The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
There has been much scholarship involving Wikipedia,[1] but Wikipedia is not a public laboratory and editors are not lab rats. Research that analyzes content or editor behavior needs no permission, because the content of Wikipedia is open to everyone. Research that is interventional—that involves surveying editors or making changes in content—should be conducted only with prior consensus at relevant WikiProjects, the Village Pump, or at the administrators' noticeboard, and should be disclosed clearly on the editor's user page.[2] Researchers must disclose outside connections under the conflict of interest guideline. If their work on Wikipedia is conducted in the course of their employment, they must disclose this under the paid editing policy. Editing directed at any goal other than building an encyclopedia, or improving the experience of editors and readers, is likely to be seen by the community as disruptive and will likely entail time-wasting cleanup by our volunteers. Editors who disregard community norms regarding design, transparency, or conduct may face sanctions, including the loss of editing privileges.
support - provides useful guidance for people who want to do research, and for the community in interacting with folks who want to do research. Jytdog (talk) 02:19, 20 October 2017 (UTC)
Support, but only with an adjustment: Change "with prior consensus at relevant WikiProjects, the Village Pump, or at the administrators' noticeboard" to "with prior consensus at WP:Village pump (proposals)".
We have a site-wide noticeboard for proposals, and that one is it. Research/experimentation proposals are not magically except from WP:PROPOSAL process.
WP:AN is not such a venue, and isn't paid much attention by anyone but admins; it's a noticeboard for dealing with admin problems, but third parties using and possibly abusing WP and its editors as lab rats is not an admin issue, it's an editorial community issue. A handful of admins are not empowered to grant permission to use the editorial community for research purposes; that is beyond the remit of adminship. It's beyond the remit even of ArbCom.
Finally, wikiprojects have no authority to approve or reject anything; see WP:CONLEVEL policy – they are just some random editors with a shared topical interest, and do not speak for anyone but those editors, including with regard to topics they feel are within their scope. It is not possible for a wikiproject about football or frogs to give permission to some university to do experiments involving editors of football or frog articles, since any editor may edit those topics at any time, and this has nothing to do with wikiproject "membership".
I raised these concerns with the drafters of this proposal, only to be ignored, so I repeat them here under no uncertain terms. If this wording is not fixed, count me opposed. — SMcCandlish☏¢ >ʌⱷ҅ᴥⱷʌ< 04:11, 20 October 2017 (UTC)
I've struck the bits, hopefully before this gets underway enough that the strike derails it. I'm fine with it either way, and Robert seems fine with it below. Jytdog may want to take a second look as the only pre-strike !vote, to see if it seems like a game changer for them or not. GMGtalk10:53, 20 October 2017 (UTC)
Support As we've recently found, we cannot leave it to well-intentioned researchers, if they have little or no experience of editing Wikipedia, to decide for themselves whether extensive additions or changes to Wikipedia content are improvements or not. They should run their proposals past the community before proceeding: Noyster (talk),13:14, 20 October 2017 (UTC)
Support per previous discussion on this, and comments above. Makes perfect sense to be clear about this. May want to also add in a link to Wikipedia:WikiProject Classroom coordination to provide those who wish to have class-based assignments on WP writing for additional advice. --MASEM (t) 13:37, 20 October 2017 (UTC)
Support - with a caveat, which is that despite the name and shortcut, this new policy is not to be interpreted as a ban on doing experiments at Wikipedia. As discussed in the talk prior to this RfC, the purpose of this policy is to stablish a much needed control and guidance over researchers doing interventional experiments, so that the experiment won't be disruptive nor unethical. It is not meant as a requirement that all researchers are doing it with an explicit desire in ther hearts to build an encyclopedia, since every editor may have their own reasons to participate; merely that we as a community expect that this improvement happens as a side-effect. Diego (talk) 13:57, 20 October 2017 (UTC)
Updated - per LZia (WMF)'s comment below, I'd prefer that the text requires researchers to be familiarized with the available guidelines and/or seek advice at the Village pump, but without obtaining consensus there being made a blocking requirement (as there may be other ways to gain consensus, and the Pump may prove an inadequate channel). Diego (talk) 11:43, 23 October 2017 (UTC)
Strong Oppose as written. Here are my main issues:
(1) Research which has no direct effect on the encyclopedia (i.e. no impact or influence on articlespace) should not be treated in the same way as other forms of research which do not affect the encyclopedia. "Surveying editors" is very, very different from "changes in content", but here they are the same.
(2) "Intervention" in the sense that includes "surveying editors" is incredibly broad, and would conceivably include anyone who asks a handful of editors a question to better understand Wikipedia before engaging in a journalistic, academic, scientific, sociological, computational, etc. exercise. These exercises are sometimes for their own sake, but more often to better understand Wikipedia, how Wikipedia works, to communicate elements of Wikipedia to the public, etc. and we shouldn't be erecting barriers to such projects that do not negatively impact the encyclopedia (since, after all, such projects are generally good for Wikipedia).
(3) Someone who does not engage in any research activity that has an impact on mainspace (and who does not attempt to influence mainspace in any way) is not a paid editor.
To be clear, the requirement I find most objectionable is the requirement that one finds consensus on the Village Pump (or wherever) before surveying even a small number of users.
Here's what I would support: requiring disclosure of affiliations/intentions in some manner, explicitly prohibiting breaching experiments, and requiring discussion/consensus before engaging in a very large or potentially disruptive project, such as when the researcher intends to contact a thousand users, or a project that involves editing mainspace in a way that prioritizes research over the quality of the encyclopedia, or a project that involves engaging users in a way that may be problematic for Wikipedia's/Wikimedia's policies about civility, harassment, etc. (i.e. potentially harming users). — Rhododendritestalk \\ 18:54, 20 October 2017 (UTC)
Well... At least in my mind, if someone is asking questions at somewhere like the help desk in order to better understand WP, or for journalistic endeavors or community education, etc., that's not really what's meant here by "research" in the sense of a structured experimental design, and it's not the kind of thing that, AFAIK, has ever exploded into the kind of ANI mess that this is intended to address.
I would also note that there are really no "rules" for finding consensus laid out, and getting the go ahead at VP would probably pretty much work like most consensus does and as common sense dictates. By this I mean that for obviously uncontroversial research (of the type with an actual structured design), like testing different new user notifications for editor retention, posting at VP would probably be little more than a public notification, and would require little if any in-depth discussion. On the other hand, for a large or potentially disruptive project, (importantly, as identified by the community and not by researchers themselves), the onus is on VP to escalate into an extended discussion or even an RfC if necessary. But if you don't have any requirement to get a second opinion somewhere, then we end up with what we've seen over and over, namely, researchers who know little to nothing about WP deciding unilaterally that their research design is overall ethical and non-disruptive, when it would be obvious to any experienced editor that that isn't the case. I'm not sure I can change your mind, but maybe this might clear some things up as far s the approach we've taken. GMGtalk19:13, 20 October 2017 (UTC)
If it's intended to be a narrower definition of "research," that's something to include before adding this to a fundamental policy page, rather than assume everybody will define it the same way. That other forms of research has not previously exploded into an ANI mess is further reason to be clearer rather than to just talk about "research" in such broad terms.
The problem with these different VP post scenarios is that it's so wildly varied, but there's the only nuance in this text has to be read between the lines. Sometimes of research will be a rubber stamp, some are just formal announcements, some will require long processes, some will go on for a week and be formally closed, etc. -- we're just leaving it up to whoever happens to see it, and until then everything that's considered "research" is subject to the same NOTLAB.
Though, again, none of this is to say this idea behind this proposal is a bad one. I mentioned what sort of thing I would support above. If we really want to do this well, we wouldn't lump it into VPP, but create something like "WP:IRB" that's linked from multiple places, including NOT and pages on Meta, where people can outline their research, people can comment, ideas can be flagged for further discussion, etc. A simple template on people's user page could direct them there.
