Wikipedia:Village pump (miscellaneous)

From Wikipedia, the free encyclopedia
Jump to navigation Jump to search
  Policy   Technical   Proposals   Idea lab   Miscellaneous  
The miscellaneous section of the village pump is used to post messages that do not fit into any other category. Please post on the policy, technical, or proposals sections when appropriate, or at the help desk for assistance. For general knowledge questions, please use the reference desk.
« Older discussions, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60

Research project: The effects of specific barnstars

I'm a researcher studying the effects of rewards on motivation. My team and I would like to run an experiment with barnstars on Wikipedia. We'll use data analysis to identify editors who are making substantial contributions to Wikipedia and we'll post barnstars on their talk pages. Our goal is to see if specific types of barnstars elicit different types of motivation reinforcement for an editor. For example, if we reward an editor for doing copy-editing work specifically, are they more likely to continue doing copy-editing work?

Our goal in all of this is to find effective strategies for increasing the long term retention of editors. To us, it appears that identifying way to motivate volunteers to stay and contribute is very important. For the continued viability of Wikipedia. And anyway, it's good to recognize people for their hard work.

I'm posting here today to announce the project and to invite you to highlight any potential issues with us posting barnstars. See metawiki:Research:How role-specific rewards influence Wikipedia editors’ contribution for our proposed study design. There, you can find our criteria for selecting editors to receive barnstars. We're interested in your insights. We're also very interested in working with anyone who would like to collaborate with us in this experiment. --Diyiy (talk) 23:20, 18 December 2018 (UTC)

One problem with your experiment seems pretty obvious; The "reward" comes from an editor without any substantial contributions to the encyclopedia: you. Anyone who'd receive a barnstar from you would immediately notice that they're participating in an experiment. Vexations (talk) 22:36, 18 December 2018 (UTC)
@Vexations Thanks for pointing out this issue. We won't post the messages using the account of Diyiy (talk). We would like to collaborate with some highly experienced Wikipedia editors and ask them to help us post the barnstar messages. If you are interested, that will be great! We're also considering posting the messages using Robert Kraut's account, a senior contributor to en_wiki from our research team. --Diyiy (talk) 23:20, 18 December 2018 (UTC)
With no disrespect to Dr. Kraut, I would not consider an editor with under 500 edits to be a senior contributor to Wikipedia. To put things in perspective, according to WP:SERVICE, a senior editor has at least 24,000 edits (although it is worth noting that service awards don't really mean all that much and are really just a measure of how much time you spend on the site). signed, Rosguill talk 02:23, 19 December 2018 (UTC)
@Rosguill Thanks! Following your and Vexations's suggestion, I just added one specific section in our research project description to look for volunteers - experienced Wikipedia editors - to help us with the posting. If you are interested, please let us know. -- Diyiy (talk) 05:13, 19 December 2018 (UTC)
  • @Diyiy: I'm wondering what your criteria for Editors in the top 10% of those making edits adding substantive content or performing copy-editing actions are. Is the top 10% just an evaluation of edit volume? I would be interested in helping hand out the barnstars, but I'm skeptical that edit volume is really indicative of quality contributions and would feel uncomfortable handing out barnstars on that basis alone. Additionally, I'd suggest an exclusion criterion that should be added: editors who are watching this page or who have participated in discussions on this page while this discussion is on the page. signed, Rosguill talk 07:15, 19 December 2018 (UTC)
@Rosguill: Great, we'll be very happy to have your help! So, top 10% will include two aspects: (1) in terms of the total number of edits in the last month, and (2) high portion of edits that are adding substantive content or copying editing. We have some prior work with Halfak (WMF) where we built machine learning models with reasonable accuracy to automatically predict what work has been done in a revision such as adding substantive content, copy editing, counter-vandalism, vandalism (You can find the details in this page). We will rank editors in terms of their total number of edits and the portion of adding substantive content edits or copy editing edits, and then select the barnstar recipient candidates from this pool. In this case, we could guarantee that our barnstar recipient candidates are making quality contributions, not vandalism or other type of edits. Your exclusion criterion makes sense - we need to exclude people who have participated in our project discussions ("Watching" seems a bit impossible to be excluded because of Wiki privacy issues). I added this to our research project description. --Diyiy (talk) 14:54, 19 December 2018 (UTC)
@Rosguill: In addition to the quantity metric, if collaborators also want to include a personal quality assessment , that would be fine. For the purposes of the experiment, the only requirements are that the editors getting barnstars should deserve them based on Wikipedia standards and that once the pool of deserving editors is identified (through a combination of automated identification and human judgment), those who get barnstars are randomly selected from the pool. Robertekraut (talk) 15:20, 19 December 2018 (UTC)
  • Sorry but I'm not happy about this. Please see "Wikipedia is not a laboratory". The proposal could be regarded as somewhat "disruptive to the community" in diluting the value of the barnstar, which we would hope is intended as a sincere expression of appreciation from one Wikipedia editor to another. Now I wouldn't pretend the proposal is some kind of life-threatening intervention, but by its nature the experiment cannot seek the informed consent of participants and as such does not set a good example. Wikipedia editors are not lab rats and should not be fed barnstars to see if they scurry round any faster afterwards! Feel free to disregard this if other contributors don't see it this way: Noyster (talk), 18:36, 19 December 2018 (UTC)
User:Noyster The experiment is essentially Nudge theory, which the CIA uses to wage psychological warfare, so I'm sure it's safe and not distruptive... Cesdeva (talk) 18:45, 19 December 2018 (UTC)
I have to assume you're being somewhat ironic. I, too, am skeptical of turning barnstar-giving into a lab experiment, which seems a "blunt tool" (to quote Hannibal Lecter), although it's quite possible that to recognize adept but heretofore unrecognized editors in any way may encourage them to continued productivity, even if they're aware of their being subjects of an experiment. Dhtwiki (talk) 00:02, 20 December 2018 (UTC)
My delivery was a bit too dry. Ya i'm against making barnstar lab rats out of us without informed consent. I think you may be right; i've never been given a barnstar and my productivity has always been more sporadic than a randy mushroom. Cesdeva (talk) 00:40, 20 December 2018 (UTC)
User:Noyster I don't see the experiment as disruptive. We are proposing to give barnstars to a subset of Wikipedians who deserve them but haven't yet received any. I don't see how doing so dilutes their value. Barnstars are an established wiki recognition system, and we are proposing to use them for their intended purpose--to recognize some overlooked Wikipedians for their meritorious service. Wikipedia already uses machine-learning methods in the form of the | ORES service to identify bad behavior; In the experiment, we'll be using machine learning to identify good behavior. Our innovation here isn't the tool to identify people deserving of barnstars. Rather it is the attempt to understand the consequences of a common activity on Wikipedia, the giving of barnstars. Robertekraut (talk) 02:26, 20 December 2018 (UTC)
If you give out more barnstars than given previously, you could well dilute their value; but the main objection I see is that, for the recipient, instead of receiving an expression of appreciation from a grateful fellow editor, you are receiving something from people who are not working in the trenches with you and are just experimenting to see how you can be made to do more, better, faster, which can be easily felt as belittling. Given the complex set of inducements that I've experienced, where barnstars play only a part, I don't see how your study would show anything more than some minor and flawed correlation of effort to reward. Dhtwiki (talk) 06:12, 20 December 2018 (UTC)
  • So, couple of thoughts. Given your exclusion criteria I'm not entirely sure how many editors are going to be left. Keeping in mind that most of the work on Wikipedia is done by a small minority of users, and a comparatively large number of those are going to have advanced permissions, and run or control bots, and virtually all of them will have at some point received a barn star. So based on just your exclusion criteria, I expect you're going to be well below the 50th percentile of most active editors that your considering, and I wonder how many of these are active enough to observe any appreciable difference in the editing habits. Also, how are you going to tell how many of ~300,000 active monthly users have never received a barnstar? GMGtalk
@GreenMeansGo: This is good point! We wrote computer scripts to look at the user page and user-talk page of each editor to see whether there is a barnstar type of message. Taking into account your concerns, we reduced the number of editors we are going to give barnstars (400 barnstars recipients and messages in total). We followed prior work in terms of inclusion and exclusion criteria [1][2]. --Diyiy (talk) 22:43, 19 December 2018 (UTC)
I have removed my barnstars from my user page as I am not a tinpot general who wants to flaunt his medals, because I want my page to be uncluttered by what I regard as merely gaudy crap, and because I don't care whether I receive barnstars or not. My motivation to edit Wikipedia is purely to provide quality information for the enrichment of readers. Knowing that I have done something useful, and of quality, is reward enough. Therefore, surveying who has barnstars, and who has not, is a flawed procedure. Lots of editors delete barnstars or move them to a subpage, sometimes, and sometimes not, leaving a link on their main user page. The fact that they have a collection of stars to their credit is not immediately apparent. Your computer scripts cannot identify deleted barnstars. — O'Dea (talk) 18:44, 1 January 2019 (UTC)

References

  1. ^ Restivo, Michael, and Arnout van de Rijt. "No praise without effort: experimental evidence on how rewards affect Wikipedia's contributor community." Information, Communication & Society 17, no. 4 (2014): 451-462.
  2. ^ Restivo, Michael, and Arnout Van De Rijt. "Experimental study of informal rewards in peer production." PloS one 7, no. 3 (2012): e34358.
Second, I'm not sure you have considered the confounding effect of the person giving the barnstar. Again, fairly small core community and many users are fimiliar with one another. If a user gets a barnstar from someone they highly respect, or even a conciliatory barnstar from someone they were previously in a dispute with, that could send a much different effective message than a barnstar from an anonymous person. GMGtalk
@GreenMeansGo: Yes, there seems a confounding factor about the relation between the giver and the receiver. It is challenging to assess whether the giver was previously in a dispute with the receiver or not. We plan to seek help from several senior editors to help us post the barnstar messages. This sort of reduces the variability in the giver's side. In this stage, we are interested in the overall effects on whether receiving barnstar is associated with higher productivity, and incorporating more contextual factors like the relation between givers and receivers will be our future research to look at. --Diyiy (talk) 22:43, 19 December 2018 (UTC)
Third, you want to give out 1,200 barnstars in one day?? Again, fairly small community, half of us have half of us on one another's watchlists, and that is well above a level that could be considered normal. I mean, I have currently 3,737 user pages on my watchlist. May users will have more. I have a hard time believing you don't run a significant chance of raising quite a few eyebrows and opening a few discussions trying to figure out what's going on. Not to mention that if I look at someone's account who is an established user, and I see that they've spent the last hour spamming 200 users with barnstars, people are going to start wondering whether these accounts aren't compromised, unless you have quite a large pool of editors handing them out, to the tune of only three or four each. GMGtalk 19:12, 19 December 2018 (UTC)
@GreenMeansGo: We mentioned this and will highlight this in our project description - "We will post around 50 barnstar messages per day (each volunteer will help with 10 barnstar messages) for one or two weeks for this barnstar experiment". In the updated project, only 400 editors will actually receive barnstars, not 1,200. Please let us know if you have any further questions --Diyiy (talk) 22:43, 19 December 2018 (UTC)
  • Well, I'll be honest Diyiy, I think it's an interesting idea, and one that could actually inform practice here, and not just serve some esoteric academic interest. So I'm trying to work through things as best I can. (Also generally it's common practice to put comments one after the other, and not interspersed, although I get that you were trying to address each point.)
I think you need to test your exclusion criteria mathematically before you get married to it. In order to get meaningful numbers, you need to be testing users who edit several times a week, although not necessarily every day. I'm sure you've heard of the Long tail phenomenon with regard to online participation. That's true here too, and most of those ~300k active monthly users only make a handful of edits per month. So I expect there's going to be too much stochasticity in their editing behavior to draw any meaningful conclusions. In other words, their editing behavior is going to be much more influenced by what happens off-wiki, than what happens on-wiki. If you're not informed by that, then you're going to draw statistical conclusions that aren't based in reality. That is, unless you target those users in particular, but you can't mix the edit-every-day users with the rest, because there is a base level of time commitment that makes them meaningfully different in their behavior.
With regard to the "giving" user, the most pure measure would be to use a bot, but that's gaining purity at the expense of the native experience. At the very least I think you should either completely exclude admins from your participants, or have enough admins distributed evenly that you can statistically control for the notoriety that that permission in particular has among the community. You probably also want to exclude other "prestigious" positions, like anyone on WP:ARBCOM and anyone who is a major project coordinator, like WP:MILHIST or WP:FAC. You probably just want a blanket exclusion criteria for "givers" that says "do you hold any position in the community that would give you particular notoriety". (Although the reaction to "prestigious" members of the community giving recognition is a worthy question in itself.)
In regard to the volume, you just gotta slow your roll. Ten barnstars in a day is still entirely too much to seem natural. No editor on Wikipedia naturally gives out ten barnstars in a day. The most giving users might give out one every couple of weeks or every month. You're gonna have to space this out over time to get meaningful results if you are committed to the volume. Often the first thing a user does when encoutering another user is to look at their recent edits, and if they see multiple barnstars in their last 50 contribution (the default number displayed), then it's going to be a suspect. Of course that depends on the size of the effect, and if the effect is large, then a smaller sample size may be adequate.
Perhaps one alternative to consider is, rather than barnstars, to examine WP:THANKS. Thanks, while ostensibly logged publicly, are extremely difficult to access the log for, and so wouldn't get spoiled by volume. GMGtalk 23:22, 19 December 2018 (UTC)
GreenMeansGo you might be interested in this project. J-Mo 21:43, 21 December 2018 (UTC)
That is quite interesting J-Mo. Not least of which is the fairly large cultural variations in the use of the feature. Incidentally, it makes me wonder whether there is any coordination happening between meta and people like Bri over at the Signpost to regularly update the editorial staff when this kind of thing is published. GMGtalk 22:01, 21 December 2018 (UTC)
To be fair, I see less of a human subject problem when someone is only offering positive interaction. Compare a design where someone offered subjects on the street $10 randomly, to observe their subsequent behavior. But again, IRB needs to make a decision on informed consent. I'm also not totally convinced that submitting the IRB approval via OTRS isn't a thing we should be doing as a matter of course. GMGtalk 01:06, 20 December 2018 (UTC)

