspective that there are 2 problems. One is people click the upload button. They upload things. They don't know that they ended up on commons.

141 00:15:56.390 --> 00:15:59.130 AntiCompositeNumber (he/him): We've gotten over time. There's

142 00:15:59.320 --> 00:16:08.160 AntiCompositeNumber (he/him): then lots of people requesting help saying I uploaded something. I have no idea where it went. I checked my contributions. It's nowhere.

143 00:16:08.300 --> 00:16:13.080 AntiCompositeNumber (he/him): but it's still in the article. I just can't find where it went. That is one problem.

144 00:16:13.490 --> 00:16:20.280 AntiCompositeNumber (he/him): The other problem is the form that let them do that causes problems on Commons as well.

145 00:16:20.350 --> 00:16:32.120 AntiCompositeNumber (he/him): so embedding the upload wizard into visual editor and source editor would solve the Commons problem of people are uploading things without sufficient metadata.

146 00:16:32.350 --> 00:16:38.300 AntiCompositeNumber (he/him): it wouldn't necessarily solve the I don't know where my uploads just went. Problem.

147 00:16:39.920 --> 00:16:44.210 AntiCompositeNumber (he/him): That is much more of a problem, especially when we talk about

148 00:16:44.420 --> 00:16:45.860 AntiCompositeNumber (he/him): things that have

149 00:16:47.220 --> 00:16:59.660 AntiCompositeNumber (he/him): maybe not clear permission states, and we say we send a talk page message saying, hey. can you say where you got this from, or why you think it's freely licensed.

150 00:16:59.810 --> 00:17:04.930 AntiCompositeNumber (he/him): and then, because it's on another wiki, it just goes up into the bell, and they don't pay attention to it.

151 00:17:06.640 --> 00:17:09.640 Liam Wyatt: Yeah. the I

152 00:17:09.869 --> 00:17:16.569 Liam Wyatt: I I acknowledge I feel the the the problem having done, and I think many people in this room done

153 00:17:16.589 --> 00:17:19.369 Liam Wyatt: outreach events or training events where

154 00:17:19.700 --> 00:17:29.860 Liam Wyatt: we're helping someone write an article and they want to add an image. the workflow to say, okay, we'll open a new tab and go to a completely different website.

155 00:17:29.910 --> 00:17:31.800 Liam Wyatt: upload it over there.

156 00:17:31.930 --> 00:17:51.510 Liam Wyatt: and then get that file name, and come back to your first term. Now insert that file is contradictory or confusing the I believe the usability side of work implies or increase the usability of Wikipedia. Editing implies that people should be able to upload their files

157 00:17:51.510 --> 00:17:56.030 Liam Wyatt: as they edit. and there's particularly mobile and things like that.

158 00:17:56.440 --> 00:18:01.260 Liam Wyatt: but that causes the kinds of confusions that you're describing

159 00:18:04.030 --> 00:18:04.830 Liam Wyatt: mountain.

160 00:18:07.640 --> 00:18:13.550 DerHexer/Martin: But this might be an issue on many by size that you

161 00:18:13.850 --> 00:18:21.110 DerHexer/Martin: do something on 1 one place and something in the background happens. So

162 00:18:21.370 --> 00:18:22.390 DerHexer/Martin: maybe he's.

163 00:18:23.970 --> 00:18:26.640 DerHexer/Martin: The initial problem is.

164 00:18:26.890 --> 00:18:39.510 DerHexer/Martin: we: we. We are guiding people to their local contributions, but they don't find their edits, including their uploads and commons there. So if they had one like global contributions, Page

165 00:18:39.530 --> 00:18:50.500 DerHexer/Martin: I shown on the the Wiki is both filtering, and such they might find more easily the files they were uploading. So this like

166 00:18:55.950 --> 00:19:01.590 DerHexer/Martin: is good by itself, but not for those who work on the global scale.

167 00:19:01.800 --> 00:19:04.830 DerHexer/Martin: I like the ones who work at Commons.

168 00:19:06.600 --> 00:19:07.310 Liam Wyatt: Yeah.

169 00:19:07.670 --> 00:19:08.420 DerHexer/Martin: it

170 00:19:09.230 --> 00:19:12.760 Liam Wyatt: definitely the the I want to point out here.

171 00:19:12.950 --> 00:19:24.720 Liam Wyatt: as also mentioned by in the in the chat here. This is not easy or unique to us. This is these kinds of user interface. Where did my thing go? Is not a unique to Wik Media.

172 00:19:24.970 --> 00:19:40.060 Liam Wyatt: I I should add in this this point that one of the big shifts in the annual plan priorities for the Wikimedia foundation technical work this year, compared to a lot of the conversations over the last, say 5 years.

173 00:19:40.070 --> 00:19:42.440 which have been heavily focused on

174 00:19:42.580 --> 00:19:44.290 Liam Wyatt: new user

175 00:19:44.390 --> 00:19:55.710 Liam Wyatt: acquisition and retention, lowering the barrier to entry, which is still a good thing, and still desirable, of course. But this year there is a deliberate focus in the

176 00:19:55.740 --> 00:20:08.900 Liam Wyatt: in the text of the annual plan and its priorities to make the the workflows and the workload of existing users, particularly power users, and I use the word power users

177 00:20:09.160 --> 00:20:11.870 Liam Wyatt: broadly and without prejudice.

178 00:20:13.110 --> 00:20:15.090 Liam Wyatt: to make their work easier

179 00:20:15.210 --> 00:20:24.060 Liam Wyatt: on the assumption that those people there's a pareto principle that a small number of people do a large proportion of the work. We all know this to be true.

180 00:20:24.880 --> 00:20:36.260 Liam Wyatt: and that, lowering their work, making their work easier, increases the quality of the the enjoyment for a lot of people, and because there's less of that

181 00:20:36.780 --> 00:20:41.770 Liam Wyatt: drudgery to get through these these backlogs that we've all seen. So

182 00:20:43.080 --> 00:20:51.630 Liam Wyatt: was their id user new user improvements that can be made. I would. I would like to

183 00:20:51.740 --> 00:21:08.720 Liam Wyatt: emphasize things that the advanced users, the power, uses. The moderators here would find beneficial to you personally to lower your workloads or simplify your workloads and and work flows.

184 00:21:09.220 --> 00:21:11.900 Liam Wyatt: as that might actually be the kind of

185 00:21:12.580 --> 00:21:18.630 Liam Wyatt: precision target that we can work on in this annual plan to benefit a lot of people.

186 00:21:19.080 --> 00:21:21.240 Liam Wyatt: IC and and and Martin.