Honestly, I agree with you nearly 100%. If I were king I would have formed WP:IRB outright. In fact, I mention exactly that in the discussion above. However, I have serious doubts that we can legislate that artificially. I really think the most we can do realistically is create the environment where VP figures out through consensus via editing, how these should be handled. That will undoubtedly raise incremental issues with the wording proposed here, which can be incrementally fixed. And once we have a precedent of participation and procedure, it may be appropriate to spin off a WP:VP/IRB in the future. But right now there is no requirement that would lead to that precedent, in order to figure out exactly how a WP:IRB would work, and the intention here is simply to set an expectation of daylight, of at the very least a signing off by a second opinion, and trust that the community is level headed enough to know what to do with it. GMGtalk21:48, 20 October 2017 (UTC)
I don't agree at all that surveying editors is less disruptive than mucking around in an article content. The difference is (in my view) that it way easier for WMF to explain that spamming editors' talk pages with survey requests in not what editor talk pages are for, and people who do surveys are generally aware of issues are spam. This is unlike people who come up with some "great idea" about an experiment using content and can just do it. But both are disruptive with regard to the work we are all trying to do. Jytdog (talk) 21:58, 20 October 2017 (UTC)
Oppose as written. Good idea, and I support the general idea of such an addition / section, but has some problems as written. So my "oppose" is just to make sure that this is done well. "Obtain consensus" is a high bar that would usually be a roadblock. Also per other objections raised above. North8000 (talk) 21:54, 20 October 2017 (UTC)
Oppose per WP:CREEP. Experiments can take many forms and so each case should be dealt with on its merits. We already have numerous policies addressing actual disruption such as WP:HOAX and so we don't need an additional meta-rule forbidding the idea of experimentation as this might equally be positive and helpful. Andrew D. (talk) 22:16, 20 October 2017 (UTC)
Oppose research has the potential to help us understand Wikipedia better to make it better. Adding an additional policy to try to control detrimental behaviour that is already covered by other policies is not needed. John Cummings (talk) 22:47, 20 October 2017 (UTC)
Oppose for a two reasons. First, it's a prima facie attempt at control over things that are/might be true about Wikipedia. For example, if a researcher wants to explore how/if editors write about the history of black conservatives, they are at risk of getting blocked by people who think that any research on race is biased, or liberals who think that this is amplifying "falsities." Let's think about any of the research that's attempted to be done on the gendergap, for example. In that context, VP becomes a battle ground for consensus, and research is done or not based on who can cultivate the most votes (or create automated sock puppets). Second, Wikipedia is the encyclopedia anyone can edit, and constitutes a public document. Therefore, just on the basis of research protocols, any behavior/happenings that occur here are not bound to privacy as laid out by most Human Subjects protocols of Institutional Review Boards. Now, interacting with editors in a survey form IS under IRB. But textual analysis or academic reflections on an editors' experiences editing are not really in any violation of any known research ethics -- codified or implied. Finally, how on earth will anyone police this? SO I didn't get consensus to study the complicated negotiations of race on Wikipedia... you know what I'll do? Do it anyway. From an IP address. My suggestion: rather than create more WP: policies, create a space to engage in conversations about how best to approach and support research (not how best to police it).Thebrycepeake (talk) 22:49, 20 October 2017 (UTC)
On the contrary: those discussions - some of which are already known to me - demonstrate that we already have the tools to deal with such inappropriate behaviour; there remains no indication of a problem, beyond that, which requires this proposed solution. Furthermore, no attempt has been made to demonstrate such a need. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits11:30, 21 October 2017 (UTC)
If you think it is OK that it took the community three ANIs over the course of two years to deal with one shitty experiment, or that somebody had a computer generate "articles" and then uploaded them, treating the editing community like lab rats to see how we react, then you and I do not inhabit the same universe. Jytdog (talk) 13:46, 22 October 2017 (UTC)
Oppose especially the notion includes " "Intervention" in the sense that includes "surveying editors". I have surveyed editors using the email link and published the results in a mainline scholarly journal (read it at http://www.americanhistoryprojects.com/downloads/JMH1812.PDF ). I asked how they got involved in the topic ("War of 1812") and nearly all responded without quibble, and explicitly gave me permission to use their real names. That is, the editors actually involved in the survey can make the decision--leaving the decision to anyone else ("prior consensus") is a bad mistake in my opinion. Rjensen (talk) 01:31, 21 October 2017 (UTC)
Undue weight: way too long a text for the perceived problem. The general idea is slightly at odds with rule 0, i.e. that Wikipedia works only in practice and not in theory. I prefer m:Experiments as a formulation of the problem. Nemo06:31, 21 October 2017 (UTC)
Oppose: Why go down this rabbit hole? The possible cons of ban greatly outweigh the possible benefits of it and it can be really difficult to find a mechanism to uphold it. The 22 articles made in the incident in question (even if bad) surely shouldn't justify it. There are many newbie or bad articles added for various reasons and many end up contributing to the system in some ways. Most newbie articles would fall in this category, but that's where new valued contributors come from. Wikipedia has so far been a great source of both knowledge and an understanding of how large-scale collaborations can work in the first place. At some point a better understanding of that may need an experiment here and there. A discussion on ethics will be a part of the design in any case. Making the predictions public to the participants may cause problems to the experiment is it may bias what people do. Pusle8 (talk) 13:40, 21 October 2017 (UTC)
Support we need a guideline for ethical, permitted and involved work that involves content. We've already experienced disruption on this front; whether because of ignorance or willful actions doesn't matter unless we've declared the boundaries of permissible behavior. Doing this proactively instead of calling up an ANI after the fact is just good self-management. ☆ Bri (talk) 23:09, 21 October 2017 (UTC)
Oppose. I was a researcher of Wikipedia long before I was WMF staff, so I think I can legitimately participate in this discussion under my volunteer account. As written, this proposal is far too broad in scope, doesn't tackle thorny issues around adjudication, review, and enforcement, and overall would have a chilling effect on research by GF people who are willing to follow guidelines, while not really hampering the work of people who don't care about the potential negative consequences of their research. If the community wanted to form a policy specifically about breaching experiments that involve changing article content, I would be happy to participate in the development of such a policy proposal. J-Mo16:57, 23 October 2017 (UTC)
Support I favor supporting research which is totally non-interventional. That sort of research is different from research interacting with Wikimedia community members or Wikimedia projects, which I oppose. I have interacted with perhaps 50 different research projects doing research on the Wikimedia community. All of these have caused problems. In general, I think that the English Wikipedia community should say NO to all researchers by default. Before I give reasons for that, I want to share that I find the discussion here to be under-informed. No one has yet mentioned that there have been efforts to establish meta:Research:Committee and to mediate research in Wikimedia projects. If there is to be research, then we need to assert some ground rules. There have been too many historical problems to say that research should proceed whenever, however, with no oversight, and with no channel for flagging concerns or problems. If anyone wants research to proceed, then those people need to identify the funding needed to provide the resources needed to clean the mess. Research makes messes and having researchers create unfortunate situations for random Wikipedians is not a reasonable plan for the future. Here are some of the major problems associated with research:
In the United States and other developed countries there are standards for research. For even the most respected universities and research institutions, top professors and academics feel comfortable abandoning all ethical standards when they work with online communities as compared to people they can see. In general, research programs have not adapted to acknowledge the advent of the Internet. If Wikipedia is to permit research, then we need a ground rule that researchers need to make a commitment to follow their own institution's own policies for ethical interaction with human research subjects. Almost no researcher is actually conscious that they should do this, or even if they get the approval, most researchers do not show respect for Wikipedia community members in any sensible way.
Most research proposals are bunk from their premise. Too many researchers are unwilling to spend their time doing preliminary research to ensure that their research is valid and relevant.
Paired with #2, those researchers who do not do their preliminary research are often entirely willing to attempt to recruit as much Wikipedia volunteer time as possible for their research. 20 hours of research preparation is scarcely enough to plan to recruit 200 hours of Wikipedia community volunteer time.
Researchers have no respect for volunteer time and greatly 10% underestimate the research pool. There are about 40,000 active editors in English Wikipedia. As a rule of thumb, for any target research demographic, only 10% of that number exist. So 10% of that 40,000 are super-editors, or WikiProject Participants, or women, or want to talk about harassment, or are conscious about their co-learning practices, or whatever. Most research projects imagine that they will interact with a pool of 100s of 1000s of potential research participants, when actually, a study that for example wants to identify gay Wikipedians making a lot of edits and who write about politics will have a potential pool of 40 editors. Wikipedia is not Facebook or Twitter and research metholodologies designed for countless masses do not work on wiki.
Time spent in research is time spent away from Wikimedia engagement. There is a cost to allowing researchers into Wiki community spaces. We should not support projects which come into Wiki community space with intent to take away our community members' time, unless we have reason to believe that good for Wikimedia comes out of it. Right now most research is bad and good does not come out of participation.
Most research does not match our publishing values. Most research projects never get published period, even after taking many volunteer hours. We should only support research which is open access, open data, and which allows the next researcher who comes along to build from it without trying to recruit the same volunteers repeatedly.
There is such a thing as "research fatigue". If study participants repeatedly get asked to be in research, they tire. We want to save people's attention for legitimate research.
The scale of research demand is never ending. There are 1000 universities in the world and every one of them wants access to the same Wikipedians every year to ask the same questions. The only thing preventing this is that we have some barriers to push back on lazy research.
We do not have any functional channel for flagging problems or communicating between researchers and the people who perceive problems. When a research problem happens, then the Wiki community is at a disadvantage to address it.
The attention of our community members is an asset. When researchers come, they market themselves as being representatives of science and progress. They use "science" and "research" as trusted names in branding and advertising. I favor good science, and I favor good research, but 99% of researchers who want to do research are not able to do legitimate research. The biggest part of the problem is not the researchers, but rather, that we do not have wiki guidelines for them to follow. If we make a list of basic research demands then most researchers will be able to follow those. Since we do not have those demands written, discussed, and in enforcement, I say no research. Blue Rasberry (talk)18:23, 23 October 2017 (UTC)
Oppose. In general, I agree that Wikipedia is platform to create and curate a specific kind of content. Other goals related to Wikipedia can have disruptive effects and create more work for the Wikipedians (e.g. for such a research: putting in vandalism in order to find out how long it takes before it is removed). I am worried that a too strict policy will make research de facto impossible. Long discussions about every case will be the result. It would be the question what behavior is really problematic - and I think that that behavior is already covered by policies such as BLP etc. Ziko (talk) 19:36, 23 October 2017 (UTC)
Weak oppose but just because of the wording. I would prefer something more concise like "While research on Wikipedia content requires no permission, interventional research performed without the informed consent of the subjects and the community more generally is not permitted. Researchers are required to disclose that they are conducting research on their user page, and must comply with all applicable policies, including those on conflicts of interest and paid editing. Disruptive interaction with Wikipedia, whether or not it serves a research purpose, is prohibited and may result in a loss of editing privileges." I'd also want a footnote better distinguishing between interventional and non-interventional research, ideally with a couple illustrations of obviously permitted and obviously prohibited research. —/Mendaliv/2¢/Δ's/ 13:25, 26 October 2017 (UTC)
Oppose proposal, not necessarily the underlying idea. To begin with, I hate the WP:NOT format overall. Add to that that what you're really talking about is WP:COI or WP:POINT or the licensing or something. Add to that the absurdity that you are arguing "Wikipedia is not a laboratory" and then laying out the conditions for getting your research to pass your notion of the local institutional review board! The net effect being not obviously much different from status quo ante. There is a valid issue in here, but you'll have to go back and reconsider the philosophy from scratch. Wnt (talk) 15:41, 28 October 2017 (UTC)
Support disclosure but oppose addition to WP:NOT. Editors must disclose paid editing or school-project editing or any other-than-personal editing. However I suggest updating Wikipedia:Paid-contribution disclosure instead to encompass disclosure of other-than-personal editing. Bright☀12:31, 2 November 2017 (UTC)
Oppose the proposal as written for reasons already offered by others. "Interventional" is not a standard term and lacks clearly understood meanings for researchers or Wikipedians. The obligation to get consensus at the Village Pump is a high bar that will deter (or be used to eliminate) good faith researchers who are not already community members. Feels like a case of using WP:NOT to restate or expand rules and policies already codified elsewhere. I'd love to help find and discuss ways to address many of the very real and very important issues and problems raised here, but I don't think adding a vague and/or redundant rule to the regulatory thicket will really help. Aaron (talk) 23:33, 2 November 2017 (UTC)
Comment. I have to share in Rhododendrite's concern, as this does feel like it's far too broad. While ideally this would be used to deal with the people who would be abusing the goodwill of Wikipedia editors and not making any improvements to speak of, this is the type of thing that would have the potential for a lot of abuse. I was going to give a long drawn out example, but my concern here is that it could make it very easy for people to argue for a block or sanctions against people who weren't the intended targets for this policy. Even if common sense wins out and the person isn't blocked in any given scenario, this doesn't make for a very inviting environment and we could end up running off people who could make some really great contributions.