Break

I fail to see why the experiment is necessary. It would be better if the researchers found the data they're looking for in the existing logs and edit histories without handing out barnstars themselves. Vexations (talk) 23:24, 19 December 2018 (UTC)
Vexations, the difference is that one is randomly selected and the other isn't. I'm still interested to see what their IRB says with regard to consent to participate in a study on the part of the participants, but you can't draw very many meaningful conclusions from non-random selection. GMGtalk 00:08, 20 December 2018 (UTC)
GreenMeansGo, I guess what I'm saying is that I much prefer an observational study to a non blind randomized controlled trial. I'm not satisfied that the results of an observational study are necessarily less reliable than this RCT, given the flaws in the design of this proposal. To put it bluntly; all the participant will know they get the placebo. Vexations (talk) 02:07, 20 December 2018 (UTC)
GreenMeansGo This research proposal has been approved by Carnegie Mellon University's IRB, waiving informed consent. The reasoning is that the experiment doesn't expose Wikipedians to any more risk than they would otherwise be exposed to (i.e., receiving an unsolicited barnstar) and couldn't be practically done without waiving informed consent. Robertekraut (talk) 02:26, 20 December 2018 (UTC)
How can Carnegie Mellon waive getting informed consent from Wikipedia editors? The University has no power over nor any relationship with Wikipedia editors. I think it is a bad idea. - Donald Albury 02:55, 20 December 2018 (UTC)
IRB doesn't have any power over Wikipedia. We're free to ignore their recommendations either way. Approval does tell us that their design has been reviewed by professional researchers who haven't identified major ethical concerns. (I'd be interested to get input here from User:EpochFail and User:Jtmorgan.) GMGtalk 11:55, 20 December 2018 (UTC)
I'm not surprised that the CMU IRB approved this study. Believe it or not, waving consent in these types of situations is pretty common practice due to the low risk -- as Robertekraut mentioned. I definitely agree with GMG that we're not under any obligation to allow the study to continue. It is up to us if we would like to allow these kinds of experiments to proceed in Wikipedia. It's my opinion that, it's in our best interests to work through the details with the researchers so their work can proceed with minimal disruptions. As GMG points out, this research has the potential to inform practice on Wikipedia. I think GMG has raised some excelling points for the researchers to review (e.g. the disruption of sending out 1200 barnstars, the unreasonable filter criteria, etc.). Noyster has relayed some important concerns about how studies like this might dilute the meaning of a barnstar. I think that if the researchers can figure out a way to adjust their study protocol to deal with these sorts of issues and we help them figure out how to minimize disruption, then we'd be better off by allowing them to continue and demanding that they make their research results publicly available gratis (and maybe also open access) so that we can take our best advantage of their work.
Either way, I think the issues raised in this conversation represent a large set of valuable insights about how barnstars work in Wikipedia and how Wikipedians expect that they will be used. Personally, I'm very interested in how barnstars are applied and I've learned a few things from this conversation. I'd like to perform some analyses to if the data aligns with Noyster's assertions about the rate at which barnstars are applied. I think it would be fascinating to look for examples of highly productive editors who are working in a barnstar desert where no one notices or sees fit to appreciate the kind of work that they do. Maybe the researchers could find a way to help us generate a list of such editors and we could be the ones to send the barnstars to these under-appreciated editors. --EpochFail (talkcontribs) 14:25, 20 December 2018 (UTC)
More just musing really, but I do wonder what measurable difference there would be between private thanks notifications and public barnstars. In other words, is it the act of appreciation that makes a behavioral difference, or it is the public "token" displayed for others. But maybe that's a question of another day. GMGtalk 15:11, 20 December 2018 (UTC)
I get the occasional "thanks" (maybe once a week), which I do appreciate, but I don't remember getting a barnstar, and if I ever did, I didn't leave it displayed. I guess I ran in a crowd that didn't give out many barnstars (these days I feel like a bit of a loner, some of my Wiki friends have dropped out, and I just don't interact much with others). My participation in WP has waxed and waned over the years, but I don't remember feedback from other Wikipedians playing a role in that. - Donald Albury 16:43, 20 December 2018 (UTC)
  • @Diyiy and Robertekraut: we don't give out barnstars to motivate people to stay, but to thank them for work we've noticed or something they've helped with. I'm concerned about the ethics of your proposal (not to mention that you may have destroyed it by discussing it in public). You plan to use the recipients by fooling them, not even letting them know they're taking part in research. In addition, lots of editors are addicted to editing Wikipedia and are not served well by anything that motivates them to stay. See this BBC report about social-media companies using social-validation feedback loops to create environments that are addictive and exploitative. Wikipedia shouldn't be involved in anything like that, at least not on purpose. Does the Wikimedia Foundation know about this proposal? SarahSV (talk) 22:11, 20 December 2018 (UTC)
  • @SlimVirgin: One goal of this research is to develop bettter methods to recognize others' work. We have developed a set of machine learning techniques to predict whether an editor has made more than average specific edits around some skills such as adding substantive work or doing copy editing. Via this, we will be able to 'thank' editors who have done much but haven't received any thanks from Wiki community. For the ethnics of this proposal, as Robertekraut mentioned, it has been approved with a university's IRB. The reason why we discuss it here in public is to get valuable feedback and suggestions from Wiki editors before doing this work - people who have participated in this discussion will be excluded from our barnstar work. We have been discussing this proposal with EpochFail from Wikimedia Foundation. --Diyiy (talk) 15:31, 21 December 2018 (UTC)
  • Well give them credit SV, we ask researchers to come & consult with us and they did so (indeed NOTLAB points researchers to the higher-profile Village pump (proposals)). But yes, the scheme would seem to depend on keeping the bulk of the editing community in the dark, with risk of harm not to individuals but to community interaction, by introducing an ulterior motive into one of our mutual recognition devices: Noyster (talk), 15:42, 21 December 2018 (UTC)
(EC) I should make it clear that I am not operating as WMF Staff when supporting this project. This project is not officially endorsed by the Wikimedia Foundation and I am not an official collaborator on the project. With that said, my professional opinion (Wikipedian-researcher who contributes content and tools for Wikipedians) is that finding ways for studies like this to proceed without causing disruption is very desirable and I think it is safe to say that we (the researchers and Research team) at the Wikimedia Foundation generally want to see good research on motivation and social reinforcement patterns to help inform movement strategy work, WMF Product development, grant making strategies, etc. I think about the patterns of addiction that SarahSV brings up quite often. I think that avoiding sending reinforcing messages to users who are putting in potentially unhealthy hours on Wikipedia is quite possible. In my work studying hours people spend on Wikipedia (pdf), there was a pretty clear divide between people who spent a few hours per day on average and people who were spending 8-10 hours every day. Maybe we should consider filtering these users out of the study so as to not risk harming them. It's hard to say as they are also likely the most deserving of a barnstar. I'm not an expert in addiction research, so it's hard to say how harmful a random barnstar could possibly be. In the end, it may be best to leave that judgment call to a real, live human rather than some quantitative threshold. If diyiy et al. were to develop a strategy to help experienced editors find editors deserving of barnstars (e.g. by producing a list with statistics), there could be a real live human deciding when to send a barnstar or not. --EpochFail (talkcontribs) 15:51, 21 December 2018 (UTC)
EpochFail, the proposal is to find a bunch of people who are doing unpaid labour—labour that might be damaging them and to which they may be addicted—and see what kind of poke will induce them to do more. Read the proposal (permalink) for yourself; they are quite clear about it. They want to know whether "activity-specific barnstars reinforce the actions that they reward".
From a practical standpoint, do you not suppose the recipients will notice something is amiss? The proposal is that 800 barnstars be handed out, which is disruptive. In fact, I wouldn't be surprised if the involved accounts end up being blocked. I thought the WMF had to approve research that interferes with the community. SarahSV (talk) 20:57, 21 December 2018 (UTC)
SlimVirgin No, WMF doesn't have that kind of authority over any Wikimedia project community (and the communities themselves would probably not want WMF to have that power). Many of the researchers who work at WMF, including EpochFail and me, try to keep an eye out for research that could be disruptive or even harmful, and discourage it. We also hear about a lot of research proposals ahead of time, through our academic networks and by monitoring proposals on the Research index, and we try to encourage researchers to follow best practices—such as bringing their projects up for review and discussion in forums like this one. We do this for three reasons: 1) we care about Wikipedia, 2) we care about research ethics, and 3) we do research on Wikipedia ourselves, and bad research erodes community trust, which makes our jobs more difficult. That's one of the reasons I got involved in the crafting of WP:NOTLAB, too. As for the current project, I can vouch for the integrity and the scientific cred of the researchers involved. It's up to the community to decide whether this project should proceed. I think reasonable concerns have been raised, and that includes yours. Sounds like the researchers are willing to work with the community to address some of these concerns, and that the research—if allowed to proceed—may be better for the conversation. So the process seems to me to be working well, right here and now, with no WMF oversight. J-Mo 21:41, 21 December 2018 (UTC)
J-Mo and EpochFail, I don't have time to look for links, but I recall several discussions over the years about researchers being told they must gain approval from the WMF for certain kinds of research.
Regardless of that, the ethical problems are obvious. They plan to go ahead without any kind of consent (even generic consent to be part of a research study), despite the fact that the study has the potential to harm its participants. Research increasingly shows how harmful social-media addiction is, and this study aims to determine whether "activity-specific barnstars reinforce the actions that they reward".
Feeding or triggering an addiction is just one of the potential harms. The study also has the potential to embarrass or humiliate the recipients, who will be chosen only if they've never received a barnstar. Experienced editors will recognize there's something odd about the barnstar, because the proposed wording smacks of "your call is important to us", and because the accounts handing them out won't have many edits and will be distributing lots of them. Less experienced editors may not notice these things, and will believe that another editor has genuinely thanked them for their work. It's disrespectful to fool someone in that way, especially in public. Imagine if that editor follows up by thanking the account that posted the barnstar and trying to engage them in conversation. What then? Will the researchers ignore them or keep up the pretence? SarahSV (talk) 22:51, 21 December 2018 (UTC)
I am deeply confused by an argument that says we should not encourage people to edit Wikipedia. GMGtalk 00:46, 22 December 2018 (UTC)
From reading of all of this I fail to see the benefits to anyone, and also find the comments from SV quite disturbing in that the points that she raises seem to invoke, the responses to date appear thoroughly unconvincing.
I strongly object to this sort of behaviour on wikipedia, from ethical bases alone. If I read the responses correctly from those conducting this work, I sure hope that someone or other raises the ethical issue other than self or SV - this is so odd, as to give it a sense that we can do this under the radar, and a sense of without having any problems about it. JarrahTree 03:00, 22 December 2018 (UTC)

───────────────────────── While I think the results of the study could be very enlightening, I agree that this sounds a little too disruptive and disingenuous: I guess I fundamentally don't agree with awarding barnstars semi-arbitrarily; if someone told me that my first barnstar was awarded to me as part of an experiment to see if my behavior changed as a result, I'd probably be a little disappointed and upset to be used as a lab rat. There has to be enough legitimate barnstars already awarded for real reasons to study passively, aren't there? Can't a study be designed as a strict observation of existing barnstar awards to accomplish the same thing? Is there some reason that the study must award its own barnstars? CThomas3 (talk) 03:14, 22 December 2018 (UTC)