187 00:21:22.350 --> 00:21:25.940 AntiCompositeNumber (he/him): So a lot of the problem is that

188 00:21:25.960 --> 00:21:30.910 AntiCompositeNumber (he/him): a lot of Commons workflows, especially for commons. Administrators are

189 00:21:31.630 --> 00:21:36.740 AntiCompositeNumber (he/him): requiring of human judgment and significant amounts of research.

190 00:21:37.190 --> 00:21:48.220 AntiCompositeNumber (he/him): If i'm going to close the deletion request. I need to satisfy myself that something is or is not, a copyright violation. which means I need to research

191 00:21:48.420 --> 00:21:57.030 AntiCompositeNumber (he/him): both the love, whatever country applies here, if that's relevant, and the sourcing circumstances of this particular image

192 00:21:57.230 --> 00:22:05.370 AntiCompositeNumber (he/him): that takes time, and that takes human judgment. That's not something that can be easily automated. We've automated all of the automatable parts like

193 00:22:06.070 --> 00:22:10.820 AntiCompositeNumber (he/him): actually deleting the files and sending messages to users and

194 00:22:11.460 --> 00:22:14.310 AntiCompositeNumber (he/him): closing the discussion itself.

195 00:22:14.490 --> 00:22:19.320 AntiCompositeNumber (he/him): That part is all automated. It's the human judgment that we haven't automated yet, because we can't.

196 00:22:20.980 --> 00:22:24.180 AntiCompositeNumber (he/him): which is why a lot of my focus is on.

197 00:22:25.160 --> 00:22:36.690 AntiCompositeNumber (he/him): If we remove. If we lower the amount of work that needs to be done. In the first place, that improves the rest of the workflow, because we can actually focus on the things that need our attention.

198 00:22:36.990 --> 00:22:39.450 AntiCompositeNumber (he/him): not the things that shouldn't need our attention.

199 00:22:42.070 --> 00:22:51.080 Liam Wyatt: Can I ask Shawn to respond to that one as someone from the done Wicked Foundation legal department sean? Can you hear us?

200 00:22:52.270 --> 00:22:53.290 Shaun Spalding: Yeah

201 00:22:58.410 --> 00:23:13.610 Shaun Spalding: trying to unmute. I I think that there was. There are 2 hypotheses that I have, and what you just said dovetailed with one of them. It's like. Either you make things easier to

202 00:23:13.820 --> 00:23:19.780 Shaun Spalding: decide when deleting certain content, or you

203 00:23:20.500 --> 00:23:23.960 Shaun Spalding: shut off some of the situations where

204 00:23:24.080 --> 00:23:39.670 Shaun Spalding: users can upload obviously problematic things that shouldn't be using user time. And so I think that's really strong piece of feedback which is let's educate users better to not cause problems at the start of the

205 00:23:40.810 --> 00:23:46.180 Shaun Spalding: pipeline. because it seems as if many of the things that

206 00:23:47.230 --> 00:23:52.920 Shaun Spalding: can be automated are already automated at the tail end of the pipeline.

207 00:23:52.970 --> 00:23:58.150 Shaun Spalding: And so I think that's really useful information. I don't know if anyone has any

208 00:23:58.160 --> 00:24:11.320 Shaun Spalding: additional feedback. As to other places, that problematic uploads are coming from, or ways that uploads are particularly problematic that

209 00:24:11.420 --> 00:24:19.490 Shaun Spalding: you see. But don't need to be. That can be changed through education or change through different

210 00:24:19.610 --> 00:24:22.190 Shaun Spalding: user sort of

211 00:24:22.370 --> 00:24:25.250 Shaun Spalding: prompts in the upward wizard.

212 00:24:25.520 --> 00:24:30.190 Shaun Spalding: etc. So if anyone has any feedback on that, that'd be useful.

213 00:24:31.670 --> 00:24:37.390 Liam Wyatt: Susannah Martin, I see your hand up already, but I think she's gonna have a a direct

214 00:24:38.370 --> 00:24:42.100 Susanna Ånäs: Yes, and i'm happy about the question, because I want to

215 00:24:42.330 --> 00:24:52.490 Susanna Ånäs: push this this question into a discussion in the Wikileaks Living Heritage project. We are working with a group of

216 00:24:52.530 --> 00:25:03.810 Susanna Ånäs: intangible cultural heritage professionals and the Wikimedia community together, in order to find ways to

217 00:25:05.030 --> 00:25:11.710 Susanna Ånäs: upload materials that probably public domain, for example, but they may

218 00:25:11.790 --> 00:25:28.820 Susanna Ånäs: be ethically questionable. But if because this this has been consulted with the with the subjects or the originating communities, this is then expressed in the upload process, and then they will be forming.

219 00:25:28.820 --> 00:25:44.340 Susanna Ånäs: let's say, an ethically checked a collection. So we are doing a a special campaign for that with extra questions, and it's a like a a, a, a a learning process that we carry up this year.

220 00:25:48.450 --> 00:25:52.570 Susanna Ånäs: and I don't know how to, you know, promote it for further

221 00:25:52.800 --> 00:25:59.210 Susanna Ånäs: the use yet, but I would very much invite anyone who is interested to to join.

222 00:26:02.610 --> 00:26:03.730 Liam Wyatt: Thank you, Susanna.

223 00:26:04.140 --> 00:26:09.070 Liam Wyatt: There are 2 other hands up. I wanted to make sure before we get to them is

224 00:26:11.490 --> 00:26:13.700 Liam Wyatt: so i'm not sure what you're at.

225 00:26:13.710 --> 00:26:16.920 Liam Wyatt: Z. Then

226 00:26:19.150 --> 00:26:21.660 Liam Wyatt: was it a direct response to the existing one

227 00:26:21.800 --> 00:26:27.440 Z. Blace: or another to the just the previous one.

228 00:26:29.740 --> 00:26:30.890 Z. Blace: I don't know if

229 00:26:32.090 --> 00:26:46.360 Z. Blace: it makes sense for to see things, maybe a little bit more complex when making a decision. If something gets deleted or processed by an admin, or or

230 00:26:49.810 --> 00:26:53.360 Z. Blace: if it makes sense to delay this decision

231 00:26:55.270 --> 00:27:01.770 Z. Blace: by passing it on to someone who has already competencies rather than

232 00:27:02.310 --> 00:27:15.310 Z. Blace: each of the people who are enrolled in a process to have all the competences for all the countries. For me it sounds like a waste of energy and and and time for

233 00:27:15.400 --> 00:27:17.430 Z. Blace: people who are doing volunteer work.