My fear is that this is very broad and in trying to deal with a very specific set of people, it runs the risk of wider, more awful repercussions. Sure, if this turns into something people abuse it can get revoked, but that's not an easy or immediate process. I just can't help recalling something someone told me when I tried to make a guideline that had very broad repercussions - that it would be like trying to swat a mosquito with a sledgehammer (or something along those lines). You'd get the work done, but you'd also do quite a bit of damage in the process. Tokyogirl79 (｡◕‿◕｡)21:02, 20 October 2017 (UTC)
I'm afraid you may have to give a long drawn out example TG, at least for the sake of argument. I really did try to imagine every possible instance of unintended consequences, and assuming our community and our admins have a good feel for common sense and ignoring rules, including this one, when they don't help the project, I don't really see much potential for abuse. There is the added burden on good faith regular contributors to post a notice on VP and on the user page, but that's still much much lower than any standard that any bona fide academic institution would place on research involving human subjects. I think in my mind, the next time this crops up at ANI, this helps to avoid the question of whether there is such a thing as good-faith disruption for the point of research. GMGtalk21:21, 20 October 2017 (UTC)
The term "making changes in content" could be used to mean just about anything, as can the term "surveying editors". So here are a few scenarios:
An editor has been here for a while. They're considered a good editor by a large amount of people, but they do have their detractors - as is common on here. At some point this person writes a research paper about Wikipedia from an editor's POV, using their perspective as an editor. They don't post at the VP because they see it as a reflective essay rather than the type of research mentioned above, but they do mention this paper on their userpage or to a few people after it's published because they're proud of it because they think it makes both them and Wikipedia look good. However another person disagrees (maybe one of their detractors) and thinks that they've violated the guideline and brings it up at ANI or somewhere similar. Their argument is that some of the content and questions the editor made during the course of their time up to that point were deliberately done in order to get a reaction or to perform research for the paper. The accused editor responds by saying that this wasn't the case and that any questions or edits were legitimately done to improve Wikipedia. The accused editor gets a short block, but this upsets them so much that they leave Wikipedia - meaning that we've lost a good editor.
An editor has openly stated that they research the Internet and makes an offhand remark somewhere that what they're doing on here would make an awesome research project. This editor is pretty analytical and is always curious to know why things are done a certain way and how it came about that way - this is just something they do in their everyday life and part of why they became a researcher. At some point someone assumes that the offhand remark means that the editor is conducting research in secret akin to what likely is intended in this policy, ie a survey research project and reports them. It blows up into a huge argument and it gains the attention of some off-WP people who draw negative comparisons to the police report in Minority Report, as they say that it's possible that the editor could turn this into a research project that would violate guidelines at some point in time, but they hadn't yet done anything along those lines. Editor in question starts editing less on Wikipedia, not because they don't like Wikipedia but because they don't like all of the attention that this is getting.
Another scenario: a student comes on to Wikipedia as part of a class assignment, where their only requirement is to add content to Wikipedia. During the assignment they ask questions, which strikes someone as a violation of this guideline since the student and/or their class (it may be that the student is doing it by themselves, possibly as part of a class) didn't get permission from the Village Pump. This gets brought up somewhere and causes a huge discussion about whether or not student work falls under the guidelines. This scares the student off so quickly that there's only a student-shaped cloud where they last stood. The student is scared off and that's another editor we've lost.
In my opinion this is all potentially capable of happening with how loose this is worded, as it could apply to a pretty wide variety of people and situations. While sure, we could argue that common sense should keep all of this from happening, we shouldn't always rely on that. I think that the wording here would have to be very specific to show who it's actually meant to deal with, to the point where this may not be the best place for this. Tokyogirl79 (｡◕‿◕｡)14:41, 23 October 2017 (UTC)
Basically what I'm afraid is that while this may have good intentions now, this could very easily turn into a policy that supports witchhunts and could be warped to fit scenarios it was never meant to be used for. As we've seen on here time and time again, guidelines and policies that have blocking and sanctions as a punishment/result are very, very intricate and specific for a reason. This should not be approved unless it's extremely specific and detailed, which we don't have at this point in time. This is so broad that it has a pretty big chance of getting abused, despite good intentions and common sense. Even if someone trying to abuse the system doesn't succeed most of the time, just going through something like this runs a serious risk of alienating people - especially people that the policy isn't aimed at. I'm not saying never for a guideline about using Wikipedia as research and finding a way to deal with people who exploit the site and its users in a detrimental manner, just that this isn't the way to deal with them. There's too much chance for collateral damage stemming from someone either misunderstanding the policy's intent or willfully misinterpreting and using it as a way to punish someone. Tokyogirl79 (｡◕‿◕｡)14:47, 24 October 2017 (UTC)
It does look as if the RfC needs to be revised: (1) the first two sentences could go, while retaining the links in Note 1; (2) two types of research are conflated here: surveying editors or making changes in content, which are very different activities and require separate discussions over what if any prior conditions should be imposed: Noyster (talk),12:25, 21 October 2017 (UTC)
Comment - I agree that research involving surveying of editors is reasonable. Research that involves making changes in content for any other reason than the improvement of the encyclopedia is the problem. There have been many experiments, including the one in point, that involved changing content other than simply to improve the encyclopedia. Robert McClenon (talk) 18:44, 21 October 2017 (UTC)
Can always go back to the drawing board for a few weeks and be sure to seek input from some of the more robust opposes before proposing again. GMGtalk16:00, 22 October 2017 (UTC)
I concur. This has been useful as feedback exercise, but is a dead stick when it comes to adopting the proposed change. A revised version will have better chances, and will actually be better in all likelihood. — SMcCandlish☏¢ >ʌⱷ҅ᴥⱷʌ< 14:50, 2 November 2017 (UTC)
User:Jytdog, User:SMcCandlish - Is there a revised version available? You are asking me to withdraw the current version, but I am not entirely sure why it is important to withdraw it rather than let it run and be closed, unless there is a new version available. Robert McClenon (talk) 03:31, 3 November 2017 (UTC)
Because this one will not pass, and we already have more than enough feedback to redraft something that addresses the concerns raised. Dragging it out to the bitter end won't be productive. — SMcCandlish☏¢ >ʌⱷ҅ᴥⱷʌ< 03:58, 3 November 2017 (UTC)
I have to agree with them. This has probably gone on long enough to get some good feedback and a few interested editors to bring to the back room discussion for their input. I don't really see a compelling reason to have lego bot pinging more people here, especially since they're probably unlikely to have been the types of folks to were watching this page to begin with, and so unlikely to be nearly as interested or offer nearly as in-depth feed back as we've already gotten. GMGtalk10:27, 3 November 2017 (UTC)
Robert: What's next is to probably have each of us reread this thread at least a couple of times from start to finish, and see if we can reconcile the feedback here, as well as the alternate proposals, and make another draft. Then ping the most robust opposes to get their opinions, and hopefully go into a second proposal with a consensus of a half dozen editors or a dozen if possible, instead of three or four, and see if it gets more traction. If at that point it doesn't then... well... go try to write a GA I guess. GMGtalk19:09, 4 November 2017 (UTC)
Yeah, I would use a sandbox for the redrafting and ping people to the sandbox. Work up something most people seem happy with, then use that as a new proposal. — SMcCandlish☏¢ >ʌⱷ҅ᴥⱷʌ< 10:08, 5 November 2017 (UTC)
Comment - Hi all. This is Leila from WMF's Research team. I understand that this is a community RfC which is seeking votes from editors. I'm going to stay out of that entirely, but I still would like to share some thoughts with you given the experience I've had in the past almost 4 years working as a research scientist on Wikimedia projects. I hope this perspective can help you to arrive at a solution that eventually helps all of us build stronger projects.
In my experience, and as mentioned by others on different parts of this RfC, there are two main categories that could be addressed (either through compilation of current guidelines/policies in one place or through that plus introducing new ones):
Research projects that do not directly aim to improve Wikipedia (or other Wikimedia projects). Although it is clear for some of you that this kind of project, when it involves change in content and work for editors, should be stopped/banned, what to do with this kind of research is sometimes not as clear cut. One reason can be that when a research effort starts, it's not always very clear from early on all the benefits/learnings/insights that can eventually come out of it. For example, a project that intends to remove content from Wikipedia to learn about human cognitive abilities may help us learn about how to create and organize content on Wikipedia, and maybe not. On the other hand, a project that aims to see if adding specific content to Wikipedia is going to improve purchases in platform x, is most probably not going to help the project(s), or if it does, it's too indirect.
Research projects that do directly aim to improve Wikipedia:
Those that can be shared publicly prior to the study: It's good to have clear guidelines on how they should be shared, with whom, do the researchers have to wait for someone or a group of people to show a green light to them, ...?
What if the project cannot be shred publicly or with a large group of people, because then that knowledge can have impact on the results? Again, guidelines are very helpful here. At least in one discussion I saw a suggestion for people to share these ideas via OTRS. That's an interesting idea that can probably work, except that your queue of OTRS tickets to process may become even longer.