I agree with CT , JT and SV these if I started getting barnstars as sort of lab rat experiment for a third party I would walk away. If I was to receive a complaint about an account just dishing out barnstars and doing nothing productive as an admin I'd warn them for being disruptive if it continued I block the account due to past experience WP:BEANS. Gnangarra 10:53, 22 December 2018 (UTC)
  • Why not thanks? - I know it's been mentioned already, but if we can help them work the thanks logs that would seem a vastly better approach. 10 thanks a day is "merely" very eager, 10 barnstars a week, let alone a day, is insanely high. Additionally, I'd be find with receiving a semi-arbitrary thanks, but agree with the potentially counter-productive barnstar issuing. Nosebagbear (talk) 11:35, 22 December 2018 (UTC)
  • Has the person organizing this research project considered vicarious learning - i.e. learning by seeing other people reinforced? It would be interesting to see whether people's motivation is increased if they see other people getting barnstars. Vorbee (talk) 17:40, 23 December 2018 (UTC)
  • Totally ignoring the negative aspects of the mass spamming of barnstars, of the devaluing what is current just just a nice thought. Lets ignore the disruption, the potential for abuse, how do you even measure what the impact has, how do you tell the difference between motivation and opportunity? Everyone here is a volunteer they contribute what they can when they can, what we do know is that everyone contributes to wikipedia for their reasons. How can you tell that a barnstar is the motivator when it may be that they injured themselves and cant do other things, maybe they just got access to an interesting collection resources that they have been trying to access for a few years. Maybe they went on holiday and discovered a new area of interest, maybe they have been snowed in(or its too hot outside) and have nothing else to do. How do you establish that difference by looking at the current, what if you give a barnstar to a disruptive editor, does that reinforce poor behavior, what methodologies have go you propose to ensure that in motivating a person you dont embolden them to be more disruptive, do even have a plan in place to redress such actions. Gnangarra 15:53, 24 December 2018 (UTC)
The reason to do a true, random-assignment experiment to control for exactly the type of problems in interpretation that Gnan and Vexations have raised. Editors' motivations and availability to participate in Wikipedia vary between people and over time within the same person. The assumption in a random assignment experiment is that the people who receive the experimental treatment (i.e., a barnstar or a thanks) are on average the same as those who are in the non-treatment, control condition. Therefore, if the treatment and control groups differ on some outcomes of interest, such as the amount and type of work they do on Wikipedia, their willingness to talk on article talk pages, etc.) one can attribute these differences to the experimental treatment. The standard inferential statistics, like t-tests and regression analyses, gives an estimate about how likely the observed difference would occur by chance because of natural variations between people and time. Robertekraut (talk) 15:57, 28 December 2018 (UTC)
In the final analysis both "barn-stars" and "thanks" are means of communication. Correct me if I am wrong but in the final analysis this proposal is about something I would characterize as false communication. No one knows what I mean if I give them a "barn-star" or I send "thanks". I am communicating but it is never clear what I am saying. (Probably here as well.) There is a degree of unclarity whenever we deploy language to communicate ideas. The suggestion here is that we increase the noise that is present in verbal communication involving "barn-stars". I believe I have to oppose that. Bus stop (talk) 16:27, 28 December 2018 (UTC)
The fallacy here is that a barnstar or a thanks is a motivating factor, when we have a pile of studies that show the motivation behind a person contributions are distributed across many reason there is no singular reason why people contribute. How do value an contribution do allow for volume of content added, or number of edits, do you discount tool usage or promote it, what about the gnome areas closng AFD, CfD, protections and NPP. No two legitimate people contribute in the same way for the same reasons for the same amount of time. How do you stop people in the control group from receiving barnstars and thanks are going to remove them if they do. The whole issue is giving someone a barnstar and them then making 10 edits doesnt prove the barnstar motivated them. Gnangarra 17:05, 28 December 2018 (UTC)
The "thanks" that I receive are for specific edits I have made, and have always been from a user that I recognize as being active in editing the specific article. Random thanks from editors that I do not recognize would strike me as strange. - Donald Albury 17:23, 28 December 2018 (UTC)
I don't think "thanks" are always from editors one recognizes. One presumes the person passing along "thanks" is highly in agreement with some input you have made. I oppose this experimentation because it adds to the uncertainty over what motivates someone to use the tools of "barn-stars" or "thanks". There are many downsides to "barn-stars" or "thanks". Nevertheless I don't advocate abolishing them. But weakening them or making their significance less clear makes little sense to me. Bus stop (talk) 17:38, 28 December 2018 (UTC)
(edit conflict) What Donald Albury & Bus stop said^^^.
To me, Barnstars awarded among Wikipedia editors and the WikiLove messages I give and receive actually mean something. To use the Barnstars (and potentially the WikiLove system) in the researchers' proposed way devalues their meaning and processes them into something different, something that I personally would not appreciate receiving (you know, like the difference between White Stilton cheese and pasteurized-processed cheese spread...). Maybe the researchers should come up with their very own "Your-work-has-been-seen-and-valued-as-meaningful-by-a-team-of-researchers-at-Carnegie-Mellon-University-and-this-particular-recognition-and-your-continued-contributions-after-receiving-this-recognition-is-being-used-as-part-of-a-CMU-motivational-research-project" Barnstar/WikiLove thingy. Shearonink (talk) 18:05, 28 December 2018 (UTC)
I feel that pasteurized-processed cheese spread is being treated unfairly in this discussion. Bus stop (talk) 18:26, 28 December 2018 (UTC)
Single wrapped slice of processed cheese.jpg

As an aside...To show pasteurized process cheese and its cheese spread that there are no hard feelings: Mmmmmmmmm...pasteurized-process cheeeeeeeese. Shearonink (talk) 21:46, 28 December 2018 (UTC)

Ahh, the delicate qualities of processed cheese! Bus stop (talk) 15:45, 30 December 2018 (UTC)

Break 2

If you want really see if barnstars work, try this find a sample size of users who meet at least the following criteria;

  1. havent edited for more than 3 months
  2. have made more than 2000 contributions, over at least a continous 6 month period
  3. were in good standing at time of last edit
  4. edited in both article and WP space

This criteria proves the people are long term editor who contributions are likely accepted, and importantly something the community would encouraged. Additionally they will be contactable by email this user. Divide the sample into 3 groups;

  • Group A gets "we miss you, please return" barnstar.
  • Group B receives and email with a similar request about being missed and that it would a good if they returned.
  • Group C receives neither.

Measure their activity 1 day, 1 week, 1 month for however long you can. Then you're able to at least make a plausible causal inference about the impact of each style of contact including a correlation of the outcomes between each group both in the number and size of edits also the length of time they edit. Gnangarra 17:40, 28 December 2018 (UTC)

  • I see that a similar type of barnstar research was done in 2014: Resvito, Michael; Van de Rijt, Arnout (2014). "No praise without effort: experimental evidence on how rewards affect Wikipedia's contributor community". Journal Information, Communication & Society. 17 (4): 451–462. doi:10.1080/1369118X.2014.888459.

Our experimental treatment was to assign an informal, peer-to-peer reward, or ‘barnstar,’ to contributors who had never previously received one ... We focused on the top 10% of Wikipedia’s contributor community ... we removed those users who had administrative or elevated privileges, and excluded ... the bottom 90% of contributors by edit volume in the prior 30 days. We also screened out users who had previously received a barnstar ... After stratifying the population, we conducted a simple random sample of 200 subjects from each tier. The treatment consisted of anonymously placing a barnstar on the subjects’ user-talk page ... The barnstar we chose ... expressed community appreciation for their contributions, but it was not tailored to any recipient-specific activities or achievements. ... Contrary to theory, rewarding less productive editors did not stimulate higher subsequent productivity. ... The experimental results show that rewards can be used to sustain productivity among highly-active contributors at the top of the distribution, yet are ineffective in this regard for less-active contributors.

I wonder whether they had consensus to do it. SarahSV (talk) 22:49, 28 December 2018 (UTC)
An earlier experiment by Resvito used a similar procedure:

We test the effects of informal rewards in online peer production. Using a randomized, experimental design, we assigned editing awards or “barnstars” to a subset of the 1% most productive Wikipedia contributors. Comparison with the control group shows that receiving a barnstar increases productivity by 60% and makes contributors six times more likely to receive additional barnstars from other community members, revealing that informal rewards significantly impact individual effort.

FYI, we contacted Resvito and asked more about the procedure in these experiments. They gave out barnstars using their personal accounts, which had very few contributions. They gave all the barnstars out during a one or two week window. We failed to ask them if got feedback and consensus from the community before conducting their experiment. Robertekraut (talk) 23:20, 28 December 2018 (UTC)

Summarizing the discussion about an experiment to examine the effects of barnstars

To try to summarize the discussion and reach consensus, here are what I think are the main points raised so far. Sorry if I’ve gotten anything wrong or missed important points, but the discussion has been long, meandering and continually changing. In the summary, I’ll generally use the general term “recognition,” where Barnstars, Thanks, and WikiLove are all forms of positive recognition, because many of the concerns raised in the discussion apply independent of the type of recognition.

  • Need for the research:
Several Wikipedians, including User:GreenMeansGo, User:Jtmorgan, EpochFail and User:Cthomas3, thought the proposed research is asking interesting and useful questions about the downstream effects of Wikipedia recognitions, which may have implications for retaining valuable editors. GreenMeansGo suggests that it might be interesting to examine the effects of different types of recognitions – barnstars, which tend to be more public and have community defined meanings, versus thanks, which tend to less public and are based on the thanker’s private criteria.   Vorbee thought the question could be extended to examine the impact of vicarious learning by observing the impact on others who merely witness someone getting a barnstar.
However, User:Vexations, User:CThomas3 and others challenge the need for experimental research when the questions could be answered through observational methods.  We will indeed conduct the observational, correlational analyses that Vexations and CThomas3 recommend. But as explained several times in the discussion, a random assignment experiment is desirable because it is the best way to differentiate the ‘effects’ of recognition from ‘selection’, i.e., that people who receive recognitions at a particular time are likely to be different from those who never received the recognition or received it at a different time.
  • Ethics: Harm to individual editors.
Several editors highlighted possible harm to the individual editors who would receive recognitions. User:Noyster, User:Cesdeva, User:CThomas3 and user:Gnangarra all mention some concerns about editors' dignity and autonomy by treating them as “lab rats” in which they are part of an experiment to which they did not explicitly consent.  This is indeed a legitimate concern and one we considered when weighing the possible benefits that Wikipedia and the scientific community would get from understanding the effects of recognition against this possible harm to individuals who receive them. Although reasonable people can disagree, user:Diyiy and I (along with our IRB) think the potential benefits outweigh these risks.
SarahSV raised the concern that receiving recognitions would harm editors more tangibly by increasing editing addiction. But this is a critique of Wikipedia’s recognition systems rather the research itself.  If receiving recognition increases editing addiction, then we should ban Barnstars, Thanks, and WikiLove and not research that tries to understand the effects of receiving them.
  • Ethics: Harm to the community.
Several people raised the concern that increasing the number of recognitions distributed will devalue them and the devaluation will be even worse if the community believes they were being distributed as part of an experiment run by a “third party”.  I think this concern about the devaluing recognitions is the most important one raised in the discussion so far.
First let me dispel the ‘third party’ claim.  Although as Rosguill pointed out, I am not a senior editor with over 24,000 edits, I consider myself a member of the Wikipedia community. I am not just treating Wikipedia as a petri dish for research but am a community member concerned about the health of Wikipedia. By starting the Association for Psychological Science's Wikipedia Initiative, by recruiting PhD economists to evaluate and provide feedback on Wikipedia articles and by assigning Wikipedia writing assignments to students in my classes since 2011 and closely mentoring them as they worked on articles, I’ve been indirectly responsible for improving scores of Wikipedia articles. Indeed the only barnstar I’ve received was for “helping your students make welcome improvements across a range of psychology articles MartinPoulter.”
People have made a number of suggestions of reduce possible harm.
  1. Distribute Thanks instead of Barnstars, because they are less visible. We are considering this possibility and would like more feedback on this decision. Our concern is that Thanks may have less powerful effects and have less clearly defined criteria behind them. If we distribute Thanks rather than Barnstars, we may need to distribute more of them.
  2. Reduce the perception of spam by cutting down the number of recognitions awarded per day.  We have modified our original proposal and are proposing to distribute only 400 recognition and can spread the distribution over a longer time (e.g., 4 or 6 weeks).  
  3. Increase the number of editors who will distribute the recognitions. We are recruiting editors to help with this.
  • Other suggestions for improving the research
Discussants raised the question of who should give the recognitions. Some of the commenters thought the experiment wouldn’t work if the recognitions come from “from an editor without any substantial contributions” to Wikipedia. Others cautioned us if the recognitions come from administrators, project coordinators and others who have prestigious" positions.  Correlational research we have done shows that when editors receive feedback on their talk pages, the effects of the feedback on their subsequent motivation are generally stronger when the person providing the feedback is an administrator rather a non-administrator [Effectiveness of Shared Leadership in Online Communities]. For example recipients make more edits after they’ve received positive feedback, task suggestions or social comments and make fewer edits if they received negative feedback. Most of these effects are stronger if admins provide the feedback. However, our own experiments and those on [informal rewards in Wikipedia] by Restivo and colleagues show that feedback even from relatively new editors changes the behavior of those who receive it. We will try to recruit experienced Wikipedians to help us distribute recognitions, but if we fail, we'll continue with less experienced ones. Robertekraut (talk) 23:05, 28 December 2018 (UTC)
Robertekraut, I think you've made a good-faith attempt at summarizing what has been said so far. I had not yet come to a conclusion myself, so I will summarize my own ideas about this research in response: I think that all forms of deception are blockable offenses. That includes attempts at influencing editor behavior, even if it is intended to be beneficial. Simply put: If I say something nice to someone because I want them to stay, that's OK, as long as I make it clear that editor retention is my goal. If I do not make it clear what I am doing (study their behavior), I am deceiving the recipient and I deserve to be indefinitely blocked for my dishonesty. I would support blocks against anyone actively participating in this study, and I would also support a full ban on all research that involves any form of psychological manipulation, that is, any research that is not strictly observational only. Vexations (talk) 00:48, 29 December 2018 (UTC)
  • I'd say this is a fair summary. I appreciate that you seem to have carefully taken in to consideration our feedback. The only other feedback I would give is that as far as rate goes, it may be helpful, once you have established participant givers, to ask them what their opinion is about how many barnstars might be given out in a "natural state" rather than trying to set some comparatively arbitrary measure. If such a study last four months, rather than six weeks, but still gathers reliable data, then that is not time lost. Regardless, I would not set a rate that would be higher than one per 50 edits, for reasons outlined above regarding the interface.
Overall, I think personally that more recognition should be given out by more people generally because we mostly do a good job at doing what we do and most editors are underappreciated. So I don't see the imminent harm in encouraging editors to do so. I don't really understand the argument that encouraging people to edit Wikipedia is somehow harmful. It's deeply counterintuitive given the ongoing editathons and workshops designed to do just that. But if anyone wants to explore the issue in more detail, my talk page is always open to all comers. GMGtalk 00:22, 29 December 2018 (UTC)
  • The issue I see is that the trial lacks any understanding of the community that contributes, the assumption that a barnstar is a motivating factor for further contributions is a fallacy. Secondly the disruption this has the potential to cause is incredibly high, see WP:BEANS. I have yet to see anything that would satisfy me that such disruption has any value to the community and I would block participants actively spamming barnstars. Gnangarra 05:13, 29 December 2018 (UTC)
  • I hope everyone realizes how extraordinarily easy it would be for someone to totally invalidate any conclusions this research project might reach. All they would have to do is follow the bot handing out barnstars, and then add a comment saying: “Note: this barnstar may have been awarded to you by a bot, as part of a research project to see how you react”. This simple comment would let editors know they are the subject of an experiment, and that would change their behavior... thus tainting your data sets. Blueboar (talk) 18:26, 29 December 2018 (UTC)
  • It might have made sense to focus on barnstars 10 years ago, when they were much more widely used. There are all sorts of prizes and rewards that can affect behavior, but I think most of them are probably trivial as compared to specific personal appreciation expressed in an individual way. . Singling out barnstars by itself is I think a poor research design, though it is certainly easier than considering multiple factors. And, if anyone does think they are meaningful, then handing them out by bot would certainly be disruptive. DGG ( talk ) 21:16, 8 January 2019 (UTC)
    • Agree with DGG that the present-day barn-star is "specific personal appreciation expressed in an individual way" and for that I hereby bestow upon DGG one virtual barn-star. Bus stop (talk) 17:54, 10 January 2019 (UTC)