234 00:27:17.640 --> 00:27:28.300 Z. Blace: And the other thing is there a reason why people who contribute to commons. Don't preemptively. Get an email as soon as they sign up.

235 00:27:28.380 --> 00:27:39.990 Z. Blace: That's Commons is much more complex than the Wiki. Let's say, Wikipedia default experience that you will be facing different criteria.

236 00:27:40.000 --> 00:27:50.150 Z. Blace: that you might have stuff deleted, that you shouldn't take it personally that if things get deleted that they can be also undeleted.

237 00:27:50.210 --> 00:27:56.270 Z. Blace: If people would get this email, people will not care of common soapy stuff

238 00:27:56.330 --> 00:28:06.040 Z. Blace: once this happens to them, because almost everyone experiences this at at certain point in time when they are starting with different types of

239 00:28:06.370 --> 00:28:10.820 Z. Blace: and just pointing them to like a pile of regulations.

240 00:28:11.050 --> 00:28:15.910 Z. Blace: is not the first thing that that that makes sense to to new users.

241 00:28:16.310 --> 00:28:24.400 Liam Wyatt: But yes, exactly. The the pile of regulations problem is something that we we suffer from across a lot of Wikimedia

242 00:28:24.730 --> 00:28:26.800 there is a

243 00:28:26.890 --> 00:28:41.510 Liam Wyatt: and Geon Palmier was responsible for the creation of some documentation. The first time you upload that it's a cartoon little puzzle piece that explains to you the very first time you go to with me to Commons.

244 00:28:41.770 --> 00:28:50.680 Liam Wyatt: If you have not already clicked, do not show me this again. That says, you know, this is an image of a public artwork. That's okay.

245 00:28:51.150 --> 00:28:56.710 but not in some circumstances. This is an image of myself. But who is the photographer?

246 00:28:57.270 --> 00:29:07.760 Liam Wyatt: Those kinds of educators were attempted to create some kind of easy education entry point. But it's very difficult to get people to read that.

247 00:29:08.730 --> 00:29:21.560 Liam Wyatt: and the number of exceptions to every circumstance. For example, public artworks are so long that everything you can say in an easy education email or or

248 00:29:22.050 --> 00:29:28.050 Liam Wyatt: pre upload warning message is followed by a series of.

249 00:29:28.550 --> 00:29:31.290 but in other circumstances it's different.

250 00:29:31.700 --> 00:29:34.390 Liam Wyatt: So it's very difficult to

251 00:29:35.570 --> 00:29:36.280 make it

252 00:29:36.290 --> 00:29:39.160 Liam Wyatt: enjoyable without also.

253 00:29:40.200 --> 00:29:47.930 Liam Wyatt: and and good without also creating a huge rule book that people will never read and ignore. In the first place.

254 00:29:48.420 --> 00:30:01.290 Shaun Spalding: I think I think something really interesting from from that comment is, it's not necessarily how to educate people on the rules as much as it is getting people ready for

255 00:30:01.320 --> 00:30:03.980 Shaun Spalding: the idea that

256 00:30:04.120 --> 00:30:15.430 Shaun Spalding: something that this will be a problem. Yeah, setting expectations. This will be a problem. It's not your fault. It's the fault of a complex system. Yeah. And I I think that there is kind of

257 00:30:15.780 --> 00:30:28.630 Shaun Spalding: something to that could be done about sort of expectation management, because people complaining people going to the help desk. That also takes a lot of time, especially when help desk questions are asking things that

258 00:30:28.860 --> 00:30:31.300 Shaun Spalding: have been answered over and over and over again.

259 00:30:32.740 --> 00:30:44.520 Liam Wyatt: Speaking of answering over and over again, that is also Dominique's question, which is in the chat. I want to hold that, and there's also there's 2 questions in the chat. I want to hold them. Martin has had his hand up for a long time.

260 00:30:45.300 --> 00:30:52.580 Liam Wyatt: and I want to come back to the asking the same question over and over again for for Dominic's point nothing.

261 00:30:53.660 --> 00:30:55.310 DerHexer/Martin: Yeah.

262 00:30:55.600 --> 00:31:00.320 DerHexer/Martin: yeah, I might be in the at cage.

263 00:31:00.610 --> 00:31:07.450 DerHexer/Martin: Sorry. and with a hundreds of thousands of photos uploaded to comments and some millions of added.

264 00:31:07.500 --> 00:31:09.750 DerHexer/Martin: But Well.

265 00:31:09.920 --> 00:31:22.820 DerHexer/Martin: I I have the feeling that a couple of users are also affected by this. So for me, upload, visit, doesn't, break, so I use external tool between the upload is the one which is working.

266 00:31:24.820 --> 00:31:28.750 DerHexer/Martin: however, for the amount of data

267 00:31:28.880 --> 00:31:29.910 DerHexer/Martin: and

268 00:31:30.740 --> 00:31:32.460 weird oh.

269 00:31:32.600 --> 00:31:35.420 DerHexer/Martin: as moderators also dependent on

270 00:31:36.910 --> 00:31:48.210 DerHexer/Martin: various other tools as scripts and gadgets, since, like visual, I I couch meant maintain comments without visual file change.

271 00:31:48.280 --> 00:31:50.960 DerHexer/Martin: and there is no any such tool

272 00:31:51.000 --> 00:31:59.080 DerHexer/Martin: on the software side. It's it's a gadget. We have the same issue. We have to a 100% dependent on catalog

273 00:31:59.190 --> 00:32:05.510 DerHexer/Martin: and master rename, for example, to to fix

274 00:32:05.740 --> 00:32:08.640 DerHexer/Martin: titles of pages or phone lines

275 00:32:08.760 --> 00:32:10.230 DerHexer/Martin: and

276 00:32:11.110 --> 00:32:15.350 DerHexer/Martin: partially, most of them reversible.

277 00:32:15.910 --> 00:32:30.070 DerHexer/Martin: However, that's not the case for the Sdc tool. So structured data, we can add a a mass level with Sc. But i'm always sitting there. These are choose the right ones, because I can't stop

278 00:32:30.500 --> 00:32:36.740 DerHexer/Martin: this account undo things I I was just adding so it's like

279 00:32:36.980 --> 00:32:49.650 DerHexer/Martin: when I noticed this after maybe 5 to 10, added, I can still stop the script, but then go to each and every page, and and added this manually. So

280 00:32:51.160 --> 00:33:09.900 DerHexer/Martin: I know there's a structure data team for commons at the we can your foundation, etc. But we still use this tool, at least for one I know, and I I don't have anything else to to you. Change

281 00:33:09.940 --> 00:33:12.790 DerHexer/Martin: data of of files.