I have a few things I'd like to ask you to consider beyond the above:
Please be mindful of your own workloads, backlogs, and queues. I get it that when things go wrong you have to spend time on them and that counts, too, but as much as you can, I'd recommend you don't make yourselves a blocker for decisions on whether research projects can run on enwiki or not, unless you see a clear need.
This discussion, whether it results in a new policy/addendum/etc or not is useful in that it creates an opportunity to gather a list of guidelines we can share with researchers when we meet them. For example, every year we organize Wiki Workshop and such guidelines can be shared with the broader research community there.
I'm going to continue reading this thread to see where I/we can help. It's not super clear to me where the boundaries are, but please be aware that I'm eager to see where I can help. --LZia (WMF) (talk) 11:21, 23 October 2017 (UTC)
User:LZia (WMF) This proposal does not seek to ban anything.
The doomsday scenario you propose about someone floating an idea and getting pounded on is unlikely. Things are talked through simply before decisions are made.
The proposal clearly notes where people interested in doing research should seek prior consensus.
Probably more importantly... for editors in the volunteer community, new content is not "drops in the ocean" but actual words and sentences and references that individual volunteers here actually read and wrestle with and try to incorporate into the body of the encyclopedia.
The time editors spend maintaining and creating content is the lifeblood of this place, and that is what I at least am concerned to protect. I am "mindful" of that. You don't seem to be mindful of that. Jytdog (talk) 19:26, 23 October 2017 (UTC)
@Jytdog: I see that what I intended to transfer via my note is read by you very differently than what I had in mind. I did not talk about any doomsday, I did not talk about anyone getting pounded, no "drops in the ocean", and I did value editors time. Though I honestly feel a bit pounded by your response. It feels like you read someone else's comment and responded to me. :( --LZia (WMF) (talk) 09:00, 24 October 2017 (UTC)
Yes. In other words, the WMF doesn't say anything useful. The WMF is part of the problem at least as often as it is part of the solution. Oh well. Robert McClenon (talk) 03:31, 3 November 2017 (UTC)
Here is a potential alternative. I agree with the sentiment above that it would likely require a separate RfC rather than try to modify this one, but for the sake of discussion I'll post it here. It's decidedly less firm than the original, but also emphasizes the main issue [from my perspective], which is that research projects shouldn't be disruptive or negatively affect articles.
Research about Wikipedia's content, processes, and the people involved[1] can provide valuable insights and understanding that benefit pubic knowledge, scholarship, and the Wikipedia community, but Wikipedia is not a public laboratory. Research that analyzes articles, talk pages, or other content on Wikipedia is not typically controversial, since all of Wikipedia is open and freely usable. However, research projects that are disruptive to the community or which negatively affect articles—even temporarily—are not allowed and can result in loss of editing privileges. Before starting a potentially controversial project, researchers should open discussion at the Village Pump to ensure it will not interfere with Wikipedia's mission. Regardless of the type of project, researchers are advised to be as transparent as possible on their user pages, disclosing information like institutional connections and intentions.[2]
Upon consideration since my initial comments above, I think this is about the strictest version I'd support at this time (i.e. few absolute requirements), though that's not to say that some combination of this and the original wouldn't find sufficient support from others. — Rhododendritestalk \\ 17:38, 24 October 2017 (UTC)
This might be turning out constructive after all. Couple of initial thoughts: I'm not sure I'm on board with removing explicit links to COI and PAID. I could maybe see adding it possibly as a footnote, rather than part of the main body if that's a compromise, but I do kindof feel it should be in there. Also, I'm not so sure I would limit the scope to mainspace. So, for example, now that ACTRIAL is well under way, and considering that there is a possibility it might move toward something more permanent in the future, I could easily imaging disruptive research testing out the limits of AfC that would be equally a waste of editors' time even though it would be entirely in draft space. GMGtalk17:43, 24 October 2017 (UTC)
One thing I'd maybe add to what I wrote is something to be clear that any research project which involves directly modifying content (even if not negatively affecting the article) is "potentially controversial"). That seems like it would cover the AfC concern (and others). And if we're already discouraging (or placing barriers to) research that affects article content, PAID and COI aren't as relevant. That's not to say irrelevant, but it places most such cases in a gray area outside of the opening line of COI defined as "contributing to Wikipedia about yourself, family, friends, clients, employers, or your financial and other relationships". Even WP:DISCLOSE says "If you become involved in an article where you..." without adding anything about engaging with Wikipedia outside of "in an article". It seems to me that "disclosing information like institutional connections and intentions" covers what we'd want people to do by pointing them to COI. Perhaps a footnote there saying that Wikipedia has policies about COI and paid editing they should be familiar with. — Rhododendritestalk \\ 17:57, 24 October 2017 (UTC)
Maybe a footnote after "Before starting a potentially controversial project," saying "Projects that are "potentially controversial" include, but are not limited to, any project that involves directly changing article content (contributors are expected to have as their primary motivation the betterment of the encyclopedia, without a competing motivation such as research objectives), any project that involves contacting a very large number of editors, and any project that involves asking sensitive questions about their real-life identities."
I agree with with Rhododendrites's criterion just above. Anything that can be investigated with what is already in WP or in the edit history or on talk pages is both fair use, and specifically protected by our license--regardless of the purpose of the research , and it does not have to be benefitting the encyclopedia. DGG ( talk ) 09:41, 25 October 2017 (UTC)
I realize this discussion has at this point died down to almost nothing, but for whatever it's worth this right here is pretty much the kind of thing I personally had in mind with this proposal. It's a simple, to-the-point notification to the community that research is being conducted. It's unlikely to receive perhaps any follow up comments at all, but in the case something did obviously stand out to someone as potentially unethical or damaging, there is an opportunity for someone to raise issue in a public forum for public discussion. GMGtalk12:32, 2 November 2017 (UTC)
Research about Wikipedia's content, processes, and the people involved[1] can provide valuable insights and understanding that benefit pubic knowledge, scholarship, and the Wikipedia community, but Wikipedia is not a public laboratory and every editorsare not lab ratshas human subject research protection. Research that analyzes articles, talk pages, or other content on Wikipedia is not typically controversial, since all of Wikipedia is open and freely usable. However, research projects that are disruptive to the community or which negatively affect articles—even temporarily—are not allowed and can result in loss of editing privileges. Before starting a potentially controversial project,[2] researchers should open discussion at the Village Pump to ensure it will not interfere with Wikipedia's mission. Regardless of the type of project, researchers are advised to be as transparent as possible on their user pages, disclosing information like institutional connections and intentions.[3]
^"Projects that are "potentially controversial" include, but are not limited to, any project that involves directly changing article content (contributors are expected to have as their primary motivation the betterment of the encyclopedia, without a competing motivation such as research objectives), any project that involves contacting a very large number of editors, and any project that involves asking sensitive questions about their real-life identities.
I think the proposal aptly sums up the general feeling. Just for the sake of appearances, I'd prefer to change "editors are not lab rats" to "editors should not be treated as experimental subjects": Noyster (talk),10:51, 26 November 2017 (UTC)
User:Noyster that was on purpose. We actually do want editors to be treated like experimental subjects. There is law and regulation around how people in experiments are treated and researchers are obligated to get prior, informed consent. Some of what has happened in the past treated editors like lab rats (which also have protections, btw, but far fewer than humans). For example, the people who uploaded computer-generated articles to see how the editing community would react to them treated the editing community like lab rats. If it is too jarring, we could replace it with "and editors are protected by norms, regulations, and laws governing human subject research." but that is bloaty. Jytdog (talk) 16:52, 26 November 2017 (UTC)
I'm fine with it, but I suppose I'm overall more concerned with getting some kind of framework in place that can be adjusted/subtracted from/built upon over time, rather than starting with perfection. So I'm generally more interested in the opinions of others than my own. Adding User:Jtmorgan to the ping list to get their input as well. GMGtalk20:09, 27 November 2017 (UTC)
Thanks for the ping, @GreenMeansGo:. The current proposal seems clear, reasonable, and well-scoped to the issue at hand. "Editors are not lab rats" smacks of passive-aggression and I would change it to the bloated (also, more professional) wording proposed by Noyster, or Jytdog's second suggestion. If this becomes a policy or guideline, I think I will be comfortable pointing researchers to the document as a first step in planning their study. Including some examples of responsible notification behavior, like the noticeboard post about the dispute resolution focus group, as well as examples of good userpage disclosure practices like @Staeiou:'s, will help researchers understand what is expected of them, and help ameliorate feelings of intimidation that outsiders commonly feel when they are exposed to Wikipedia's community governance. J-Mo23:55, 27 November 2017 (UTC)
As far as I can tell, this is what I wrote above (with the extra line/footnote from the discussion thereafter) and with added labrats :) (please let me know if I'm missing some other substantive change), so yes I would support this. (had this tab up unsaved in the time labrats changed -- I do prefer the new language, too). — Rhododendritestalk \\ 14:38, 28 November 2017 (UTC)
The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
version history articles
Some software version history articles have been deleted citing WP:NOTCHANGELOG while others have been kept (Firefox version history vs Google Chrome OS version history). Could the policy article explicitly mentioned this category of articles as allowed or disallowed so these articles can be kept or deleted with confidence? Personally I tend to err on the side of keep, especially for articles I don't see value in deleting. But I would rather the policy be clear such that it can be applied consistently for better or for worse. --Saledomo (talk) 14:47, 10 January 2018 (UTC)
The Firefox article is definitely against NOTCHANGELOG. It shouldn't read like a change log; we don't need to document every minor version update and feature/bug fixes. These article should focus only on major updates, and discuss in text the major features or changes made. For example Windows XP#Service packs is the high level coverage of software changes we should have. --Masem (t) 15:07, 10 January 2018 (UTC)
Yeah, the policy doesn't give clear guidance. That much is obvious given that people keep on arguing about whether lists of changes to software deserve a place on Wikipedia. The problem isn't limited to full articles, either .... it also happens in "history" sections of software articles (e.g. Darwin (operating system).) Even tables that have reliable secondary sources have been summarily removed because it "violates NOTCHANGELOG". Problem with that is..... NOTCHANGELOG doesn't demand prose instead of lists (not anymore, anyways), nor does it prescribe what level of detail is appropriate. It only demands sources and common sense. That's it.