Relation to Facebook

Diyiy, Your user page User:Diyiy links to your homepage http://www.cs.cmu.edu/~diyiy/ where you state "I am supported by Facebook Fellowship" Can you clarify if and how Facebook benefits from the research that you propose here? Thanks, Vexations (talk) 20:49, 28 December 2018 (UTC)

Ah yes...thanks Vexations for pointing that out. I'd be interested to know this info as well. Shearonink (talk) 21:30, 28 December 2018 (UTC)
Let me respond, since I'm User:Diyiy's PhD adviser. Facebook fellowships are the result of a international competition to identify, encourage and support promising doctoral students who are engaged in innovative and relevant research in areas related to computer science and engineering. Facebook won't benefit at all from the research we've been describing. More information about Facebook fellowships are at [[1]]. Robertekraut (talk) 22:13, 28 December 2018 (UTC)
Robertekraut, if someone is being paid to do this, either directly or indirectly, whether as a result of employment, a grant, etc, then that person is a paid editor and subject to WP:PAID. This is a policy derived from the WMF terms of use. SarahSV (talk) 22:25, 28 December 2018 (UTC)
SarahSV, I'm not following your logic in this response. First, I think you are misstating the WMF's terms of use and the related policy, which is that editors must disclose their employers or clients if they are making paid edits. For example, when EpochFail edits articles or builds tools as an employee of the WMF, he is not violating Wikipedia's policy. Second, although Diyiy and I have jobs, Diyiy is not being paid by Facebook and I am not being paid by the National Science Foundation or my university to edit Wikipedia. Therefore, we are not making paid contributions, which are ones that involve contributing to Wikipedia in exchange for money or other inducements. Robertekraut (talk) 23:41, 28 December 2018 (UTC)
From an outsiders point of view, reading this conversation - this getting quite confusing - not being paid to edit wikipedia - but actually being paid to induce editors to respond to externally funded research? JarrahTree 23:59, 28 December 2018 (UTC)
Robertekraut, the terms of use and WP:PAID say "you must disclose your employer, client, and affiliation with respect to any contribution for which you receive, or expect to receive, compensation." Note: "and affiliation". Any contribution is subject to this, whether editing articles, posting on talk or handing out barnstars. The compensation can be direct ("here's $200; go and edit that Wikipedia article"). Or it can be indirect ("here's a research grant; using it to do research on Wikipedia would be an appropriate use of it"). When EpochFail acts on WP as an employee of the WMF, he discloses; if he didn't, he would be violating the terms of use.
Perhaps you and Diyiy could explain why you want to conduct this research and in whose interests. It seems unlikely to produce any result that you'll notice. If you do notice one, that could be because you've triggered someone's addiction. During the 2014 research, giving barnstars to less-active contributors made no difference, which the addiction hypothesis would explain. This is like going to a local bingo hall, finding the most active players, and doing something to encourage them. Will encouraging them make them play more bingo? Yes, just about anything will make some of them play more, and perhaps a lot of them play more, because some or a lot of them are addicted. Therefore, don't do that research. Similarly, don't tweak it to say "but we want to know whether doing X will make them play more in a certain way". That's just as unethical and just as unlikely to produce a meaningful result. Diyiy, what is your PhD topic? SarahSV (talk) 00:13, 29 December 2018 (UTC)
Robertekraut, perhaps it is helpful to link to the Research Summary at [2] Vexations (talk) 00:17, 29 December 2018 (UTC)
JarrahTree, Vexations and SarahSV, to make it clear, I am not paid by Facebook to edit Wikipedia; the proposed work has nothing to do with Facebook. This barnstar project is a follow up on our previous Wiki work about developing algorithms to identify editors' expertise and skills, and an extension on prior research Resvito, Michael; Van de Rijt, Arnout (2014). "No praise without effort: experimental evidence on how rewards affect Wikipedia's contributor community". Journal Information, Communication & Society. 17 (4): 451–462. doi:10.1080/1369118X.2014.888459.. --Diyiy (talk) 02:15, 29 December 2018 (UTC)
Diyiy, thank you for the response. Could you say what the point of this research is, and in whose interests it's being done? It seems related to finding ways to entice people into the behavioural loops so beloved of Facebook and others. What would be the benefit to Wikipedia? SarahSV (talk) 02:42, 29 December 2018 (UTC)
SlimVirgin As stated in our research proposal, the point of this work is to determine whether rewarding Wikipedia editors through barnstars for the type of editing they do most regularly, such as adding substantive content to articles or copy-editing them, increases their willingness to continue that type of work in Wikipedia compared to editors who are eligible but did not receive these rewards. By doing so, we can better understand whether barnstars on Wikipedia can motivate editors to upload their engagement. It also allows us to understand whether giving people barnstars that fits their behaviors increases editors’ contribution more than giving people generic barnstars. We also hope that this can provide some insights on how to timely recognize editors’ work (especially editors who do a lot of work but are unrecognized by the community) and foster their willingness to contribute to Wikipedia. Benefits to Wikipedia: (1) A number of editors who are doing great work will receive barnstars showing the community’s recognition to their contribution to Wikipedia, which may increase their willingness to contribute to Wikipedia in the future. (2) Our research can provide insights into Wikipedia with the development of barnstars guidelines. For example, if people who receive barnstars that fit their editing behaviors make more contribution in the future compared to people who receive barnstars that are too generic. --Diyiy (talk) 04:00, 29 December 2018 (UTC)
Communities never rely on one means of communication but rather have countless pathways for moderating the behavior of others. Barn-stars are one of a panoply of communication devices. Do barn-stars correlate to other encouraging or discouraging aspects of communication such as Talk page banter and edit summaries? Bus stop (talk) 04:36, 29 December 2018 (UTC)
Diyiy, can you reply, please, to the part of SarahSV's question where she asks "in whose interests it's being done?" For my part, I want to know why Carnegie Mellon wants to know about Wikipedian behaviour. What benefits accrue to the university? And is the experiment to be of benefit to any of the great mainpulators of public behaviour such as Facebook, Google, Twitter, or anyone who desires to sharpen their sophisticated tools even further? Does the university have corporate, government, academic, or other partners who seek to benefit from barnstar-motivation studies? Are you, yourself, a ripe candidate for recruitment by Facebook or similar, based on your current social experiment activity, or arising out of your Facebook fellowship? I am seeking full transparency about any hidden partners or researcher motivations. Cui bono? Thank you. — O'Dea (talk) 19:11, 1 January 2019 (UTC)
Someone referenced the "natural state" and I think this is a significant concept. The proposal is to tamper with the "natural state". The knowledge of newly coined tokens of recognition will be known to some and not to others, at which point we will have the "natural state" for those in the dark and the "enhanced state" for those knowledgeable of the proposed project. This state of affairs would be analogous to some of us using the standard English alphabet of 26 letters and another group using an English alphabet consisting of 30 letters due to the addition of 4 bogus letters. I don't think we should tamper with our communications system. Bus stop (talk) 00:53, 29 December 2018 (UTC)
They're not paid to edit, they're paid to research. There is a difference, and this is not forbidden by WP:PAID or anything. Headbomb {t · c · p · b} 02:05, 29 December 2018 (UTC)
They plan to edit Wikipedia by giving hundreds of editors barnstars. There was a situation recently of an editor who was paid by researchers to close RfCs (or similar; I forget the details). There was a community discussion, opened by that editor as I recall, and it was agreed that it would be subject to WP:PAID, which entailed disclosure. SarahSV (talk) 02:31, 29 December 2018 (UTC)
They are paid to perform edits on Wikipedia, whether you call it research is irrelevant if they are actively spamming the community they will get blocked. The WMF has a team specifically employed for Community development any such active experiment that will impact the retention of the Community should have WMF support before proceeding. The whole experiment should be open with full details of what is taking place and declaration of funding sources and beneficiaries, every account sending Barnstars should clearly state they are participants. The outcomes should be freely published and accessible to the community, ethically everyone should be able to opt out of being spammed and having their contributions monitored. Gnangarra 05:12, 29 December 2018 (UTC)

Incompatible statements relating to Facebook

@Robertekraut wrote above at 22:13, 28 December[3]: Facebook won't benefit at all from the research we've been describing.

@Diyiy wrote above at 02:15, 29 December[4]: the proposed work has nothing to do with Facebook.

However, https://research.fb.com/programs/fellowship/#Eligibility_Criteria says very clearly: "Applications Must Include: 250-word research summary which clearly identifies the area of focus, importance to the field and applicability to Facebook of the anticipated research during the award (reference the research areas below)". (underling added by BHG)

(The acceptable topics are set out further down that page at https://research.fb.com/programs/fellowship/#Research_Areas)

Those statements by Diyiy and Robertekraut relating to Facebook appear to me be incompatible with the terms of the Fellowship.