282 00:33:13.120 --> 00:33:14.760 DerHexer/Martin: And this regard

283 00:33:15.170 --> 00:33:22.370 DerHexer/Martin: so specifically having an undo feature on the Sdc tool. The structure of our Commons tool is

284 00:33:22.430 --> 00:33:28.080 Liam Wyatt: missing. but I also saying that the variety of scripts.

285 00:33:28.200 --> 00:33:37.800 Liam Wyatt: some of which you named visual file change catalog and mass rename is a problem or an advantage, or there is an improvement that could be made to that

286 00:33:38.510 --> 00:33:46.850 DerHexer/Martin: as long as they run. That's okay. And I myself created a lot of work around

287 00:33:46.960 --> 00:33:52.230 DerHexer/Martin: for myself, and I educate other users on using these tools, but

288 00:33:52.320 --> 00:34:05.520 DerHexer/Martin: they are not very familiar. They are not very easy to find. We even had a mass download tool then that is deprecated, and since that I wrote my own

289 00:34:05.550 --> 00:34:16.270 DerHexer/Martin: Api down them all solution to download for the files and and and and a massive way. So it's like

290 00:34:18.100 --> 00:34:27.409 DerHexer/Martin: we comments Uses are pretty creative and using tools, and we're thankful for the tools we have, but it's pretty doesn't feel like it

291 00:34:27.670 --> 00:34:29.880 DerHexer/Martin: a solution. And

292 00:34:31.659 --> 00:34:35.710 DerHexer/Martin: we, for those who really maintain

293 00:34:36.080 --> 00:34:37.199 DerHexer/Martin: hundreds.

294 00:34:37.350 --> 00:34:44.630 DerHexer/Martin: hundreds of thousands of files, their own ones are in specific categories.

295 00:34:46.159 --> 00:34:54.190 DerHexer/Martin: They they need tools, we we need tools, we need tools. We can rely on a 5, 10, or more years.

296 00:34:55.820 --> 00:35:04.440 DerHexer/Martin: and I wrote some myself, so i'm. I'm also part of the problem and the solution, and

297 00:35:04.530 --> 00:35:07.560 DerHexer/Martin: but maybe not the real solution.

298 00:35:09.450 --> 00:35:12.130 DerHexer/Martin: I'm i'm looking forward to.

299 00:35:15.150 --> 00:35:25.220 Liam Wyatt: I've tried to note this in the in the chat. By the way, so keep. please read my notes as i'm writing them to make sure that your points are being well represented. I see

300 00:35:25.620 --> 00:35:32.870 AntiCompositeNumber (he/him): this is a continuation of what Martin was saying. Yes, we have a lot of tools currently.

301 00:35:33.290 --> 00:35:37.800 AntiCompositeNumber (he/him): Most of them are fairly old and fairly fragile.

302 00:35:38.500 --> 00:35:43.860 AntiCompositeNumber (he/him): which is not because their maintainers have left.

303 00:35:45.200 --> 00:36:02.870 AntiCompositeNumber (he/him): That's not something that can be easily fixed with a one-year. W. Muf comes up that comes around and fixes all the tools up to current standards. Although I would love that to happen that doesn't solve the problem of our critical infrastructure. Tools are unmaintained.

304 00:36:03.440 --> 00:36:14.800 AntiCompositeNumber (he/him): For example, visual file change has not changed significantly. In the last 5 years the most recent change was me fixing a media wiki Api change

305 00:36:15.330 --> 00:36:16.440 AntiCompositeNumber (he/him): that

306 00:36:17.020 --> 00:36:21.090 AntiCompositeNumber (he/him): broke the entire tool with one week's notice

307 00:36:21.310 --> 00:36:28.660 AntiCompositeNumber (he/him): before it happened. We didn't really fake. Think it was going to break anything. It broke everything, and then I had to fix it quite quickly.

308 00:36:31.550 --> 00:36:41.480 AntiCompositeNumber (he/him): That's not something that can be fixed by one year. Wf. Does a lot of work, which is why it's not really on my list of things I want changed.

309 00:36:42.220 --> 00:36:49.230 AntiCompositeNumber (he/him): There's a great tension between. Here is something that was independently built, and it's an important tool.

310 00:36:49.370 --> 00:36:51.820 Liam Wyatt: and it breaks because that

311 00:36:52.490 --> 00:37:03.240 Liam Wyatt: tension of If the Wikipedia foundation steps into to fix the thing, does it mean that we can be? The foundation now owns the thing, and therefore there is an expectation that it is part of core work

312 00:37:04.060 --> 00:37:11.920 AntiCompositeNumber (he/him): right, and that Hasn't been the case with deployed extensions. So we shouldn't expect that to be the case for

313 00:37:12.600 --> 00:37:16.280 AntiCompositeNumber (he/him): javascript that's maintained by the community. Traditionally.

314 00:37:19.650 --> 00:37:20.550 Liam Wyatt: Dominic.

315 00:37:20.890 --> 00:37:23.520 Liam Wyatt: I want to call back to your

316 00:37:23.990 --> 00:37:29.800 Liam Wyatt: question in the chat, which also refers back to a previous comment someone else has made as well.

317 00:37:32.340 --> 00:37:44.800 Dominic Byrd-McDevitt: Sure. Yeah, I I mean, I have kind of a 2 part question, because this is also getting back to the conversation, and what I think the and others are saying about setting expectations

318 00:37:46.880 --> 00:37:59.910 Dominic Byrd-McDevitt: feel like I have to ask if if the community is really on board with those expectations that you want to set. because my experience is the Commons community is really oriented toward

319 00:38:00.310 --> 00:38:01.350 Dominic Byrd-McDevitt: warning.

320 00:38:01.450 --> 00:38:02.980 warning all the time.

321 00:38:03.070 --> 00:38:16.570 Dominic Byrd-McDevitt: And, like Delete, this delete first kind of attitude, and so we can try to set expectations. But I don't know if the solution to the problem is necessarily

322 00:38:16.690 --> 00:38:19.960 Dominic Byrd-McDevitt: to tell users that all of their work might get

323 00:38:20.070 --> 00:38:24.390 deleted if they go away for a week, and they'll have

324 00:38:24.610 --> 00:38:36.850 Dominic Byrd-McDevitt: lots of warnings and shaming after that, and that you kind of have to be. If you upload something, you kind of have to be active for the rest of the time, and

325 00:38:37.090 --> 00:38:50.690 Dominic Byrd-McDevitt: available to defend your actions in a like 7 day time period. which is what it can feel like this. This is a problem that that I come up with come up against, because

326 00:38:51.820 --> 00:38:55.410 you know. through the pot that I run deeply. But

327 00:38:55.680 --> 00:39:05.820 Dominic Byrd-McDevitt: we've uploaded more than 3 million file. and from from many different contributors contributing cultural heritage institutions.