The problem with demanding "common sense" from editors is that such a thing doesn't really exist (WP:NOCOMMON, etc.) What I may believe is an appropriate level of detail is going to differ from the next person. (There's a reason why Urban Dictionary's #1 definition of common sense is "What I think others should know"...) The "common sense" phrase in this policy was added as part of a larger rewrite in 2013, but the use of the phrase was never discussed. I'm ---><--- this close to just removing the phrase outright since it doesn't actually translate into actionable (or enforceable) guidance for editors. It'd be much more useful to bulk up the language around demanding that there be no original research (i.e. people observing the software on their computer and writing content to match), and that we shouldn't merely regurgitate what the software author says. Yes I know that would be repeating what's already said in WP:NOR.... but I'm looking at the 450+ references in Windows 10 version history and see that more than 200 of them are Microsoft blog posts and support articles, many of which are used for nothing more than for release date / version numbers. The current reading of NOTCHANGELOG says this is not okay, but WP:PRIMARY says it is. — Warren. ‘ talk ,21:28, 18 January 2018 (UTC)
We can write guidance into NOTCHANGELOG that says changes must have a secondary (or independent, I'm sure WAID can correct me) source, regardless of PSTS or other guidance. --Izno (talk) 21:41, 18 January 2018 (UTC)
That said, the correct policy/guidance to apply there isn't PRIMARY, it's WEIGHT. If half of the citations are primary, you are emphasizing or adding the wrong information to the page in question. --Izno (talk) 21:42, 18 January 2018 (UTC)
It's a list of version numbers and release dates and literally nothing more. Not sure how you got "emphasizing the wrong information" out of that, or that somehow there's an issue of differing viewpoints that we're giving undue weight to. — Warren. ‘ talk ,21:52, 18 January 2018 (UTC)
The answer to your implied question is that the inclusionist wants WP:INDY sources, and the exclusionist wants secondary sources. For any commercially important or otherwise widely used software, there will always be a trade rag (=independent) that mentions updates. Actual analysis (or similar signs of secondary-ness) is rarer, so if you picked that standard, it would reduce the number of subjects that could be covered. WhatamIdoing (talk) 01:57, 23 January 2018 (UTC)
RFC for Wikipedia Is Not a Laboratory Version 2
Closing RFC, as it's clear no further comments are forthcoming (last !vote was on 18 January). Measuring consensus when determining the outcome of a Wikipedia discussion is explicitly not done by headcounting; the strength of arguments, grounded in Wikipedia policy and guidelines, are also taken into account. In this case, though, I can't reasonably discount or assign a differing weight to either sets of arguments for / against the proposal, as both are equally strong. Therefore, when all other things are equal, we fall back to noting that the majority of opinions expressed are in favour of the proposal, and conclude that the consensus is in favour of adding the paragraph to WP:NOT. fish&karate15:08, 5 February 2018 (UTC)
The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
Following the discussion above here and via the prior, withdrawn-for-revision RfC here, am proposing a revised version.
Research about Wikipedia's content, processes, and the people involved[1] can provide valuable insights and understanding that benefit public knowledge, scholarship, and the Wikipedia community, but Wikipedia is not a public laboratory. Research that analyzes articles, talk pages, or other content on Wikipedia is not typically controversial, since all of Wikipedia is open and freely usable. However, research projects that are disruptive to the community or which negatively affect articles—even temporarily—are not allowed and can result in loss of editing privileges. Before starting a potentially controversial project,[2] researchers should open discussion at the Village Pump to ensure it will not interfere with Wikipedia's mission. Regardless of the type of project, researchers are advised to be as transparent as possible on their user pages, disclosing information like institutional connections and intentions.[3]
^"Projects that are "potentially controversial" include, but are not limited to, any project that involves directly changing article content (contributors are expected to have as their primary motivation the betterment of the encyclopedia, without a competing motivation such as research objectives), any project that involves contacting a very large number of editors, and any project that involves asking sensitive questions about their real-life identities.
So it is handy, the prior version that was failing to gain consensus is here:
The following discussion has been closed. Please do not modify it.
There has been much scholarship involving Wikipedia,[1] but Wikipedia is not a public laboratory and editors are not lab rats. Research that analyzes content or editor behavior needs no permission, because the content of Wikipedia is open to everyone. Research that is interventional—that involves surveying editors or making changes in content—should be conducted only with prior consensus at relevant WikiProjects, the Village Pump, or at the administrators' noticeboard, and should be disclosed clearly on the editor's user page.[2] Researchers must disclose outside connections under the conflict of interest guideline. If their work on Wikipedia is conducted in the course of their employment, they must disclose this under the paid editing policy. Editing directed at any goal other than building an encyclopedia, or improving the experience of editors and readers, is likely to be seen by the community as disruptive and will likely entail time-wasting cleanup by our volunteers. Editors who disregard community norms regarding design, transparency, or conduct may face sanctions, including the loss of editing privileges.
Oppose - needs more work on the content and the wording before ready to propose as a policy -- would putting it out now not be be violating it's own prohibition against experimenting ? ;-) Markbassett (talk) 01:32, 5 December 2017 (UTC)
Comment - is Village Pump really the best place to ask for permission? Since this would be research involving human subjects there might be restrictions involved that a "volunteer" discussion cannot address. Volunteer Marek 02:54, 5 December 2017 (UTC)
Support This has been reworded fairly and succinctly. The difference makes it clear why the previous version failed to get consensus, because it is hardly if the people are opposing the substance of the addition. –Ammarpad (talk) 09:43, 18 December 2017 (UTC)
Object as before. Furthermore, the clause including "asking sensitive questions about their real-life identities" would be interpreted as including any survey with an opt-in "leave your email address for a copy of our results/ so that we may contact you with follow-up questions" or "leave your name and contact to be entered into the draw for gift vouchers" part. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits16:50, 3 January 2018 (UTC)
Oppose The current rules are good enough to handle the linked case and I don't see an argument being made why this policy is needed to deal with a case like the linked case. Increasing the number of text of our policy makes it harder for newcomers to read through our policies and understand them and keeping the length of 'What Wikipedia is not' short to keep is valuable. Furthermore I don't think that a paragraph inside 'What Wikipedia is not' is the best way to go about this. I would prefer a new guideline page for doing research with Wikipedia. I add further thoughts in the comments about what such a page should contain. ChristianKl ❪✉❫ 13:09, 5 January 2018 (UTC)
Oppose, again, because of the wording. The amount of detail here is unnecessary and makes for a rule that could be confusing (or more likely, subject to wikilawyering). It should be simple: Interventional research requires preauthorization by the community at large through public discussion, and, when involving direct contact with individual editors, informed consent. Non-interventional research requires no authorization from the Wikipedia community. The Wikipedia Community understands the paid editing disclosure requirements as applying to research projects. Conduct that disrupts Wikipedia, regardless of the intention behind that conduct, is prohibited. There should be a footnote illustrating the difference between interventional and non-interventional, and a footnote warning prospective researchers to be aware of their institutions IRB policies and that WMF authorization may also be required in some cases. The entire sentence about Wikipedia being a valuable research resource does not belong in a pronouncement of policy. —/Mendaliv/2¢/Δ's/ 21:19, 5 January 2018 (UTC)
Oppose as redundant. Seems to already be covered by existing policy, in particular WP:DE. While I might agree with the gist of the wording, expanding this section with prose that is already covered just makes it longer. Dennis Brown - 2¢15:46, 18 January 2018 (UTC)
Oppose I'm not really sure what the point of this is or how it could be enforced beyond what we already do. If they are being disruptive, they get a disruptive editing block anyway. TonyBallioni (talk) 16:21, 18 January 2018 (UTC)
I have read the discussion: it’s not clear to me this would solve anything or even be a decent principle that we could point to. Like I said: we can deal with disruption by blocks already, and I’ll add that if there are issues with mainspace articles, we likely already have valid deletion grounds. I don’t think adding this to the policy would help anything, and it’s likely to turn into an ALLCAPS that no one ever uses. That’s enough for me to oppose: I don’t think we need it. TonyBallioni (talk) 20:23, 18 January 2018 (UTC)
Discussion
@Jytdog: - May want to put a "version two" or something in the header maybe. With the low participation so far, I suspect some who watch this page may be thinking they've already weighed in on the discussion. GMGtalk07:02, 4 December 2017 (UTC)
Hmm, seems a bit shy of the intent. I'd think it should say to more clearly delineate and broaden the difference between passive analysis and any active experimentation which does or might effect users, article content or data, or wikipedia operations. Also wordsmith a bit here -- change the "not typically controversial" to "acceptable", and then rework the whole ending. Either experimentation is a policy and so just a mentioning it at the pump does not alter it, or else it is a guideline and then is not grounds for action. I'd fall into the side of making it policy and never OK to guide folks to wikipedia as an encyclopedia or as a source of data for analysis and nothing more. Because anything that is letting any experiments in is weakening the policy and ... eventually will go wrong. Cheers Markbassett (talk) 01:42, 5 December 2017 (UTC)