So it seems to me that that:

  1. Diyiy explicitly disclosed to Facebook that this area of research would have no benefit to Facebook, or
  2. The proposed research is outside the scope of the Facebook Fellowship which funds Diyiy's research, or
  3. Diyiy and Robertekraut's statements above are false, or
  4. Diyiy's work on this particular project from funded by some other source which has not been disclosed

There may of course be some other explanation for this apparent incompatibility. But whatever the explanation is, I think it is important that Diyiy and Robertekraut promptly clarify exactly what is going on here. --BrownHairedGirl (talk) • (contribs) 16:20, 31 December 2018 (UTC)

Diyiy and I should have been more precise when saying ‘the proposed work has nothing to do with Facebook’ and ‘Facebook won't benefit at all from the research we've been describing’. We should have said that Facebook does not benefit directly from our research and does not benefit more from this knowledge than do other online platforms. We started this research on the influence of social roles in Wikipedia in collaboration with the WMF and our first paper[1] on the topic was published in 2016 before Diyiy received a Facebook fellowship . The proposed research should lead to generalizable knowledge about the consequences of bestowing recognition and the influence of social roles in online groups. This generalizable knowledge could be useful to many different types of online groups, including Wikipedia, open-source software development communities, online health support groups, peer-to-peer lending groups and many others, including Facebook's online groups.  Robertekraut (talk) 18:30, 2 January 2019 (UTC)

References

  1. ^ Yang, D., Halfaker, A., Kraut, R., & Hovy, E. (2016). Who Did What: Editor Role Identification in Wikipedia. Proceedings ICWSM: International AAAI Conference on Weblogs and Social Media (pp. 446-455). Palo Alto, CA: Association for the Advancement of Artificial Intelligence. pdf: https://www.aaai.org/ocs/index.php/ICWSM/ICWSM16/paper/view/13077/12764

Opt-out

Diyiy andRobertekraut, what is the opt-out protocol for this research project? For example, is there a template users can place on their user page to exclude themselves from your research? Vexations (talk) 17:34, 29 December 2018 (UTC)

User:Vexations. In search of a more permanent solution, i've just created a category called Category:NOTLABRAT. Whether it is an effective deterrent to being involved in research or not lies in the honesty of researchers. I'm hoping to highlight this in a brief sentence in WP:NOTLAB. Cesdeva (talk) 00:46, 30 December 2018 (UTC)
I've highlighted the category with this edit to WP:NOTLAB. The edit wasn't the result of a consensus to make the edit, but it does reflect consensus among some editors (those listed) that they would like to be left out of research. Cesdeva (talk) 19:09, 31 December 2018 (UTC)

Yes to most research. But no thanks to this

This was mentioned on my talk page, so I came here with an open mind to it check out. All our contributions to Wikipedia are publicly visible, and much research has been done on them. Some of it is very valuable. In principle, more research is fine.

But having read the discussion, I am strongly opposed.

The good side is that the researchers came here to set out their proposal, and have engaged in a collegial manner with those who commented. Good marks on that score.

Points lost for the obvious folly of claiming that an editor with <500 contributions in near 12 years is a senior contributor. Others have explained why that is not so, but it raised alarm bells with me as an indication of a very poor understanding of the editorial community. On the day this proposal was posted, the editor ranked at slot number 10,000 in the list of editors by edit count had made 8201 edits. The numbers are no criticism of the editor with <500 contibs, whose efforts are evidently made in ways which don't accumulate a high edit count ... but this sort of data should have been well-known to the designers of the research project. It is a flashing red light that the researchers have done seriously inadequate homework about the nature of the community they are researching.

Homework aside, it's clear that the substance of this project is to monitor the effects of a program of systematic deception.

The project is misleadingly labelled as research, a broad spectrum term which includes observational and analytical processes, as I mistakenly assumed this to be. In reality, what is proposed is a social experiment: to measure how people behave when manipulated as part of an RCT. More points lost for not explicitly labelling the project as a social experiment.

This deception experiment is destructive, because Wikipedia is a trust community. It is composed overwhelming of volunteers who contribute without possibility of tangible reward, and mostly anonymously. We don't hear voices or see faces, let alone any of the social aspects of community of most collaborations. All we have is letters on a screen, and trust which is built up slowly on an assumption of good faith. We react very strongly against editors who deceive the community, whether by socking or by faking references or claiming false credentials.

So when someone gives a thanks or barnstar or a welcome or a friendly note, we assume that is out of some sort of good intent. As others noted above, there are complex permutations around good intent, but the core of it is that some human is trying to make a nice gesture. The community has repeatedly rejected proposals for a "welcome bot" — so much so that it is listed both as a perennial proposal and as a frequently denied bot. The reason is simple: human gestures are valueless unless genuine, and fake ones can be deeply corrosive.

This is a proposal to target hundreds editors with fake acknowledgements. Sure, they will be delivered by a human, but the decision will be made by some sort of bot. Every editor delivering such a message will be deceiving the recipient into believing that the decision is their own.

@Robertekraut defended this by repeating that the barnstars will be given to editors who deserve them but haven't yet received any. I think that misses the whole point of barnstars, Most editors know that >99% of the good stuff they do en.wp goes unpraised; the significance of the barnstar is not in the work done, but in the fact that a human being has been watching and decided of their own free will to extend praise.

Such deceit is highly corrosive. It doesn't just sour the relationship betweeen the two individuals involved; it sours the recipient's undestanding of the whole community. One fake reward makes all rewards suspect, and even the news that this is being planned will make editors more wary.

I am saddened to hear that CMU's IRB waived informed consent. It seems to be adhering more closely to the minimal requirements of the US's National Research Act than to the much higher principles of the Nuremberg Code. The deceit and manipulation involved here is miniscule compared to the deceit and manipulation involved in all the commercial, political and military uses of manipulative psychology to which universities turn a blind eye; but it is on that dark side of the spectrum.

I hope that Facebook's sponsorship of @User:Diyiy's work is at sufficient distance that there is no possibility of direction from the company. https://research.fb.com/programs/fellowship/ is vague about selection personnel and criteria, so I will AGF. But given the extent of Facebook's manipulation of its users, its spectre hovers overs this project, even if its malign influence is only as an indirect signal to the educational culture that its bread is buttered on a particular side.

Wikipedia has a lot in common with how the internet was in its pre-commercialisation days in the early 1990s: anonymous, quite anarchic, and driven by altruism rather than by money. This proposed social experiment reeks of the manipulation of users and and monetisation-of-retention which underpins the giants of the commercialised web.

Many thanks to Diyiy and Robert for their courtesy and civility in this discussion. But FWIW, my view is that this proposal doesn't belong on en.wp ... and that if it proceeds, sanctions should be applied to any editor who assists it without full transparency. --BrownHairedGirl (talk) • (contribs) 15:39, 30 December 2018 (UTC)

PS In the discussion on meta, I have just posted[5] a request to @Diyiy to a) explicitly relabel her proposal as a social experiment, and b) to prominently disclose her personal sponsorship by Facebook.
I was shocked to find that the word "Facebook" is mentioned nowhere in m:Research:How role-specific rewards influence Wikipedia editors’ contribution. My ability to sustain an assumption of the good faith of both @Diyiy and @Robertekraut is fading fast, and I think that they both owe the en.wp community a fulsome explanation for this extraordinary lack of transparency.
(Personally, I also think that CMU should also be concerned about the lack of transparency, but that discusison belongs elsewhere.) --BrownHairedGirl (talk) • (contribs) 16:54, 30 December 2018 (UTC)
BrownHairedGirl, thank you for your post. I agree with your view of this. Diyiy, again, thank you for taking the time to explain.
I see the Signpost wrote about the 2011/2012 study: Restivo, Michael and van de Rijt, Arnout (2012). "Experimental Study of Informal Rewards in Peer Production". PLoS ONE, 7(3): e34358. doi:10.1371/journal.pone.0034358 The Signpost explained:

During the experiment, it was noted on the Administrator's noticeboard/Incidents page that a seemingly random IP editor was "handing out barnstars", which led to some suspicion from Wikipedians. The thread was closed after User:Mike Restivo confirmed he accidentally logged out when delivering the barnstars. He did not, however, declare his status as a researcher, and the group's paper does not disclose that the behavior was considered unusual enough to warrant such a discussion thread.

See Mike Restivo's contribs for 21 February to 29 April 2011, when he handed them out. He posted this note on his user page on 17 February 2011 as an explanation of what he was doing, with no mention of a study. On 29 April he added a sentence about random selection, but still no mention of a study. Boing! said Zebedee opened an AN/I on 29 April in which several editors expressed concern. Others expressed concern on Mike's talk page, where he told them in an edit summary that he was handing out the barnstars "randomly", which wasn't the whole story. The paper said they had targeted the top 1% of users, minus those who had received a barnstar or were admins. "We then took a uniformly random sample of 200 users and through random assignment either awarded a barnstar (100 cases) or withheld the award in the control group (100 cases). Finally, we observed all 200 subjects’ actions for 90 days." Several recipients thanked him for the barnstars on his user page and on his talk page.
As Brownhairedgirl said, this subverts and undermines Wikipedia:Barnstars, which was started after the original barnstar was created in (I believe) 2003. These are genuine awards, given in good faith and often after a lot of thought about which one to choose and how to word it. A 15-year history of people being thoughtful and kind toward each other in what can be an aggressive environment is something to cherish and respect. That's just one objection among several others, including that the researchers don't seem to know what kind of editor they're targeting; that an unknown proportion of those editors may be addicted; that several may be under the age of consent; and that this kind of research (how to tempt people into behavioural loops that may not be in their interests) is something that Wikipedia, and I hope the Wikimedia Foundation, should distance itself from. SarahSV (talk) 04:10, 31 December 2018 (UTC)
Thanks, @SarahSV. I think you are right to draw such attention to the issues of addiction. Facebook's business model appears to encourage addiction, but Wikipedia should not treat its editors in the same way.
That is why I am so concerned that the @Diyiy's Facebook Fellowship was not disclosed either on her page on meta about this research or in her initial post on this page.[6]. It was only revealed to this discussion thanks to the diligence of @Vexations, who raised it in a post[7] ten days after @Diyiy's initial post.
My own further checking (see above: #Incompatible statements relating to Facebook) has given me further concerns about the candour and transparency of Diyiy and @Robertekraut. It seems to me that whenever more checking is done by en.wp editors, the story told by Diyiy and Robertekraut's story unravels a little more.
I understand that they may both be enjoying a break over the holiday season, but clarification is urgently needed on their return.
In the meantime, I think that Wikipedia's policies in this area need to be tightened up. The issues which I see so far include:
  1. Clarifying the policy on WP:COI/WP:MEAT to explicitly forbid edits on behalf of a third party unless that external relationship is disclosed
  2. Tightening policy to require that academics explicit disclose the source and terms of their funding.
  3. Developing a policy on addiction. So far, all I see is a humorous essay at WP:Wikipediholic
I don't think that we need to wait for a consensus on this particular proposal to address the policy issues which it has revealed. --BrownHairedGirl (talk) • (contribs) 16:47, 31 December 2018 (UTC)
BrownHairedGirl, I don't know how to interpret that WP:ADDICTED hosts a humour page, but we definitely need a serious page about it. We should also start a discussion once the holidays are over about banning this type of research and/or setting up an ethics committee of our own. SarahSV (talk) 17:04, 31 December 2018 (UTC)
@SarahSV. Yes to both. I particularly like the idea of our own ethics committee, and hope that it is an area where WMF support would be available. --BrownHairedGirl (talk) • (contribs) 17:08, 31 December 2018 (UTC)
There has been previous discussion regarding setting up an internal ethics review in the form of a WP:IRB. See discussions that established NOTLAB as policy at Wikipedia talk:What Wikipedia is not/Archive 55. See also current discussion at Wikipedia:Village pump (idea lab)#IRB Review of Research on Wikipedia to be submitted to OTRS. There's a few issues surrounding that:
  1. We don't want unethical research conducted (obviously)
  2. We need an independent verification of IRB approval
  3. Although external IRBs may have a good understanding of scholarly research, they may have little or no understanding of Wikipedia
  4. We don't want to discourage ethical research that could help us improve the project
  5. We don't want to make following the rules so cumbersome that researchers forgo any pretense of doing so and just do what they want regardless (AKA the classic COI conundrum)
  6. We should have proposals evaluated by editors who have some experience actually working with ethical evaluations of research involving human subjects. AKA, we don't want legitimate ethical research derailed because Wikipedia:Randy in Boise decided to turn community oversight into an ANI thread.
I've always been in favor of such a proposal. Whether it's logistically feasible and sustainable is a different issue though. GMGtalk 17:15, 31 December 2018 (UTC)
Thanks, @GMG. A few thoughts on that:
  • Yes, obvs, to points 1–3
  • You are almost certainly right about IRBs having little or no understanding of Wikipedia. This is an area where the WMF could help, by a program of systematically contacting IRBs (and whatever collective organisations they use) to explain the policies and ethical issues relating to Wikipedia.
  • I would want to qualify or tighten yoir statement that We don't want to discourage ethical research that could help us improve the project. I get a whiff from this discussion that some WMF staff saw the potential gains for en.wp from this research project, but did little scrutiny of the downsides. I suspect that most research proposals will have some positive potential, and that the main issue will be weighing the potential gains against against the possibilities for harm. In other words, this is a an area replete with trade-offs, and we need to be clear about which issues are bright lines, and for other issues how the trade-offs are weighed. For example, I can imagine that some breaching experiments might be considered if carefully constructed.
  • Yes to simplicity and clarity. That's commonly a failure of en.wp, where most policies and guidelines have grown far too verbose and unclear. There are some exceptions, such as WP:NOT, but in general instruction-creep and obstruction-through-verbosity are structural consequences of our WP:BOLD policy.
  • Yes to an ethics committee of people with verifiable expertise. However, no matter what structures we set up WP:Randy in Boise or anyone else can launch an ANI thread whenever they like on whatever they like. (Hence the perennial question about ANI: does it drives its participants insane? Or merely attract those already insane?) What matters is that the ethics committee attracts enough community support to ensure that such threads are shut down quickly, as is mostly the case with Randy's zomg-arbcom-is-evil threads.
However, I am troubled by the apparent demise of m:Research:Committee, and the lack of explanation for its dormancy. Whatever happened to it should inform this work on en.wp. @Halfak (WMF): is there any writeup of what happened? --BrownHairedGirl (talk) • (contribs) 22:08, 31 December 2018 (UTC)
I don't think proactively contacting IRBs is at all feasible. That's...I dunno...tens of thousands of IRBs globally, all of whom are under no obligation to listen, and many of whom will have no interest at all. As to this proposal, I'm broadly in favor of it, if done cautiously so as to not be disruptive. I'm of the opinion that we should encourage community engagement and mutual recognition, and that most of us don't do it enough. I think the ethical risks are minimal and that similar experiments with other online communities would also easily get an informed consent waiver based on risk to human subjects, and would be carried on without any input from the community at all. I think we need to carefully consider that the "advise and consent" role of the community here is done with the goal of minimizing/eliminating any potential disruption, with a keen eye that if we are too...verbose maybe...in that role, we effectively push research underground and maximize the disruption caused by uninformed research. I honestly think the online addiction thing is mostly a red herring, and that while it is perhaps an issue that we should have a response to as a community, it's not directly relevant to this proposal, and if followed to its logical conclusions, calls into question all types of community efforts, like workshops and editathons, all designed to increase community participation and longevity. GMGtalk 22:27, 31 December 2018 (UTC)
Thanks for writing this, @BrownHairedGirl:, I felt a general unease reading this proposal and your edit captures exactly why. Wikipedia editors have been handing out barnstars for 15 years now. Over this time I think one could gather enough data to form and show correlations re: whether getting more barnstars leads to more or fewer future edits, or whichever other hypotheses we can form, and let us know about their conclusions. If falsifiable results are needed, one could simply construe these conclusions as hypotheses and see if they pan out on 1-2 years of future data. By getting involved in giving out barnstars the experiment would obviously go quicker, but also create problems -- making this a blinded experiment in our environment would be tough job even for someone with a lot more inside knowledge than the people proposing this. We'd have to get very good at deceiving our peers, or we'd have a study of how Wikipedia is impacted by thinly disguised gamification ...unless these are precisely the aims of the experiment.
As tempted as I am to leave on this bombshell, I have to add that Diyiy's and Robertekraut's comments don't evoke someone with such sinister plans, rather people who still have a rather shallow awareness of WP's functioning and can't foresee these consequences. Still, even the more useful of these two by-product goals would have a limited use to Wikipedia, as senior editor retention is not our key problem. We've got thousands, probably tens of thousands people making thousands of regular edits on Wikipedia over 5-10 years or more, something few online communities can boast of. People who stay on Wikipedia for so long do so or quit due to their own internal motivation, and even genuine peer approval (let alone a calculated reward they can as senior editors see through the fastest) is not going to be a likely factor in it. This is a genuinely stressful hobby to devote time to; people who come here don't stay just for tea and cookies. IMO it would be better for us to focus such efforts into new user retention, e.g. to use gamification as a training wheel for new editors' motivation until they garner enough interest to withstand the rough ride of editing Wikipedia on their own. DaßWölf 20:29, 2 January 2019 (UTC)