328 00:39:06.210 --> 00:39:14.190 So with that many uploads, even with a very tiny error rate, You know, I I have to contend with a lot of deletion

329 00:39:14.260 --> 00:39:16.140 Dominic Byrd-McDevitt: nomination, like

330 00:39:16.320 --> 00:39:18.400 probably on a weekly basis at least.

331 00:39:19.960 --> 00:39:22.160 Dominic Byrd-McDevitt: but they fall into like many

332 00:39:22.200 --> 00:39:24.960 predictable categories.

333 00:39:25.260 --> 00:39:37.400 Dominic Byrd-McDevitt: There's and and you can. I looked back at my talk Page archives, and I would say less than half of them are actually successful. So i'm I'm

334 00:39:37.450 --> 00:39:38.570 oftentimes

335 00:39:38.840 --> 00:39:40.390 Dominic Byrd-McDevitt: arguing

336 00:39:40.550 --> 00:39:44.900 at these these deletion nominations where there's not a lot of people there.

337 00:39:45.000 --> 00:39:57.790 Dominic Byrd-McDevitt: Do you kind of have an obligation to show up, or otherwise your thing might get deleted, even if there shouldn't be. But it's the same for like years now the same types of discussions over and over again, of like

338 00:39:57.990 --> 00:40:06.070 Dominic Byrd-McDevitt: some basic things that that come up a lot, such as either people not understanding that

339 00:40:06.370 --> 00:40:09.310 Dominic Byrd-McDevitt: Wiki source or other projects value

340 00:40:09.420 --> 00:40:30.900 Dominic Byrd-McDevitt: a page that you think is a blank page, but it's part of the larger work, or things like that non content. images or things that you think are. But there's nothing of value there, but it's part of the larger a larger work. It's a a photo of space taken by NASA. Sorry it's black. I have those kinds of conversations, a lot or things like people that aren't taking

341 00:40:31.170 --> 00:40:37.980 Dominic Byrd-McDevitt: aren't, valuing the copyright assessment that was made by the institution that holds an an object, and they're like

342 00:40:38.040 --> 00:40:42.160 asking for proof. Sorry I don't have proof that that

343 00:40:42.370 --> 00:40:46.230 Dominic Byrd-McDevitt: an actual museum said that the the copyright of

344 00:40:46.350 --> 00:40:47.100 they own

345 00:40:48.200 --> 00:40:54.370 with I mean, we can debate how to handle that. But it's more the issue of that. These are really repetitive discussions.

346 00:40:56.200 --> 00:41:05.350 Dominic Byrd-McDevitt: and I think it's a time thing for the people that are closing those discussions, but also a time sync for the people that have to show up to and

347 00:41:05.380 --> 00:41:06.180 does

348 00:41:06.390 --> 00:41:08.100 Dominic Byrd-McDevitt: again again.

349 00:41:09.410 --> 00:41:13.380 Liam Wyatt: I think we all, especially in the Glam world, had

350 00:41:13.610 --> 00:41:22.020 Liam Wyatt: cases like you of having to defend. Look that that comes from that collection it's in the collection.

351 00:41:22.750 --> 00:41:27.910 Liam Wyatt: This is that's that's where it comes from. But equally we've seen institutions.

352 00:41:28.350 --> 00:41:30.590 Liam Wyatt: hallucinating copyright statements.

353 00:41:32.200 --> 00:41:44.020 Liam Wyatt: and it's very hard for individuals, and particularly Commons, uses UN associated, UN affiliated with that with that institution to know how to deal with it. Commons seems to have become

354 00:41:44.510 --> 00:41:49.890 Liam Wyatt: the world's largest clearing house of intersectional copyright law.

355 00:41:50.980 --> 00:41:53.680 Liam Wyatt: and we're all self taught in it.

356 00:41:54.780 --> 00:42:07.400 Liam Wyatt: Is there a. Since this conversation today is coming from the perspective of the product and tech department at the locomotive foundation, and we do have someone from the legal department here listening as well.

357 00:42:07.580 --> 00:42:09.290 Liam Wyatt: Is there a way

358 00:42:09.500 --> 00:42:13.530 Liam Wyatt: that this challenge you're expressing of repetitive

359 00:42:13.750 --> 00:42:17.640 Liam Wyatt: work. Either Indonesian nomination or

360 00:42:17.920 --> 00:42:21.270 Liam Wyatt: deletion Defense upload of defense

361 00:42:21.730 --> 00:42:26.550 Liam Wyatt: can be addressed as a workflow from a technical

362 00:42:26.630 --> 00:42:35.030 Liam Wyatt: perspective, not merely from a better documentation and legal training perspective.

363 00:42:38.950 --> 00:42:51.580 Dominic Byrd-McDevitt: I think so. Maybe I think that there might be some workflow solutions. I'm a little unsure of like what specific

364 00:42:51.760 --> 00:43:01.450 Dominic Byrd-McDevitt: suggestions to make about that. The one thing that I would make. maybe in the opposite direction is to be wary when you're

365 00:43:01.530 --> 00:43:03.060 when you are making.

366 00:43:04.490 --> 00:43:09.190 Dominic Byrd-McDevitt: and just to the workflow that that automate or make things easier for the

367 00:43:09.450 --> 00:43:21.030 Moderators to be wary of taking things in the other direction, because I think, from the perspective of new users, and and also people that do a lot of uploads.

368 00:43:21.060 --> 00:43:36.830 Dominic Byrd-McDevitt: It can feel like the Commons community already has to your finger. So I worry if we make things easier like what what the fallout might be if we make it too easy, or or at least easier, to make it

369 00:43:38.880 --> 00:43:40.410 Dominic Byrd-McDevitt: give. Give people

370 00:43:40.520 --> 00:43:43.660 on the other end of this a negative. Okay.