VM, there was a good deal of debate as to what the correct venue would be. VP seemed to be generally preferred over WikiProjects or AN. There was some talk about maybe implementing a WP:IRB, but that's probably some time down the line. GMGtalk22:38, 5 December 2017 (UTC)
Concerning oppose vote by @Markbassett:. Please can you suggest the the needed "rewording"?. Perhaps you should preferably give it in draft, or highlight the words you think should be redacted or replaced so as to improve the final version generally. Thanks. –Ammarpad (talk) 09:50, 18 Decembe r 2017 (UTC)
The comment was that it needs more work on the content and the wording before ready to propose as a policy. Just seems like this has been throwing out a succession of semi-random candidates here, not clearly or jointly working on the wording, and not having a clearly stated basis or even a fully stated RFC. (Saying oh go look at the past discussions and following the breadcrumbs this ver 1 then here now ver 2 did not provide a summary or focus for this RFC.) I could offer my own dings against it, but the main one is it just has not showed the work, and for a guideline I want the matters clearer than usual.. Markbassett (talk) 01:54, 20 December 2017 (UTC)
I would assume community. Everything in the content section is potentially a valid argument at AfD, while I don't see this ever being one. GMGtalk14:07, 20 December 2017 (UTC)
I think this is the wrong way of going about regulating research on Wikidata. Instead of addining another paragraph to 'What Wikipedia is not' it would be better to write a guideline that actually helps improve the quality of the research about Wikipedia and towards which prospective researchers can be directed. In medical sciences preregistration of clinical trials is an important tool to establish research quality and it would be good to have an analogues process for Wikipedia that's structured the way that every announcement of a trial has it's own page with a stable permalink and not just a village pump entry. Afterwards the actual academic paper that's produced should link back to the preregistration on Wikipedia. ChristianKl ❪✉❫ 13:20, 5 January 2018 (UTC)
Further discussion
So, having opposed an October 2017 proposal, I find that that RfC was closed on 1 December, and a second RfC opened, which itself was closed on 20 December and no notification was issued to those involved in the first discussion, nor were the opposes considered in closing the second RfC. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits12:08, 3 January 2018 (UTC)
I agree that this was improper, especially considering the time of year. Such an RfC should stay open for 30 days and be listed at {{centralized discussion}} but neither of these things were done. I shall mark the supposed policy as disputed. Andrew D. (talk) 14:00, 3 January 2018 (UTC)
I think that it went through sufficient process and obtained sufficient support to be in there as-is without a disputed tag. It can be revisited in a new process. North8000 (talk) 14:45, 3 January 2018 (UTC)
Done.I choose to re-open it since the participants of the prev. RFC ought to have been mass-informed which was not done.Now that it's reopened, feel free to drop a messsage on the t/p of all the participants of the 1st RFC.Winged BladesGodric16:11, 3 January 2018 (UTC)
This is a good point. The second proposal was an improvement over the first, but still failed to take into account several comments, such as Wnt (IIRC) noting that there is an apparent contradiction in saying "it's not a place for X, but you can do X provided that Y". --Nemo21:26, 3 January 2018 (UTC)
Definitely no need to add a hyper-specific "don't do this" example to WP:NOT, in particular when disruptive editing already has an explicit guideline page of its own. The core of this addition is research projects that are disruptive to the community or which negatively affect articles—even temporarily—are not allowed and can result in loss of editing privileges, which can fit easily in WP:DISRUPT without attaching a preamble, lead-out, and three footnots. Sheesh. Bright☀09:47, 10 January 2018 (UTC)
Good points. I supported the change and still think that its content is good. But in hindsight, I think that it is misplaced here. WP:not is about the general nature of Wikipedia and its content, not a list of all of the things that one should not do on the Wikipedia website. North8000 (talk) 13:35, 10 January 2018 (UTC)
The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
Data
I made a bold edit to WP:INDISCRIMINATE, with the reason the title says "information", WP:NOTEVERYTHING says information. BOLDly changing "data" to "information". It was reverted with the reason thanks, but all the examples are data. This is clearly false: "Summary-only descriptions of works" isn't data in the usual sense, neither are lyrics, which is why the section is not about data but about indiscriminate information in general. Its title says so, WP:NOTEVERYTHING says so. The initial proposal says so, many subsequent discussions say so. The point about "raw data" was added by Diego Moya, apparently without discussion prior to the change. "Raw data" was actually previously proposed as a bullet point in the section, not as a representation of the section as a whole.
All this clearly points that the section is about information in general, not only data in the narrow sense.
Note that the original discussion you pointed out says "items of information" (as in, "a pile of facts"), not "information" in general. What difference in practical effects do you see in using the word "information" or "data"? I would prefer to agree on how we want the policy to work, and then find words that mean just that, instead of fighting over the preferred form on subjective terms.
For me, it's imperative that WP:INDISCRIMINATE doesn't get used as an approved policy stand-in for WP:IDONTLIKEIT, as unfortunately that's a very common way in which it's invoked. The more general the wording, the easier it is for people to misuse it that way. Saying that (all) information must be "put in context with explanations" seems excessive to me, as sometimes information provides enough context and explanation about itself.
What I expect of IINFO is that for all content we can answer the question "how is this information relevant to the article's topic?". For data in tabular or list form, it may be enough to provide good labels and organize the content with categories provided by independent sources. For more general non-tabular facts such as summaries and trivia, it's OK to require some explanation on how they are significant to the topic.
But then I think you may get a problem of infinite regress; since those explanations about the content are information as well, should they be put themselves in context with explanations about the explanation? And what about those? See, the problem with talking about unqualified "information" is that there's no way to limit when it applies and when it doesn't other than by personal preference, which has always been a *huge* problem with this policy. I don't oppose a change that clarifies to what kinds of content it applies, but merely changing "data" to "information" does not solve that problem. Diego (talk) 14:21, 15 February 2018 (UTC)
it's imperative that WP:INDISCRIMINATE doesn't get used as an approved policy stand-in for WP:IDONTLIKEIT - completely agree. I think you may get a problem of infinite regress; since those explanations about the content are information as well, should they be put themselves in context with explanations about the explanation? - disagree:
Context is finite
Context starts off with answering basic questions. Using the article wood as an example: "the earliest known plants to have grown wood is approximately 395 to 400 million years ago", "Recent use of wood has been enhanced by the addition of steel and bronze into construction", and "year-to-year variation in tree-ring widths and isotopic abundances gives clues to the prevailing climate at the time a tree was cut". "Earliest known" requires no further context; it answers a basic question about wood: when did it come about? "Recent use" and "year-to-year variation" are contextless. Nothing is explained; the reader has to make of it what they will. "Recent use" could fit into a paragraph about woodworking; woodworking needs no futher context because it answers the basic question "how is wood used?" If steel and bronze are actually discussed in reliable sources about woodworking (I'm sure they are), you have your context. "year-to-year" can be put in a section about the biology of trees. A section about tree biology answers the basic question "What are trees" and requires no further context. As it stands, these sentences are just dumped in the "history" section, as contextless single-sentence paragraphs. Giving them context does not require infinite regress, because eventually you get to the basic questions. What is wood? When was wood first made? How is wood formed? How is it used? And so on. Those basic questions require no more context other than "this is an article about wood; let's give a summary about wood, starting with the basics." Any information that's doesn't directly follow ("Harry Potter's wand is made of wood) needs to be couched in further context, that needs to exist in reliable sources. Otherwise, it's apparently indiscriminate because it's contextless.
TL;DR: Context is finite, it starts with the Five Ws or similar basic questions, and any information that's more intricate needs to be couched in context; the "context of the context" ends with the basic questions. Context must exist in reliable sources and be given due weight.
Are you complaining about "A Wikipedia article should not be a complete exposition of all possible details"? Usually I like detail, but there is a limit, and some examples of excess are given, eg we don't want a diary of every day's activity for a notable person. I think the sentence is vague enough to allow individual case discussion. Graeme Bartlett (talk) 04:52, 13 March 2018 (UTC)
Predicting outcome of US elections
I think it was 2016 that Neddy1234 (talk·contribs) started adding "prediction" sections to US election articles. Since then other eds have included a "prediction" section, including JocularJellyfish (talk·contribs), and various IPs. Every prediction section I have seen is a naked list of who predicted what outcome. In 2018, an IP is citing this earlier history to claim such context-free lists of predictions are "routine". For example, I have doubts the following table is appropriate.