Manipulated Wikipedian behaviour cannot produce neutral science

The Barnstar system whereby editors spontaneously award the stars will be corrupted and distorted by turning the system into an experiment. It will deliberately manipulate formerly innocent and well-meaning Wikipedian behaviour, and turn it into a reward-seeking, unspontaneous activity. The purpose of the encyclopaedia is not to manipulate contributors, especially without their knowledge and consent, which is unethical "science".

A researcher should properly study the system as it is, as it operates now. Once the researcher becomes, instead, an experimenter — and she openly admits how her research actively seeks to change Wikipedian behaviour — she interferes with the system she proposes to study. This is not merely bad science — it is not science at all.

There should be no burden placed on Wikipedians, generous with their time, to opt out of manipulative experimentation. The default should be that they may be invited to opt in to being manipulated. And bear in mind, that an experimental subject, conscious of being observed, is liable to behave unnaturally, is liable to have his activity distorted by self awareness, which is therefore a distorted behaviour, and therefore presents activity for study which is no longer authentic. Thus, the behaviour to be studied is no longer valid, "clean" experimental data, and "scientific" conclusions drawn from it would be unreliable, to put it mildy. — O'Dea (talk) 18:12, 1 January 2019 (UTC)

...As someone who only has mediocre experience in research design, I'm going to field a guess and say that this comment is not at all informed by experience in research design. It does however reinforce my intuition that crowd sourced ethics evaluation is not feasible, and we should instead institute some type of formalized review process. GMGtalk 18:34, 1 January 2019 (UTC)
GreenMeansGo, are you proposing to abandon consensus-based decision-making in favor of appointing a panel of experts? Vexations (talk) 18:44, 1 January 2019 (UTC)
Specifically, I'm proposing we build a community based consensus to appoint a panel of experts. Alternatively, if we cannot do so, then we should consider scrapping the process of community review all together, or just keep it on meta off in a quiet corner. When a discussion like this degrades into broad musings on the philosophy of science, and not specific methodological critique, it is no longer productive in any meaningful way. GMGtalk 19:18, 1 January 2019 (UTC)
I think the 'broad musings' were just a long-winded way of pointing out that the researchers risk becoming confounding variables in their own experiment, and that perhaps passive observation would be better. Discussions like these are important and shouldn't be had in quiet corners. Cesdeva (talk) 19:30, 1 January 2019 (UTC)
GreenMeansGo, allow me to take a leaf from your book and "field a guess," as you say, that what sways you most is not the argument made, but any "barnstars" achieved in life by the proponent of the argument. Your ad hominem brushing aside of so-called "crowd sourced ethics evaluation" is not at all an honest, rigorous engagement. Anyone who pronounces from such a lofty self-erected perch may be incapable of anything other than mediocre experience in research design (your description). Fielding "guesses" and tossing out lazy ad hominems does not accord with the spirit of Wikipedian civility, not to mention respectable discussion. — O'Dea (talk) 19:36, 1 January 2019 (UTC)
Civility is not synonymous with entertaining nonsense for it's own sake. And it is not an ad hominem to say that your argument makes no sense and therefore I conclude that you have no idea what you're talking about and no experience in the areas on which you have expressed an opinion. GMGtalk 21:53, 1 January 2019 (UTC)
GreenMeansGo, you are guilty, once again, of incivility—you merely doubled down on it—by insulting a reasoned argument as nonsense without justification, and by speculating without foundation about an editor's experience (ad hominem attack). You merely dismissed an argument based on your pure fantasy about qualification or lack thereof, as opposed to the merit of the argument. That is incivility, as well as feeble practice. — O'Dea (talk) 22:09, 18 January 2019 (UTC)

Barnstars are of no real-world value, and in practice are awarded almost randomly. They are also open to counterfeiting, since anyone can make a "barnstars" page (or section on their user page), and put any barnstars they want on it. No one goes around checking to see if the barnstars added to such pages are legitimate. Probably no one looks at such pages at all. The idea that they influence editing behavior is therefore absurd on its face. bd2412 T 23:15, 1 January 2019 (UTC)

I would just like to echo GMG a bit -- what we should be concerned with is not whether an experiment will produce scientifically viable results or whether it asks a silly question. Rather, we need to ask whether it comports with Wikipedia policies and the community's ethical consensus. To put it succinctly, science is their area; Wikipedia is ours. That said, a very happy New Year to all. Dumuzid (talk) 23:22, 1 January 2019 (UTC)

  • I'm going to take a different tack here. I believe that granting barnstars - traditionally given by one knowledgeable Wikipedian to another for a *significant* accomplishment - is very much the wrong thing to do. Maybe adding a "Wikilove" would be okay, and I wouldn't particularly have a problem with using the "thanks" feature. But giving the same copy-editing barnstar I earned for actually copy-editing an article on its way to featured status to someone who only fixed a few typos is, frankly, devaluing the hours of work I put in that resulted in *my* barnstar. Every single barnstar I have came as the result of significant effort on my part. I don't understand why the researchers have decided to grant what is, essentially, one of the highest interpersonal symbols of respect on the project to people who have not made the level of contribution that the rest of the community would expect to see when a barnstar is granted. It's like throwing a parade in recognition of successfully empyting the trash baskets, very disproportionate. If you're going to do something along the line of recognizing new users for their early contributions, please do so in a proportionate way that doesn't denigrate the barnstar-level recognition that Wikipedians give to Wikipedians. My personal belief is that it is insulting to the rest of us who *have* put in the hours and *have* worked hard for the recognition we've earned - and would devalue the giving of barnstars to the point that they would become meaningless in the future. Don't take this away from us. Use something else. Write a talk page post. Use the "thanks" feature. If you really, really have to, use Wikilove. But leave barnstars for serious, experienced Wikipedians to recognize other serious, experienced Wikipedians. Risker (talk) 23:53, 1 January 2019 (UTC)
  • My concern is with the authenticity of barn-stars. Perhaps they are not 100% authentic but we should not take steps to degrade authenticity, which is to say that we should not do anything with barn-stars. The concept of "counterfeiting", mentioned by BD2412, is interesting because in my opinion the proposal is akin to counterfeiting barn-stars. I understand that BD2412 was speaking of Users counterfeiting their own barn-stars but I think the concept of counterfeiting is equally applicable to the proposal. Would any harm come from instituting the proposed experiment? Probably not. I don't mean to be narrow-minded but I don't think any good could come from it either. My concern is akin to saving species threatened by extinction. Are they of any super-well-known super-importance to mankind? Probably not. But "preservation" is a value that is intrinsic. Unless the costs of preservation are too high we should err on the side of preservation. In other words—leave our barn-stars alone. But I am one admittedly ignorant voice. If the proposers of the experiment have something important to say in support of their proposal I think we should all listen. Bus stop (talk) 00:26, 2 January 2019 (UTC)

Victim protection

How can the experiment protect the victims of testing on them, firstly in the active stage when encourages addictive outcomes, dirves peope away, or devalues the efforts of others. What happens if it draws attention on an editor in a not so friendly environment who wants to remain under the radar, we have had editors in prisoned and those who have died as result being an editor. What about in the future when a editor puts their hand for RfA and some points out that the small recognition they got had been part of an experiment and wasnt any indication that the community appreciated their contributions. What about the damage to Wikipedia as an independent reliable source of information when its editors are being manipulated for scientific research. Gnangarra 06:19, 2 January 2019 (UTC)

Withdrawing research proposal

Even though we think that our research would answer questions about the impact of recognition that are both scientifically interesting and useful to Wikipedia and that the research could be conducted in an ethically responsible way, we’ve decided to withdraw this research project.

The initial discussion we had with the community was valuable and constructive. It evaluated the value of the research (mostly positive), raised ethical concerns (e.g., concerns about devaluing Barnstars) and suggested ways to improve the research, by mitigating the ethical concerns (e.g., using Thanks instead of Barnstars as recognition) and by increasing the likelihood that the research would lead to meaningful results (e.g., by having multiple experienced editors issue the recognition).

However, the most recent discussions suggesting that the receipt of barnstars causes Wikipedia addiction and our research is corrupt because Diyiy has a Facebook fellowship were not very productive. Despite our best intentions for scientific research on the effects of barnstars and recognition, we are sorry to see that some of the comments above contained accusation of us lying either to the Wikipedia community or to Facebook, which we consider ad hominem attacks. The most recent discussions were taking too much time, exacting too high an emotional toll and not going anywhere. Taking into account everyone’s opinions and the time we have available for this research, Diyiy and I have decided that we’ll withdraw our research proposal and discontinue this proposed project. We thank all of you for your active participation and discussion about this research proposal. Robertekraut (talk) 18:45, 2 January 2019 (UTC)