371 00:43:44.610 --> 00:43:47.220 Liam Wyatt: that's that's the constant challenge. And

372 00:43:47.660 --> 00:43:50.910 Liam Wyatt: referring back to what we were discussing earlier by making it

373 00:43:52.730 --> 00:43:58.900 Liam Wyatt: easier, fewer warnings and fewer blockers in the way results in potentially more

374 00:43:59.820 --> 00:44:05.220 Liam Wyatt: bad content, which creates more workflow for the existing admins. But also

375 00:44:05.440 --> 00:44:18.070 Liam Wyatt: that's just about creating roadblocks to make life difficult. I recall that the original prior to that cartoon based upload a helper with it that I mentioned before

376 00:44:18.300 --> 00:44:23.800 Liam Wyatt: there was always a trap suggestion in the upload form

377 00:44:24.600 --> 00:44:30.550 Liam Wyatt: that did then. Is this your own photograph? Is this an old painting? Is this something you found on the Internet

378 00:44:30.720 --> 00:44:31.560 Liam Wyatt: which

379 00:44:31.670 --> 00:44:46.700 Liam Wyatt: was put in there deliberately to tell people to go away, which is rare as a user interface in the Internet, to to make volunteer or to make people wanting to contribute to leave because you misunderstood the instruction.

380 00:44:47.740 --> 00:44:51.000 Liam Wyatt: but that does so for valid purpose.

381 00:44:58.120 --> 00:45:03.600 Liam Wyatt: Sean. There's a I think you're you're coming in. The chat is a direct. Would you like to reply, Say that yourself?

382 00:45:05.700 --> 00:45:06.370 Shaun Spalding: Oh.

383 00:45:11.820 --> 00:45:16.270 Shaun Spalding: it takes like 10 s for this mute to come off? I'm so sorry

384 00:45:16.300 --> 00:45:20.350 Shaun Spalding: I press the button, and I think it's like not gonna

385 00:45:20.580 --> 00:45:34.420 Shaun Spalding: quick. And then okay, anyway. there the one of the impetuses for the conversation was to talk about that particular fabricator ticket that was referenced.

386 00:45:34.460 --> 00:45:44.570 Shaun Spalding: That suggested that a copyright violation tool. So, for example, a 10 I reverse image search should

387 00:45:45.160 --> 00:46:01.930 Shaun Spalding: potentially be triggered upon. For example, a contributor uploading something in the upload Wizard and I'm. Curious whether people believe that is one of the ways to STEM the tide of

388 00:46:03.650 --> 00:46:19.080 Shaun Spalding: problematic stuff coming in at the start of the process. So if somebody uploads a I don't know a picture of Spiderman or something like that, and the reverse image search tool comes up and says, hey, this might be likely to be

389 00:46:19.210 --> 00:46:20.800 Shaun Spalding: copyright infringement

390 00:46:20.910 --> 00:46:23.000 Shaun Spalding: in

391 00:46:23.340 --> 00:46:24.740 Shaun Spalding: your country.

392 00:46:24.850 --> 00:46:28.030 Shaun Spalding: Consider not uploading this.

393 00:46:28.120 --> 00:46:37.670 Shaun Spalding: Do you think that would be helpful? Or do you think that that would be sort of? Not the type of thing that would move the needle much?

394 00:46:41.000 --> 00:46:45.620 Dominic Byrd-McDevitt: This is the kind of thing where I think what I was just talking about, about.

395 00:46:47.110 --> 00:46:57.830 Dominic Byrd-McDevitt: that the potential pitfalls, because certainly, like one of the discussions that I have a lot like I was saying is people that are

396 00:46:57.930 --> 00:46:59.450 disregarding

397 00:46:59.460 --> 00:47:14.450 Dominic Byrd-McDevitt: museum or cultural heritage, right statements, right? So like it's true that they can be wrong. that they're created by humans. But I think once in an institution once the expert it's already on an image has made it

398 00:47:14.540 --> 00:47:19.040 Dominic Byrd-McDevitt: determination. Then the burden of proof is kind of on.

399 00:47:19.290 --> 00:47:23.370 Dominic Byrd-McDevitt: I should, I should be on the the

400 00:47:23.380 --> 00:47:31.940 contesting it. But oftentimes it's kind of like commons, has this perverse burden of proof where it's like. You can come along and say the

401 00:47:32.290 --> 00:47:38.780 Dominic Byrd-McDevitt: this this author dive in the State. Therefore it's not in the public domain. and i'm kind of like.

402 00:47:40.420 --> 00:47:42.930 Dominic Byrd-McDevitt: The Museum also had that information.

403 00:47:43.170 --> 00:47:45.720 and so they probably have some reason, like

404 00:47:46.220 --> 00:47:59.950 Dominic Byrd-McDevitt: the person donated to to them, and donated the right. and it's in a contract somewhere that's not public knowledge right? But if I have to, if I have to like, argue just based on things that are available to

405 00:48:00.130 --> 00:48:16.150 Dominic Byrd-McDevitt: only to a common person on the Internet, but not things that are real knowledge that's available to the Museum. Then I can't win that argument. So there there is. That's part of my fear with something like this. Automating it to like this is we would want to make sure that it's not.

406 00:48:16.330 --> 00:48:22.890 Dominic Byrd-McDevitt: and encouraging people to make assumptions that are overriding. You know the authorities.

407 00:48:29.230 --> 00:48:34.660 Liam Wyatt: Thanks so many. I'm. I'm not quite sure how to summarize that down. But I I I feel your point about

408 00:48:35.460 --> 00:48:38.840 Liam Wyatt: the burden of proof to prove a negative often

409 00:48:40.040 --> 00:48:52.390 Liam Wyatt: but equally in the chat. And and you we personally have also seen where the institutions. different institutions, have absolutely different understandings of what counts as copyright.

410 00:48:52.580 --> 00:48:54.590 Liam Wyatt: you know, particularly for public domain

411 00:48:54.780 --> 00:48:56.740 Liam Wyatt: paintings that have been scanned

412 00:48:57.360 --> 00:49:01.850 Dominic Byrd-McDevitt: just just really, quickly.

413 00:49:01.880 --> 00:49:06.610 I agree with that. That. I'm. I'm not just that. I I think I'm just making the

414 00:49:06.970 --> 00:49:15.240 Dominic Byrd-McDevitt: complimentary point, which is that you also Don't, have to be a copyright expert to started to deleting nomination on common.

415 00:49:18.000 --> 00:49:22.130 AntiCompositeNumber (he/him): See, I I won't really get into that side of the point more. But

416 00:49:22.190 --> 00:49:26.940 AntiCompositeNumber (he/him): yes, I would say that upload filters, even if they weren't

417 00:49:28.040 --> 00:49:32.610 AntiCompositeNumber (he/him): deny filters, even if they are only warnings. are

418 00:49:32.790 --> 00:49:41.500 AntiCompositeNumber (he/him): probably not going to be useful on commons just because of how many of the works that are on Commons are

419 00:49:42.820 --> 00:49:58.040 AntiCompositeNumber (he/him): intentionally taken correctly from other websites, like, even if we had a perfect a detector that detected only the exact duplicates. and that checked for a copyright statement on the

420 00:49:58.240 --> 00:50:02.910 AntiCompositeNumber (he/him): on the Source website, and it's only if it's at all Rights Reserve Well, that would still

421 00:50:03.000 --> 00:50:06.520 AntiCompositeNumber (he/him): fail. If you're taking works from

422 00:50:07.460 --> 00:50:17.290 AntiCompositeNumber (he/him): various British institutions that love to claim copyright on their paintings of for or on their photos of 400 year old paintings.