In my mind, this naked table violates the spirit of WP:CRYSTALBALL and risks adding/subtracting reader motivation to go to the polls. The naked table omits any discussion of how the predictions were arrived at, what sources were used, or what significance the prediction might be having in the broader world. The same basic table (and sources) is repeated on various 2018 US election articles. Two of the references are to WP:PRIMARY sources, one of which has HQ in Virginia. One of the IPs fighting to keep this in geolocates to Virginia. Since the IP is echoing the table on other articles, it also smells of WP:PROMO. But the main issue is CRYSTALBALL..... QUESTION - is it appropriate for our election articles, prior to the election, to include a context-free, discussion-free, listing of various ultimate-winner predictions?NewsAndEventsGuy (talk) 14:44, 6 March 2018 (UTC)
I can see there's other issues, but let's assume that we use something like CNN's pre-election-day polling to predict the winner for an election and the method used is established, so there's no question of how they got to that conclusion; this leaves only the question whether it is appropriate to post before/on the day of the election. And to that answer I think it is fine, given our other general disclaims (eg we are not a legal advice book, we are not a medical advice source, etc.) That here, CNN did a poll to predict a winner is not a CRYSTALBALL issue (the poll was done), and we should not presume that WP articles are going to influence the election or should be taken as verified facts by readers. It is completely fine to post valid RS pre-election predictions before an election. --Masem (t) 14:58, 6 March 2018 (UTC)
Quoting reliable and respected polling data is not itself a violation of WP:CRYSTALBALL, so long as the source of the data is explicitly cited, which it is in the example above. I have zero problem with this information in Wikipedia articles. --Jayron3215:38, 6 March 2018 (UTC)
I will just note that I did not begin this precedent nor have I actively worked to expand it. When I was creating articles, I simply used the format present on another page that already had the predictions section on it without any notion that it was counter to en.wiki policy. I do think that having the boxes is not a problem and is good info to have on the page. – JocularJellyfishTalkContribs18:37, 6 March 2018 (UTC)
The validity and news-worthiness of predictions is very short-lived, and they have zero lasting impact. This is against WP:NOTNEWS and WP:NEWSEVENT, and therefore has no place in an encyclopedia. IMO, it should not be included. But if after an election there is a major/surprising upset, then it can be added that the election outcome was contrary to expectation. -- P 1 9 9✉14:24, 7 March 2018 (UTC)
Not necessarily; predictions that end up correct are just as valid data points (showing the election was no surprise) just as much as those that get it completely wrong; these become historically useful. Which predictions to include is a matter of RS and UNDUE (we don't need every pundit's, but things like CNN polling is good), but not a WP:NOT issue. --Masem (t) 14:36, 7 March 2018 (UTC)
Not convinced. Nobody cares about predictions and polling after elections, unless there is an upset. Nothing could be more fleeting and ephemeral. -- P 1 9 9✉15:42, 7 March 2018 (UTC)
Will have to agree to disagree then. I recognize the value of predictions clearly in a case like the last US Presidential election, but then to me, if I want to check that against the previous election (Obama's reelection), I'd want to see that it was not a surprise result. It's a combination of political interest, media, and the current political climate that the lack of a surprise is still worthy of being mentioned. --Masem (t) 16:12, 7 March 2018 (UTC)
Indicating CRYSTALBALLs of notable sources, cited to those sources, is definitely not a violation of WP:CRYSTALBALL. After all, we aren't predicting the future, we're only indicating the past in which other people attempted to indicate the future. Other policies may be relevant (UNDUE, NOTNEWS), but not CRYSTALBALL. עוד מישהוOd Mishehu14:48, 7 March 2018 (UTC)
Not a CRYSTAL issue, as others have pointed out but if a guideline would be useful on: if, when, and what source(s) - please feel free to start one perhaps in an essay, and at the relevant project pages (politics? statistics?) Alanscottwalker (talk) 15:16, 7 March 2018 (UTC)
This is better to include than raw polling data, as it's generally informed analysis from experienced journalists, rather than raw data. I do agree that it shouldn't be discussed too prominently (i.e. not in the lede), but having a section seems fine. power~enwiki (π, ν) 01:24, 14 March 2018 (UTC)
CRYSTALBALL merely says that any predictions about the future should be sourced, not made by Wikipedia editors. Predictions are part of the story and should be included when they have significant coverage. TFD (talk) 02:21, 14 March 2018 (UTC)
Agree with Alanscottwalker that it's more a case of 'guidelines on what to include' not 'never include them'. Predictions are very much a thing (particularly in US politics) analogous to polls. I think there is a strong case that predictions are not actually that predictive, but it's not really our place to dive into that debate - we just describe. I would suggest: only include predictions made by sources made by reputable media organisations or polling companies, exclude predictions where there is not otherwise much substantive content in an article, and simply report the predictions rather than using them as a basis for the narrative of the article. The Land (talk) 23:31, 15 March 2018 (UTC)
Apparently not very strict. I could see the one page being defensible if it is a significant Buddhist term, similar to the way we have articles on some Latin phrases that are important in science, law, or Catholicism, but the glossary article is a bit over the top. Beeblebrox (talk) 18:47, 17 March 2018 (UTC)
Just to double check I looked int he history of both pages, and this page explicitly said dictionary definition or lists of definitions are part fo what Wikipedia is not on the day the glossary was created. That being said, deleting entrenched cruft like this isn’t easy. Beeblebrox (talk) 18:53, 17 March 2018 (UTC)
Basically this. The meaning of policy in practice with regard to things like N and NOT, is the sum of how the community actually applies it at AfD... Ideally this will approximate the letter of the policy, and we'll amend policy when they get too out of synch, but that's an imperfect process. GMGtalk19:31, 17 March 2018 (UTC)
Feedback request: student homework, and NOTHOSTING
That looks like a two-year old rejected proposal. If a change is wanted now, please explain what the proposal is. Johnuniq (talk) 00:17, 5 April 2018 (UTC)
Can someone please tell me what is it? Because Wikipedia vets keep erasing kits history in some articles for this exact reason, even though the section does not being an issue in other Wikipedias (es, pt) even if the section was collapsible (did not cause damage to the article). Also why it only been a problem for those 3 national football teams (IDN, MAS, SIN). Thanks. – Flix11 (talk) 02:44, 11 April 2018 (UTC)
In-line with WP:NOTGALLERY and AfD consensus (which basically says you shouldn't have sections/articles just being kits) kits have been removed form a large number of national team articles. If there are others which have kit galleries remaining then feel free to remove them as well. WP:OTHERSTUFFEXISTS. GiantSnowman10:28, 11 April 2018 (UTC)
Hey Flix11. Part of the issue with NOTGALLERY, is that if there is enough media available to make a high quality gallery, then there is a place specifically for that, Wikimedia Commons. Just like if there is enough content to make a collection of pertinent quotes by or about an article's subject, then there is a place for that over at Wikiquote. All these work together to provide robust coverage of a subject, but they do it in different ways. GMGtalk11:03, 11 April 2018 (UTC)
To @GreenMeansGo: & @GiantSnowman:, I think if all kept in the Commons alone it would be a mess because the kits are dissected (kit body, left arm, etc.), thus did not inform anything at all. Some kits also have different versions by different makers, thus can make confusion about which one is the closest to resemble the real one. Besides many kits - especially those never touched by me - are not categorized to the respective national teams nor apparels (eg. Adidas specific patterns). Some national team categories are don't even exist (eg. Malaysia). It makes the effort to track the kits history in Commons much more difficult. I felt this about last year, when I did not want to make my own. It is because some kits are not properly named from FIFA trigramme (eg. ESP, MAS). Some are named in the maker's local language (eg. "Spanje" for Spain, made by a Dutch thus named so) and there are Malaysian kit named "blackmyszz". If this is about football clubs, I can agree with that, since there are too much kits to cover in the main club's page. They belong to the respective season's. If they do not belong to the team's main page, there should be distinctive pages of their history then put the kits history there (like in Spanish Wikipedia's Spain national team for example). And my question about why only those three teams affected still not answered. – Flix11 (talk) 14:28, 11 April 2018 (UTC)
It's often fairly easy to track down images based on their use (or prior use) in existing articles, even non-English ones, although sometimes it may take a little investigation to pull together. And for whatever it's worth, the limits on gallery creation are pretty open ended compared to really anything on Wikipedia. Also, something you may not have considered, is that as a sister project, if you create a gallery as a companion to an English article, it can also be linked to from the corresponding article in every language, since Commons is a multilingual project. So it may be a bit of extra work, but you're likely to get a lot of added value for it. GMGtalk14:36, 11 April 2018 (UTC)
Sure Flix11. As far as the software goes, it works almost exactly the same as creating a Wikipedia article. You just type in a page name on Commons, and if it doesn't exist already, it should give you the option to create it. The big difference is the type of content that gets added. Compare the gallery for c:American football, which pretty neatly sorts the images by broad subject, with explanatory captions for each. Then when you're done, you can connect the gallery with the associated Wikipedia article by using Template:Commons in the external links section.
Compare the article I'm working on now, Charles P. Mattocks. I ended up creating a Commons category, instead of a gallery, because categories are more appropriate when there isn't very much content, and not very much explaining to do. But I also made an English Wikiquote page, and all this comes together to support the Wikipedia article and add additional content for interested readers. GMGtalk15:25, 11 April 2018 (UTC)
I tried one and I found that the {{football kit box}} can't be accessed. Did it mean that they will appear separately instead of together as one as in Wikipedia? Then if I want to make a whole kit like in Wikipedia do I need to make adjoined version? Is it OK? – Flix11 (talk) 17:27, 11 April 2018 (UTC)
Do be aware that while most kit images can be considered free images and thus used at Commons, some may include copyrighted elements (particularly if they include advertising marks) and may not be acceptable at Commons. --Masem (t) 15:27, 11 April 2018 (UTC)
Ah. I see the problem. Template:Football kit box doesn't exist on Commons. AFAIK it's usually pretty straight forward for someone who knows what they're doing to transfer templates between projects. Unfortunately I don't, but I know User:Galobtter helped me last time I needed to transfer a template to Wikiquote. Maybe they can help here too. GMGtalk16:58, 11 April 2018 (UTC)
The key point of WP:NOTGALLERY just seems to be that we don't want "Photographs or media files with no accompanying text. If you are interested in presenting a picture, please provide an encyclopedic context..." So, a gallery of kit pictures would be fine provided that there are suitable captions and a narrative section covering the history of the kits. Andrew D. (talk) 11:59, 11 April 2018 (UTC)
Cuz it passes le'GNG and will probably pass SUSTAINED. I love how people can't wait a few days to create an article on a shooting or terror attack, yet 3 days have gone by and as bet I can tell no one has created an article on this much more impactful event. Thanks, L3X1◊distænt write◊13:33, 16 April 2018 (UTC)
The GNG specifically notes that a burst of news coverage is not sufficient for a standalone article, per NOT#NEWS. And yes, unless there are lawsuits that stem from it, it's not likely going to have long term effects. (2012 peanut butter recall is not a great example, and I think it should be deleted, but it demonstrates a long term impact, the closure of a company). That said, a list of major food recalls may be a possible topic of its own. --Masem (t) 13:41, 16 April 2018 (UTC)
I've begun a draft, and have noted that the 2010 egg recall is a redirect to a section on the incident in the principal corporation's article. From the reports I have been reading, it does appear that this egg recall is many times larger than the previous one. I hope we never have to make WP:NEGG. Thanks, L3X1◊distænt write◊13:49, 16 April 2018 (UTC)
You can probably add details of the egg recall to that existing article, but I'd keep in mind if there's not much more to this outbreak (contract to 2012 outbreak of Salmonella which is a legit large scale problem that a standalone makes sense) this might be merged into that list at a later point. --Masem (t) 14:19, 16 April 2018 (UTC)
WP:NOTMIRROR and open license text added to Wikipedia
Another important topic of discussions in Wikipedia is user's conduct. These discussions should never turn into forums about Wikipedians' political views, race, ethnicity, etc.