I'm sorry that you were subjected to this, Diyiy and Robertekraut. And I share GreenMeansGo's expressed concern that being too verbose/prescriptive/bureaucratic about research review will simply drive more research underground. This discussion is something of a case in point. Two researchers have followed best practices for performing ethical research on Wikipedia: they created a research proposal, they secured approval from Carnegie Mellon's Institutional Review board, they brought the proposal up for discussion at the Village Pump, and they actively engaged in conversation around community concerns and attempted to resolve those concerns. In response, they were subjected to accusations of bad faith, public shaming, and conspiracy-mongering. On top of that, some folks here have claimed—without evidence—that this study a) is not research (I don't even know what to do with a statement like that) and b) that the researchers clearly did not understand anything about Wikipedia because either "barnstars don't work" (they do, sometimes; social motivation is complex, more research is needed) AND/OR "500 edits doesn't make you an experienced Wikipedian" (it does, in fact, make you an experienced Wikipedian, or at least it is as good a threshold as any we've developed). Also, the notion that Robert Kraut doesn't know anything about Wikipedia is naive at best. He is one of the most cited scholars on online collaboration and Wikipedia in particular. He literally wrote the book on how open collaborations work, including a whole chapter on motivation that might be illuminating to some of the participants in this discussion.
Many people have engaged in this discussion in good faith. But—putting aside the question about whether or not this particular study was a good idea, or well-designed, or of scientific or practical value—the vigorous dogpiling that these researchers have received here will certainly cause other, less ethical, researchers to think twice before informing any Wikimedia volunteer community that they plan to conduct research that involves interacting with editors. A great deal of research, much of it substantially more risky and problematic than this study, has been conducted and published without any suggestion that the community as a whole was notified or involved. I've run into this pretty frequently when I'm reviewing recently-published literature on Wikipedia. Some of it is happening right now. It's unsettling, but ultimately it probably cannot be prevented in any systematic way. The best we, as a community, can do is to work with people who are operating openly and in good faith, as Diyiy and Robertkraut are.
So, by all means write more policies and set up committees. But if this discussion is any indication, you shouldn't expect those efforts to have any appreciable effect on the behavior of the people who are putting individual editors, or the community as a whole, at risk. And you should expect that those policies will have a chilling effect on the kind of research that has both scientific value and value to Wikipedia.
Also, to clear up any confusion, the Wikimedia Foundation does not and has never had the authority to regulate research on Wikipedia unless that research is conducted by or in collaboration with Wikimedia Foundation Staff and/or involves access to non-public data. J-Mo 20:38, 2 January 2019 (UTC)
I don't know what the solution is, but if this is how these discussion are going to be conducted, it ain't this, this faithless mess half full of nothing but vacillation between accusation and inanity. The only thing we're going to buy for ourselves in this direction is a bunch of researchers that don't seek input from the community, and then proceed with designs that are both more disruptive and probably of lower quality if successful. GMGtalk 21:09, 2 January 2019 (UTC)
The discussion is reflective of the climate we are in. After Cambridge Analytica etc, people are more concerned than ever about being manipulated. Without the WMF or an en-wiki review board of experts checking these proposals, it has been left to editors at the Village Pump to ask the questions and be the one-and-only barrier. So is it really any surprise that the response to this proposal has been so 'verbose' and pointy?
You want quality discussion? Help form a panel of experts to analyse these proposals. Otherwise next time there will just be a repeat of the above. Not everyone is content with putting their head under the sand. Cesdeva (talk) 21:36, 2 January 2019 (UTC)
Cesdeva I already analyze these proposals, and work with researchers to refine them, and notify the community about them—as I have pointed out in my posts above. And EpochFail has probably logged more hours than any other individual helping to vet and shepherd research proposals. And GreenMeansGo has put forward a straightforward proposal to independently verify whether IRB approval has been secured, which sounds like a promising first step. So I'm not sure who, by implication, is content to have their "head in the sand" in your statement above. But you can rest assured that neither I, nor any "panel of experts" I formed, would be among their number. However, if this is the way the current coalition of the willing on EnWiki wants to talk about research ethics, I'm not in any hurry to recruit more researchers to either vet proposals or submit them. It doesn't sound like it would be a good use of any of our time. J-Mo 21:53, 2 January 2019 (UTC)
Jtmorgan I guess what confused me is why this research proposal (which has been torn apart) even made it to the VP. Why was the Facebook fellowship (which is clearly relevant) not immediately highlighted and disclosed to the community? It seems the WMF has dropped a bollock in regards to due diligence. Luckily editors at the VP are thorough. Cesdeva (talk) 22:20, 2 January 2019 (UTC)
As was said above, the Foundation does not have the authority to officially determine whether research is allowed on a local project. Also, the discussion is here because of the local policy we adopted last year requiring community notification for potentially controversial research that involves interacting with the community. This isn't the Foundation's problem; it's our problem. No one is sticking their head in the sand. We've been having this discussion in various forms for months now. GMGtalk 22:30, 2 January 2019 (UTC)
Just wanted to join in and say thanks to the researchers for the way they went about this. While I think some of the resistance encountered was off-base, for lack of a better term, some of it I understand on a gut level. Either way, I hope Wikipedia will be a continued field of fruitful study and that they're willing to come back and discuss new proposals in the future. Cheers, all. Dumuzid (talk) 21:21, 2 January 2019 (UTC)
@Jtmorgan, I am surprised to see your comment that Diyiy and Robertekraut and followed best practices for performing ethical research on Wikipedia. As I acknowledged elsewhere, they got some things very right, but some things badly wrong:
  • They did not disclose upfront their source of funding
    • When it was uncovered, their initial responses were somewhere between misleading and false
    • Robert's statement on withdrawal made no apology for the shortage of transparency and candour, and instead dismissed it as ad hominem
  • They seemed ethically untroubled by framing supportive communication between volunteer people as a goal-driven management tool
  • They were also unconcerned by the deceptions which they were going to ask editors to perform
Sorry, Jt, but I think that Wikipedians are entitled to ask for better. And I would hope that since you are also a WMF employee, you will be conveying back to other WMF personnel that that many aspects of this proposal were controversial. My post above #Yes_to_most_research._But_no_thanks_to_this got more way thanks notifications than I have received for any single edit I have ever made, so while I know that plenty have disagreed with me, I'm fairly confident that I am not taking some marginal and extremist view. --BrownHairedGirl (talk) • (contribs) 10:26, 4 January 2019 (UTC)
  • I am disappointed that the researchers did not address the concern about "diluting the value of the barnstar" as expressed by Noyster. This was also my primary concern. Bus stop (talk) 21:59, 2 January 2019 (UTC)
  • Thanks for listening to the community and withdrawing this research proposal. As someone who has both given and received quite a few barnstars in my time here, and who has also been involved in many discussions about research, I for one would be happy to see some research done on Wikipedia's barnstars. But rather than research based on an experiment I would like to see research done noninvasively, by measurement. Barnstars have bern around for more than a decade, it would be interesting to know what if any effect they have had on retention, civility and encouraging particular types of behaviour. It is no secret that various WikiProjects have created barnstars in order to promote either their project or the behaviour that their project endorses. So as well as looking at barnstars and their effect on editor retention, it should be possible to be more specific and measure things like the effect of getting the article rescue barnstar at encouraging recipients to rescue further articles. There may also be a halo effect, with other visitors to a particular person's talkpage seeing that they have received a barnstar for a particular action. This sort of digital archaeology requires a slightly different mindset than the experimental research ethos of collecting data through experiment. But as both archaeologists and astronomers can affirm, research by looking at actual data is every bit as much research as research by creating your own data, and without the pitfalls of devaluing the thing you measure or missing such a key datum as who the donor is. ϢereSpielChequers 06:07, 4 January 2019 (UTC)
  • Thanks to @Robertekraut & @Diyiy for withdrawing this proposal. It was clear that it would have been divisive. I appreciate your efforts to modify the proposal based on feedback, but I am unpersuaded that any of them addressed my core concern: recruitment of meatpuppets to act without undisclosed COI in their interactions with other editors. I remained disappointed both that such a proposal was cleared by an IRB and that WMF did not raise concerns about that core issue.
And I am also disappointed to see Robertekraut's defensiveness about the Facebook connection.
First, as a general principle the funding of any research project should be disclosed upfront to other stakeholders, who have a right to know who is paying for the piper. That is basic transparency, and the lack of it when approaching Wikipedia editors was a foolish omission, because this is a large group of people whose common purpose is research. Wikipedians are likely to find out, so better to disclose at the outset.
Secondly, it is a particularly serious omission when the funder is Facebook. The ethics of Facebook have been a major issue of public controversy in both the US and the UK for over two years, and recent discoveries by a select committee of the House of Commons have shown major ethical failings at the very top of Facebook. So it is unsurprising that a connection to Facebook would be controversial and scrutinised.
Thirdly, the initial statements which both Robertekraut and Diyiy gave in response to the discovery were at best incomplete. Robertekraut's subsequent clarification confirmed some of what I had inferred from the public info, and added the crucial context that this is part of a workstream begun before Diyiy revived the Facebook Fellowship. That reduces my concerns about the direct influence of Facebook, but reaffirms my concern that this is part the broader intellectual collusion of academia and business in the science of manipulation. There is plenty of scholarly work on that, but George Monbiot's article a few days ago is a handy layperson's introduction to the extent of the problem.
So I am left feeling disappointed that Robertekraut describes the questioning of him about this as ad hominem attacks. The questioning was solely a consequence of the researchers' choice to leave this info to be dragged out of them in multiple rounds. I stand by comment above that I have concerns about the candour and transparency of these researchers. I hope that in future, WMF will advise researchers to be much more candid in their initial approach, and to respond to any questions or suspicions with enthusisastic transparency rather than the defensiveness displayed here. --BrownHairedGirl (talk) • (contribs) 07:54, 4 January 2019 (UTC)

The order of professions in biography articles

Hello! Is it a common practice to organize the professions of a person in the introduction of a biography article according to relevancy? For example, in Jacqueline Kennedy's article, the intro says "Jacqueline Lee Kennedy Onassis was an American book editor and socialite who was First Lady of the United States during the presidency of her husband". Are there guideline for this order of professions (book editor then politician)? Or is it left to editors?--Reem Al-Kashif (talk) 15:59, 12 January 2019 (UTC)

Don't know if there's a guideline, but it seems common sense to order them in rough order of notability, and that is what is normally done. So the JO one just seems silly. She didn't become a book editor until quite late in life (about 46) anyway, and was probably not notable as that alone. Fashion icon may belong in that sentence. Johnbod (talk) 16:25, 12 January 2019 (UTC)
Thank you for your reply. That's indeed why I am getting confused. And she only worked as a book editor for two years or so.--Reem Al-Kashif (talk) 16:28, 12 January 2019 (UTC)
No, that was just the first job - she seems to have done it until her death, although I somehow doubt she put in a 40-hour week all those years. Johnbod (talk) 16:30, 12 January 2019 (UTC)
Oh, I see. Generally speaking, it seems like whenever a person is taking up more than one job, which is the case for many of the people for whom we have long articles, there is confusion on which profession should be put first. This is also reflected in the categorization of the articles. If one politician performs a minor role in a movie, they seem to be eligible to be in the "actors" category.--Reem Al-Kashif (talk) 16:36, 12 January 2019 (UTC)
I think a minor acting role qualifies an already-notable person for inclusion in a Category for actors. If there is disagreement this should be resolved on the article Talk page. I would concede that it is possible that the person should be omitted from the acting Category but I don't think there should be a rule or even a guideline about this. Involved editors should weigh in on the article's Talk page. That constitutes the all-important consensus. Same thing for listing professions in the lede. There isn't a right or wrong, in my opinion. Bear in mind that consensus can change, but probably not a "recent consensus".

Looking at the sentence I think it is fine: "Jacqueline Lee Kennedy Onassis (née Bouvier /ˈbuːvieɪ/; July 28, 1929 – May 19, 1994) was an American book editor and socialite who was First Lady of the United States during the presidency of her husband, John F. Kennedy, from January 1961 until his assassination in November 1963." I would say the internal link highlights the role of First Lady of the United States, not to mention the name John F. Kennedy. I also think the construction of the sentence benefits from getting the minor role of "book editor" out of the way before addressing the more complicated part of the sentence. Bus stop (talk) 20:39, 12 January 2019 (UTC)

Image identification

Donald Trump visiting Suresnes cemetery.jpg
Hello english team,
I don't know if I ask on the good village.
I need some help identifying the third character on the left after Donal Trump and William M. Matz. I read somewhere it is Superintendent Keith Stadler. But I'm not sure, due to his Linkedin Profile. Someone can help me please ? Thanks a lot. --Gpesenti (talk) 23:46, 12 January 2019 (UTC)
I don't know the answer to that question, Gpesenti, but I think Reference desk/Humanities would be a good place to ask that question. Bus stop (talk) 00:36, 13 January 2019 (UTC)
Ok thank you Bus stop, I proceed. --Gpesenti (talk) 01:42, 13 January 2019 (UTC)

Notification of BAG nomination

I am just writing this to inform Village pump (miscellaneous) that I have requested to join the Bot Approvals Group (BAG). I invite your thoughts on the nomination subpage, which is located here. Thank you for your time. --TheSandDoctor Talk 05:41, 14 January 2019 (UTC)

Link to Special:RandomInCategory for a specific category

Is there a way to encode a wikilink so that it calls Special:RandomInCategory with a specific category? The Help:Special page entry for that call doesn't say, and the form at the link looks like it is using the Post method. Praemonitus (talk) 22:58, 15 January 2019 (UTC)

Praemonitus, put the category name (no namespace) as a "subpage": Special:RandomInCategory/Category theory will give you random articles in Category:Category theory, for example. This works with most special pages with arguments: Special:Random/Talk, Special:ListUsers/foo (but Special:ListUsers/sysop), and so on. Enterprisey (talk!) 00:16, 16 January 2019 (UTC)
Enterprisey: Thanks! Unfortunately it doesn't appear to descend the category tree, so it's not that useful. Praemonitus (talk) 02:09, 16 January 2019 (UTC)

How to add reference about a guinness world record without an official blog post?

This question has been moved to WP:HELPDESK. User670839245 (talk) 07:52, 17 January 2019 (UTC)

Wiki Loves Africa 2019 around the corner : your help more than welcome

Hello everyone. In 2 weeks time, the latest (5th) edition of Wiki Loves Africa will launch on the theme Play for one month. Please check out the contest main page for details about the theme if you are directly interested and want to contribute pictures and join the fun online or locally.