423 00:50:18.760 --> 00:50:19.670 AntiCompositeNumber (he/him): or

424 00:50:19.920 --> 00:50:27.800 AntiCompositeNumber (he/him): this file came from flickr, and it has. It's a free license. But it's not machine readable readable.

425 00:50:28.830 --> 00:50:32.830 AntiCompositeNumber (he/him): That's a much more likely circumstance, then

426 00:50:35.830 --> 00:50:46.900 AntiCompositeNumber (he/him): then actually finding a copyright violation, because we know that freely liced images are used everywhere. and we know that nobody does attribution correctly

427 00:50:47.120 --> 00:50:50.800 AntiCompositeNumber (he/him): and definitely. Nobody does machine readable attribution correctly.

428 00:50:51.130 --> 00:51:02.090 AntiCompositeNumber (he/him): So even if we got it working perfectly, it would still fail way too often for it to be useful it from my perspective, and I think Jenny will have even more to say on that point.

429 00:51:04.000 --> 00:51:04.810 Liam Wyatt: Jimmy

430 00:51:14.030 --> 00:51:14.860 Liam Wyatt: Kenny.

431 00:51:19.940 --> 00:51:22.140 wikigeni: Hello, yeah. What

432 00:51:22.430 --> 00:51:26.230 wikigeni: I was more concerned that this is

433 00:51:27.080 --> 00:51:31.830 the problem is, as soon as you build a filter people want to start using it for other things, and

434 00:51:32.250 --> 00:51:37.220 wikigeni: some of those people are the you, some of the Russian Government. Some of them are China.

435 00:51:38.040 --> 00:51:50.470 wikigeni: So if you even think about going this direction it you have to be very careful. It has to be absolutely under control our control. So tin is out. So you're trying to build an entire copyright, scanning BoT from scratch.

436 00:51:50.470 --> 00:51:59.350 wikigeni: because, as you get involved in third parties, it's not open source. You've lost control, and other people can now censor all the key, and you would like to ask why we now not censoring terrorist content.

437 00:52:01.790 --> 00:52:14.980 wikigeni: So yeah, just for me to be careful of in terms of false copyright claims. Yeah. And all over the place Bunch of people take stuff off commons, upload it on their own site, claim copyright or simply don't give us credit.

438 00:52:15.080 --> 00:52:17.760 And yeah, your bots can start running to that as well

439 00:52:20.290 --> 00:52:24.300 Liam Wyatt: and flicker washing and other such workarounds deliberate work around.

440 00:52:24.550 --> 00:52:28.540 wikigeni: Yeah, you can go in your direction. Obviously claim this is free when it's not.

441 00:52:30.580 --> 00:52:32.830 Liam Wyatt: I've seen several people speak

442 00:52:32.900 --> 00:52:38.530 Liam Wyatt: against such an upload filter and not filter a copyright violation.

443 00:52:39.380 --> 00:52:52.660 Liam Wyatt: wanting to built into the upload. not with saying that that it has been requested and suggested and come up in various Wish list locations.

444 00:52:53.170 --> 00:52:58.230 Liam Wyatt: Was there anyone in this in this room who wanted to like to speak

445 00:52:58.790 --> 00:52:59.760 Liam Wyatt: 4

446 00:53:00.430 --> 00:53:03.940 Liam Wyatt: such a thing, notwithstanding the the that the

447 00:53:04.020 --> 00:53:07.430 Liam Wyatt: the valid concerns that have been raised. Susannah.

448 00:53:08.500 --> 00:53:15.720 Susanna Ånäs: I would just like to bring to your attention a proposal by the open futures

449 00:53:15.750 --> 00:53:23.760 Susanna Ånäs: who have proposed to have a public domain copy filter

450 00:53:23.900 --> 00:53:27.410 Susanna Ånäs: a like a commons

451 00:53:27.480 --> 00:53:33.700 Susanna Ånäs: guarding and patrolling the use of openly licensed images in the world.

452 00:53:33.930 --> 00:53:35.920 Susanna Ånäs: a kind of an inverse. So

453 00:53:36.130 --> 00:53:37.680 Liam Wyatt: could you please link to that.

454 00:53:37.780 --> 00:53:41.870 Liam Wyatt: and we'll put it into the

455 00:53:43.430 --> 00:53:51.520 Liam Wyatt: We are at time for the for the official duration of this meeting. So please don't feel obliged to stay longer than

456 00:53:52.570 --> 00:53:59.780 Liam Wyatt: then you have to. I will leave the notes here open and unlocked for anyone to edit for another

457 00:54:00.140 --> 00:54:16.150 Liam Wyatt: day or few hours, and then i'll change it to no further editing. and we'll link to that from the video that's been uploaded here are all copied across, or something like that. But if you you have some time to reread any notes to make sure that your points were

458 00:54:16.720 --> 00:54:19.460 Liam Wyatt: adequately addressed or summarized.

459 00:54:22.280 --> 00:54:26.070 Liam Wyatt: Did anyone have any comments in the chat that I did not

460 00:54:27.210 --> 00:54:30.650 Liam Wyatt: raise. I can't mention. I can't

461 00:54:31.180 --> 00:54:32.060 Liam Wyatt: remember.

462 00:54:47.790 --> 00:54:52.150 Liam Wyatt: There was one question in the in the comments with regards to

463 00:54:52.750 --> 00:55:01.210 Liam Wyatt: statistics, off into the workflows and and use of audio and visual playback statistics.

464 00:55:02.810 --> 00:55:06.660 Liam Wyatt: Could you? Could you start us that with you? I believe. Could you

465 00:55:08.170 --> 00:55:09.130 Liam Wyatt: speak to that?

466 00:55:14.650 --> 00:55:22.270 Liam Wyatt: Okay, in the chat? The question about media file usage statistics. Are you considering access to audio and video playback statistics?

467 00:55:22.330 --> 00:55:37.390 Liam Wyatt: So things like, how many times was this video played? How long was it played for. and so forth? I'm not even sure if those statistics to the degree of granularity that say Youtube, or spotify.