But this message is also a pre-warning that hords of new and unexperienced users from Africa will join and upload pictures, that are often wonderful (but not really categorized or described) and sometimes terrible (blurred, irrelevant, promotional, copyrighted etc.). I would like to extend an invitation to you guys to help with image tracking and clean-up during the contest. If you plan to help heavily, please add your name to the team page so that we know where to go and who to ask if we have issues. But also, more generally, please help in keeping an eye on the general category where images will be uploaded (right now there is only a test picture from a painting from my daughter... but soon... we can expect probably around 10k images to look at). Your help will be tremendously and gratefully welcome.

Anthere (talk) 14:33, 17 January 2019 (UTC)

Language: Distinctions without differences

I launched a discussion in Wikipedia talk:Noticeboard for India-related topics#"Please use Indian English" asking for specific guidance on how to conform to the standards for Indian English, there being no guidelines for how to do so. I was dogged, some said annoying; some said trolling. I was given Trinidadian English as an analog. There are all of nine articles marked in their talk pages as being in Trinidadian English. Certainly the two-island state has a distinctive patois, but it's not appropriate for encyclopedia articles. Ask a Trini. I am willing to bet that British English would be the recommended standard for an encyclopedia article. My suggestion here is not a perennial request to standardize spelling, Wikipedia:Perennial proposals#Enforce American or British spelling, but a request to discuss why it is appropriate to make the distinction among twenty-one varieties of English when there are basically only two standards for expository English writing: with or without Oxford spelling, with or without the Oxford comma. Where numbers are concerned, there is already a standard: unless a number is part of a quotation, zeros should be grouped in threes and the decimal point is a full stop (period). Wherever this topic is discussed, an assertion is made that spelling may differ from both American English and British English and so may syntax. I haven't seen it. It just seems to me that this is a distinction without a difference. I am told that "Reality is more complex than that." It may be, but I am a simple person. I would like someone to explain to me how the entreaties to use one of twenty-one varieties of English without any instructions for how to do so are valuable to the encyclopedia. Rhadow (talk) 02:47, 18 January 2019 (UTC)

The problem is that there is no actionable proposal. What would conforming to the standards for Indian English actually mean? Please link to three articles and quote examples of inappropriate text along with the replacement text that would be recommended. My take on the problem is that WP:ENGVAR is fine for settling disagreements about spelling (color/colour, organize/organise) and date formats (mdy/dmy) but is not useful for a disagreement about text. I see text like "In the year 1998 such and such happened" where it would be standard at enwiki for the underlined "the year" to be deleted. More examples are needed to define what is proposed. A guideline that says "conform to Indian English" is useless without guidance about what that means. By the way, it is not a good idea to describe identifiable editors as dogged/annoying/trolling. Johnuniq (talk) 03:09, 18 January 2019 (UTC)
It looks like I misunderstood the OP. At the other discussion (link above) I somehow got the impression that a couple of people wanted "use X English" to mean more than "use date formats and spelling appropriate for X". It might be best to acknowledge that wikignomes expand templates and categories to fill all possibilities and I don't think anyone has ever explained what "use X English" means beyond what I mentioned, namely color/colour, -ize/-ise and mdy/dmy dates. My guess is that people don't feel comfortable putting {{Use British English}} on an article about an Indian topic, so {{Use Indian English}} was created. Johnuniq (talk) 09:24, 18 January 2019 (UTC)
Your example of numbers is in fact one of the distinguishing features of Indian English, where lakh (1,00,000) and crore (1,00,00,000) are used rather than million (1,000,000). I do agree that if these templates are used there should be some accompanying instructions, and that 21 different varieties seems like overkill. Phil Bridger (talk) 09:06, 18 January 2019 (UTC)

Draft proposal

Okay, then. I have a proposal that think would pass muster by SMcCandlish. Their analysis at Wikipedia talk:Identifying and using style guides#Style guides from around the anglosphere is a great start. The section MOS:ENGVAR should be expanded slightly to recognize all twenty-one dialects of English. A search for WP:Indian English takes you there in any case, implicitly suggesting that the English language tree has two trunks, after which the specific branch you choose is relatively insignificant. In that way, we would not offend the proponents of a tag for every regional dialect. I suggest that for every dialect we construct a short guide whose model sounds like this:

Trinidadian English is a dialect of English stemming originally from British English, enriched by native, Spanish, and French influences. In spoken form, it is a rich patois. For encyclopedia articles, formal language rules apply. In the absence of a published style guide as exists for American, Canadian and U.K. lects, a British style guide, for example Hart's Rules, is a reference for WP editors.[1] The nation uses the metric system, therefore metric units are preferred, with conversions to other units as appropriate. The spelling standard is Oxford Spelling (wp:EngvarB), although American spellings are common.

When twenty-one such paragraphs are published, it will become quite clear that the number is too high. In time then, the disused templates will become candidates for deletion. In my opinion, a gradual reduction in dialect templates is a better trend than the creation of a plethora. Any move to simplify the MOS and its templates in Wikipedia is a long-term plus.

A draft paragraph for Indian English follows. I searched for a style guide and did not find one, therefore the guidance is eerily similar to Trinidadian English.

Indian English is a dialect of English stemming originally from British English, enriched by native influences. In spoken form, it can vary substantially from its origin, including frequent use of the present continuous tense. For encyclopedia articles, formal language rules apply. In the absence of a published style guide as exists for American, Canadian and U.K. lects, a British style guide, for example Hart's Rules, is a reference for WP editors.[1] The spelling standard is Oxford Spelling (wp:EngvarB). The nation uses the metric system, therefore metric units are preferred, but imperial measures (e.g. acres and miles) are common and conversions should be provided. India uses a numbering system including the crore and lakh which require a nonstandard grouping of zeros in large numbers. When quoting or paraphrasing, these terms are fine, although an editor is entreated to convert or explain these numbers for readers unfamiliar with the units.

That's my two cents. Rhadow (talk) 13:32, 18 January 2019 (UTC)

References

  1. ^ a b https://www.ox.ac.uk/sites/files/oxford/media_wysiwyg/University%20of%20Oxford%20Style%20Guide.pdf

Continued discussion

This seems a lot like instructional WP:CREEP. There's established practice that exists at the respective WikiProjects already. Cesdeva (talk) 13:55, 18 January 2019 (UTC)

By 'this' i mean the draft proposal above. Cesdeva (talk) 15:59, 18 January 2019 (UTC)
  • Oppose WP:CREEP, also note that Wikipedia is built for the readers and standardisation must be avoided if it is detrimental to the readers. Regards. << FR 14:08, 18 January 2019 (UTC)
  • Comment I could not agree more with FR. Hanging one of twenty-one tags specifying the dialect of English complicates the work of an editor. That itself is WP:CREEP. As described in the MOS, two are sufficient. Rhadow (talk) 14:58, 18 January 2019 (UTC)
It doesn't, and it isn't creep. Your sarcasm and ignorance just highlights the shallowness of your 'grievance'. I doubt anyone will try to engage with you on this topic now, after seeing your comment which borders on trolling. Cesdeva (talk) 15:59, 18 January 2019 (UTC)
  • It is common courtesy to post the link to this discussion at the page where you first posted your musings and where, through a long series of patient corrections by others, you acquired the knowledge which you have so glibly posted above. Fowler&fowler«Talk» 15:24, 18 January 2019 (UTC)
  • Oppose The 42 templates of the 21 regional varieties of English (which include Irish-, Scottish-, Jamaican-, American-, South African-, Australian-, New Zealand-, British-, Singapore-, and many others) appear in at least 300,000 WP articles. That system has worked for at least 12 years. Why should I even read anything written by someone with little knowledge of the underlying issues, whose motivations, as exhibited in his posts seem to be based on a fixed idea that there are only two varieties of the English language, British- and American-? What are the chances of something like this receiving WP-wide approval? Why should I waste my time? See Radhow's earlier efforts at: Talk:Asa_Wright_Nature_Centre and Wikipedia_talk:Noticeboard_for_India-related_topics#"Please_use_Indian_English" Note, especially Guettarda's insightful remarks about WP:IDIDNTHEARTHAT. That alone raises the prospects of endless discussions here where people are talking past each other. The above exchange with FR is a good example. . Fowler&fowler«Talk» 15:55, 18 January 2019 (UTC)
  • Comment All of the criticisms are fair enough. It strikes me that there are three options with respect to language tags: (1) the status quo, twenty-one identified varieties, none of which have a clearly defined distinction for the purposes of article editing between, say, Jamaican English and Trinidadian English, leaving the editor in a no-guidance situation, (2) and not much different, to allow the number of {use xxx English} templates to grow, each supporting another small variation on the language (adding {use Barbadian English} even if it is a matter of national pride, for example), or (3) to limit the number of templates to those lects for which there is style guide and dictionary to which an editor can refer.
My question was genuine and proposal respectfully submitted. I am mystified by the number of negative responses, "ass," "little knowledge," "trolling," and "bumbling, random, musings." I just don't hear any other suggestions on how to improve the situation. Rhadow (talk) 16:56, 18 January 2019 (UTC)
Sadly, your history thus far has been one of ignoring in any conversation anything that is inconvenient for your theory. The half a dozen linguistics references on Indian English I posted, earlier, you dismissed by suggesting that the term register applies to only spoken language. See here, not to mention that four or five of those references were not about registers at all. (See also OED: register: Linguistics. In language: a variety or level of usage, esp. as determined by social context and characterized by the range of vocabulary, pronunciation, syntax, etc., used by a speaker or writer in particular circumstances.) Fowler&fowler«Talk» 17:37, 18 January 2019 (UTC)
The existence of a style guide is not a prerequisite for the existence of a style in any variety of English. Nor are style guides comprehensive. In written American English for example, "likely" is now used as an adverb in fairly formal settings (e.g. "According to the National Weather Service, the hurricane will likely make landfall in the vicinity of XYZ, Florida." I haven't checked, but most likely this is not mentioned in style guides; at least if didn't use to be. If a WP article says, "This article is written in American English," it doesn't mean that any contributor needs to look up a style guide and write in the manner of a native speaker of AmE. All it means is that certain lexical or syntactical or stylistic features are acceptable in AmE, which speakers of other Englishes will not commonly employ in their own speech or writing, though they will very likely understand them. Such features should be respected in such an article, as long as they are not wildly confusing to others.
It is important to note that there are higher level features of any English that lie beyond the pale of any style guide. Would you like Americans to alter the sentence patterns of any BrE speaker editing an AmE tagged article, even though nothing he has written violates the Chicago Manual of Style? I think you are misinterpreting what "This article uses Indian English" means. It doesn't mean that you will need to pick up a hypothetical Mumbai manual of style, and write Indian English in the manner of an Indian. It doesn't even mean that the patterns that might seem peculiar to you will necessarily be mentioned in that Indian style guide. Yet is is undeniable that there is such a thing as Indian English, that a Martin Amis cannot write in the style of a Salman Rushdie. A hundred years ago, the Fowler Brothers, in The King's English were bemoaning the use of American expressions introduced by Kipling (who had written his Jungle Books in Brattleboro, Vermont). In those days there weren't style any guides for American English. These days no BrE style guide will be so prescriptive. Fowler&fowler«Talk» 19:14, 18 January 2019 (UTC)
  • Oppose any change. The vast majority of Wikipedia editors will continue editing using English that is recognisable as English everywhere, but with the occasional mistake due to their unawareness of differences. Those who know and care about the different varieties will correct things where necessary. That's the process that has worked well for many years, so why change anything? If the proposer of this change can't distinguish between standard Indian English and errors then that's fine - just let someone else fix it. Phil Bridger (talk) 19:31, 18 January 2019 (UTC)

Closing discussions

Hi, I asked this question on the Wikipedia:Help desk, but did not get a reply. Not sure if it should go in the Policies section or not as it involves and unclear policy. My question: does a discussion on a Talk page have to be closed by an uninvolved third-party editor or can one of the editors involved in the discussion close it when it is clear that further discussion will be unproductive? It is not clear from Wikipedia:Closing discussions if it has to be an uninvolved editor. Also, if five editors are in agreement and one is not, does this count as a rough consensus WP:ROUGHCONSENSUS? I know consensus is not a head count, but I'm not sure if this is enough to assert a consensus. My apologies if this type of question is not appropriate for the Village pump. Thanks for your help. - Epinoia (talk) 17:45, 18 January 2019 (UTC)

There is some text in WP:RFCEND in this regard. WP:Consensus should probably have something similar. This question is fine here; the WP:Teahouse or WP:VPPOL can also be used. --Izno (talk) 18:58, 18 January 2019 (UTC)
Retrieved from "https://en.wikipedia.org/w/index.php?title=Wikipedia:Village_pump_(miscellaneous)&oldid=879314653"
This content was retrieved from Wikipedia : http://en.wikipedia.org/wiki/Wikipedia:Village_pump_(miscellaneous)
This page is based on the copyrighted Wikipedia article "Wikipedia:Village pump (miscellaneous)"; it is used under the Creative Commons Attribution-ShareAlike 3.0 Unported License (CC-BY-SA). You may redistribute it, verbatim or modified, providing that you comply with the terms of the CC-BY-SA