468 00:55:37.710 --> 00:55:42.640 Liam Wyatt: have I'm not sure if they exist for our on commons. Because

469 00:55:43.620 --> 00:55:54.860 Liam Wyatt: if a video is, you or file is used in a Wikipedia article. because that's when someone presses play

470 00:55:55.320 --> 00:56:05.610 Liam Wyatt: on the Wikipedia article. Does that count as one play, even though they only may have listened to 1 s? How long of the file do you need to

471 00:56:06.680 --> 00:56:13.230 Liam Wyatt: count as a as a full. Listen. especially if people embed a a section

472 00:56:13.410 --> 00:56:15.970 Liam Wyatt: of the video or the file.

473 00:56:16.590 --> 00:56:25.760 Liam Wyatt: About 30 s is standard since Boris, so i'm not sure if we keep a statistics I am ignorant of on what that is true or not, Sam. Do you know?

474 00:56:28.220 --> 00:56:29.660 Sam Walton: Absolutely no idea.

475 00:56:33.950 --> 00:56:46.150 Liam Wyatt: But I can try and find out. I don't think that counts, or if it is, if it does exist and can be made public, and and we just haven't for whatever reason.

476 00:56:46.190 --> 00:56:58.130 Liam Wyatt: even if we do, I don't think that necessarily counts as an improved workflow. Just an improvement itself. That's a good thing to have. But there have been lots of conversations

477 00:56:58.590 --> 00:56:59.300 Liam Wyatt: and

478 00:56:59.750 --> 00:57:05.790 Liam Wyatt: people wanting to improve the workflow and handling of multimedia files. Not just, not just images

479 00:57:06.030 --> 00:57:09.770 Liam Wyatt: in general statistics of that is

480 00:57:10.590 --> 00:57:12.520 Liam Wyatt: is is important, too.

481 00:57:17.360 --> 00:57:23.480 Liam Wyatt: I don't think there is any further comments or new comments in the chat. Did anyone want to add anything

482 00:57:24.640 --> 00:57:26.690 Liam Wyatt: before we leave?

483 00:57:33.590 --> 00:57:38.880 Liam Wyatt: And yes, in the chat. when a file is simply made available, an image is.

484 00:57:39.420 --> 00:57:50.120 Liam Wyatt: you know, Wikipedia article, and that Wikipedia article is loaded by someone that's counted as a view, even if the images way down the page, and no one ever actually saw it.

485 00:57:50.630 --> 00:57:53.600 Liam Wyatt: Or if it's used in a template somewhere.

486 00:57:54.180 --> 00:58:04.900 Liam Wyatt: I've had that as a problem of had approved to Glam, their content is actually being seen. because the most popular image might actually be used as a thumbnail

487 00:58:05.060 --> 00:58:08.740 Liam Wyatt: in a template, and and we never, actually you can

488 00:58:08.830 --> 00:58:13.580 Liam Wyatt: prove that the the reader saw it or clicked on to make it large

489 00:58:15.610 --> 00:58:20.280 AntiCompositeNumber (he/him): for Media. That should be easier because the

490 00:58:20.430 --> 00:58:26.510 AntiCompositeNumber (he/him): file itself is only downloaded if you click on.

491 00:58:26.770 --> 00:58:34.200 AntiCompositeNumber (he/him): I I think that tool mostly does what you want within the limitations of

492 00:58:34.220 --> 00:58:39.120 AntiCompositeNumber (he/him): the Wikimedia type of analytics, of not collecting that much.

493 00:58:40.110 --> 00:58:43.020 AntiCompositeNumber (he/him): If If we had javascript that reported

494 00:58:43.130 --> 00:58:51.890 AntiCompositeNumber (he/him): everyone's exact specific watch times, I think people would actually complain about that. So I don't know how much of that data is

495 00:58:51.940 --> 00:58:54.310 AntiCompositeNumber (he/him): possible to collect within our framework.

496 00:58:54.490 --> 00:58:56.810 AntiCompositeNumber (he/him): That that was a good point.

497 00:58:57.640 --> 00:58:59.520 AntiCompositeNumber (he/him): that that tool should get you close.

498 00:59:01.070 --> 00:59:14.510 Dominic Byrd-McDevitt: Yeah, Just to to add to that, it's a question of the tools that we have. as well as the data that we have, because the Api is so show you that. But what you're describing Leon is a lot of the

499 00:59:14.590 --> 00:59:18.930 Dominic Byrd-McDevitt: the main tools that we have for showing like category, level or aggregate

500 00:59:18.970 --> 00:59:22.400 page views for comments. Files

501 00:59:22.490 --> 00:59:33.780 Dominic Byrd-McDevitt: are not counting the media the request they're counting the page views of the articles that use an image. even if there there was never a request made

502 00:59:33.920 --> 00:59:36.430 to that for that media on the server

503 00:59:38.460 --> 00:59:39.300 Liam Wyatt: yet. By

504 00:59:41.440 --> 00:59:47.580 Liam Wyatt: okay, I will wrap it up there. We are slightly over time. I'll just put one more link in the chat here.

505 00:59:47.660 --> 00:59:50.740 Liam Wyatt: which is again to the talk page

506 00:59:50.850 --> 00:59:54.520 Liam Wyatt: of the We commit a foundation annual plan.

507 00:59:55.070 --> 00:59:59.480 Liam Wyatt: Okay, Ours. The objectives and key results.

508 00:59:59.490 --> 01:00:11.340 Liam Wyatt: This conversation, this video is, and not a replacement for the top page. It is a supplement to it, and that top page is divided up into sub headings based on the various

509 01:00:12.690 --> 01:00:22.980 Liam Wyatt: goals that are proposed to the annual plan. So that's where the conversations link that is their home page. This video will be published on comments, I hope.

510 01:00:23.120 --> 01:00:32.310 Liam Wyatt: and I can link it back into the documentation of various kinds of conversations that have existed across the Wikipedia movement as part of the annual plan.

511 01:00:32.400 --> 01:00:47.490 Liam Wyatt: and I will also point the notes we've made today to the various teams and responsible people for the relevant subsections of that annual plan, so they can see this feedback themselves as they take this feedback and

512 01:00:47.960 --> 01:00:57.610 Liam Wyatt: turn the annual plan into a a practical. What does my team do this year. Workflow. What roadmap?

513 01:01:00.480 --> 01:01:04.690 Liam Wyatt: Yes, there's there's always a new Tla

514 01:01:05.140 --> 01:01:07.790 Liam Wyatt: TLA a three-letter acronym.

515 01:01:09.120 --> 01:01:12.890 Liam Wyatt: Thank you for your for the time. Everyone, and

516 01:01:13.010 --> 01:01:17.050 Liam Wyatt: good evening, all Good afternoon, all good morning, depending on where you are.