Rationale: The previous version was related to discussions of article topics only. But discussions of user conduct sometimes turn into ugly bickering about their political views, ethnic background, etc. Often as an "explanation" or guesswork why users behave this way. IMO this type of forums must be warned against as well. Misbehavior is not supposed to be "explained" or "justified" or otherwise rationalized: we must pinpoint it and warn against it. Period. Staszek Lem (talk) 18:11, 16 May 2018 (UTC)
I have reverted the above good faith edit as it represents more than a minor change to one of our more important policies and it should be discussed here first. Assuming there are no objections it can be re-added in a few days. Otherwise we will need to wait for consensus to form. I take no position on the proposed addition. -Ad Orientem (talk) 19:05, 16 May 2018 (UTC)
It sounds as a well intentioned addition, but it can easily turn into a problem. Yes, when it first comes to it, we try to focus the discussions in the article itself and not the users. But for more complex conflicts, discussing the conduct of a certain user may be required. And when we have a POV Warrior, yes, that includes discussing the way he tries to advance his agenda (which may be a political agenda, an agenda related to race or ethnicity, etc.). Those users may then equal their agenda with their own personal aspects (for example, a user trying to whitewash the "Black Lives Matter" or the Black Panther Party articles may complain "You are doing this because I'm black!"), and claim that this policy would protect them. Cambalachero (talk) 19:22, 16 May 2018 (UTC)
We have to discuss editor's edits, not his views, and especially not his ethnicity or gender. The discussion should never ever go, like, "of course, since he is a Pole, his edits are Ukrainophobic". Never. It always opens a can of worms of mutual hatred. because I am black -- this argument is exactly what my policy forbids, not protects: this statement is essentially an accusation in racism, and WP:BOOMERANG is in the air. Arguments involving personality traits and convictions are invalid in Wikipedia, and we have a policy for that, and they must be stopped as soon as possible. We discuss behavior. I can say "your edits are pushing homophobic agenda", but I cannot say "you are a homophobe". We must tirelessly educate people to understand this difference, to decrease the amount of personal/political drama. Staszek Lem (talk) 20:16, 16 May 2018 (UTC)
I think that it should be carefully readded. We already have the link to talk page guidelines there which should cover avoiding discussing editors rather than contributions/content. But if this needs to be made more clear, it should be along the lines that discusses about editors themselves should be only held at appropriate dispute resolution and/or adminstrator boards, not on normal talk pages. --Masem (t) 20:39, 16 May 2018 (UTC)
That is generally true, but is it necessary to specify this in policy? The idea is sound enough as a best practice to strive for, but the reality is that once something gets enshrined in policy it almost certainly will be used against good-faith users in ANI threads to obfuscate the root problem in disputes. @Staszek Lem: Tirelessly educating people doesn't strike me as a sound method of decreasing drama on Wikipedia. Fact is, we do have racists, sexists, anti-Semites, Islamophobes, homophobes, and all manner of other troublemakers and trolls. Whether practicing deliberate disruption or pushing a hateful agenda, their conduct here shouldn't result in a productive user who calls them out in good faith being at risk of sanction simply because they didn't exercise the best judgment in their choice of words. RivertorchFIREWATER19:08, 18 May 2018 (UTC)
A good-faith editor calling someone racist without proof is inevitably calling for flame wars. I find it ridiculous that demands to keep up with one's words are described as "obfuscating". Even AN/I threads are supposed to be about conduct, not views. "Tirelessly educating" is the only way to decrease drama. "Tirelessly blocking" was decided to be unacceptable by the community. I understand it is too late to educate seasoned "drama kings", but a guideline helping to stop their numbers from growing makes sense to me. Staszek Lem (talk) 22:45, 21 May 2018 (UTC)
Did someone call demands to keep up with one's words "obfuscating"? If you thought I did, please read what I wrote again more carefully. Obfuscation is a common tactic at ANI. People try to divert attention from their own behavior to that of others. Some of them are right to do so, but usually they're highly problematic users who are trying to wikilawyer their way out of a predicament by citing the letter of some policy or other while while ignoring the spirit of the great body of policies that enable this project to work. Sometimes they're successful in such an attempt, usually not, but it nearly always wastes time and energy.
I do, by the way, wonder if demands is the word you meant to use. We are all volunteers. If anyone starts demanding things of me around here, I'm liable to find a polite way to tell them to go fuck themselves. This has a way of happening in off-wiki contexts, too. Asking nicely tends to elicit a better response than demanding, or at least that's what I've found. And you mention a guideline, but we're discussing policy. As I'm sure you know after eight years of editing, there's a significant difference between the two. RivertorchFIREWATER07:21, 22 May 2018 (UTC)
Question related to proper listing of company employees in articles
I recently deleted a section containing names and titles of top company employees from Cylance [[10]], since I thought that WP:NOTDIR discouraged adding these types of vanity lists. I just read this policy essay and WP:NOTDIR#7 provides a carveout discouraging "...employees (except CEOs, supervisory directors and similar top functionaries)". If that's the case, should I have left the info there? It was sourced by two primary sources: the company web site and a press release. Similarly, company infoboxes often include multiple names of employees, but I tend to favor showing just the CEO, President and Chairman, unless the person is notable on their own. The company infobox policy is actually clearer, saying Up to four key individuals closely associated with the company. and Generally list prominent, current executives within the company holding key positions such as chairman, chief executive officer, president, VP of design, etc., but do not list roles below chief officer level if they are not notable. Do include top executives even not individually notable, but do not wikilink them. Using the template verbiage as a starting point, might we fine tune/clarify this policy to provide additional guidance for listing multiple names of executives within articles? TimTempleton(talk)(cont)19:01, 21 May 2018 (UTC)
Templates aren't policy, they are just guidance from one or more people. I probably would have deleted that section as well, and maybe moved one or two names in the infobox. prnewswire.com is not remotely reliable, btw, they publish what the companies tell them (pay them) to publish. Using their own website is fine for sourcing the one or two names. If you really see any inconsistency between the template and policy, the template should be changed. Dennis Brown - 2¢21:55, 21 May 2018 (UTC)
In this case, it's the policy that seems more permissive; the infobox template parameters are more restrictive and specific. The policy term "supervisory director" is a bit vague; it may be an English versus American term. We could easily remove that. Otherwise, it might be seen as condoning loading various executive names into articles. I also sometimes see board members listed, often also not that notable. If revised, it could simply say:
7. Simple listings without context information. Examples include, but are not limited to: listings of business alliances, clients, competitors, employees (except CEOs and similar top functionaries), equipment, estates, offices, store locations, products and services, sponsors, subdivisions and tourist attractions. Information about relevant single entries with encyclopedic information should be added as sourced prose. Lists of creative works in a wider context are permitted.
If the template is more restrictive, then consider it guidance, not policy. My experience has been that templates can be more restrictive for good reason, as a template isn't the sole solution to a good article, and often people try to do too much with them, instead of relying on well written prose. Dennis Brown - 2¢14:13, 26 May 2018 (UTC)
I'd have left the first para - founders, presumably still large shareholders, are worth having, plus a couple of current top execs. In a much larger company more might be justified. Johnbod (talk) 14:32, 26 May 2018 (UTC)
also wiki is not a place to find interesting, leading-edge research findings not yet supported by other research or secondary research
This gist is the "15 minutes of [almost entirely regional] fame" nature of the story. Just the fact that it's basically a human interest story, and not a subject anyone will care about in 100 years (or 5), is also implicated. I suggested to Thinker78 in user talk to ask for userspacing if the AfD closes against it, since it's possible the subject might actually get some more non-trivial coverage, or eventually set a national record. — SMcCandlish☏¢ 😼 23:49, 2 June 2018 (UTC)
Not done There is no request here. Please make requests in the form of Please change some text to some other text. or Please add/remove some text.Jbh Talk13:25, 10 June 2018 (UTC)
Concision proposal
This page's layout has bugged me for years. I propose to shorten the "Wikipedia is not a foo" headings to "... a foo" versions (retaining the lengthy originals as anchors of course).
While, yes, we do have [a lot of] shortcuts, these are for experienced editors' convenience on talk pages, and should not be thrown as alien alphabet soup at new users. In particular, all our policy, guideline, and process pages shouldn't be using them at noobs, since those pages are primarily intended as introductory materials for the recruits.
But it's severely tedious to have to format something like {{section link|Wikipedia:What Wikipedia is not|Wikipedia is not a publisher of original thought}} again and again. This would be so much easier as {{section link|Wikipedia:What Wikipedia is not|... a publisher of original thought}}.
And there'd be a reader-facing benefit as well: the actual URLs (in the URL bar and in the mouseover tooltip provided by many browsers to tell you where that link will take you) would show a much shorter and less pointlessly redundant path. — SMcCandlish☏¢ 😼 23:44, 2 June 2018 (UTC)
I do not see a reason to object so long as no existing links are broken. I deal with the issue of long/repetitious linking by using a browser based keyboard macro package called ProKeys. For example typing notpromo [crtl]-[space] yields [[WP:NOTPROMO|Wikipedia is not to be used for promotion or advertising.. There are a couple of downsides: It only works with the desktop versions of Chrome and Opera so I still must type everything out on my mobile devices; and, very infrequently, an experienced user may get a bit offended because AGF or similar will come out wikilinked or something gets linked repeatedly because it is easier to use the macro with linking than to type out the unlinked phrase. Regardless, it makes my interactions with new users more understandable and convenient for them and significantly less repetitive and frustrating for me. Jbh Talk13:54, 10 June 2018 (UTC)
Using the {{anchor}} template you can create anchors of any title that work to provide section links. If the main headings are changed then these will be required anyway to preserve the incoming links, so the discussion is really just about what titles are displayed on the page. I have a very slight preference for the long headings as I think it will be very slightly less confusing to arrive at a section titled "Wikipedia is not a publisher of original thought" from a link "...a publisher of original thought" than vice versa. Thryduulf (talk) 11:07, 14 June 2018 (UTC)
"Wikipedia is not censored" (and then Wikipedia tells you how it is censored)
Just stating that Wikipedia complies with US laws, when determining how to censor itself is an admission that it censors. The censorship is simultaneously self-censorship and censorship by the US government. Please don't lie, Wikipedia. It's not good form. Admit that this site is censored like everything else is. — Preceding unsigned comment added by 107.77.192.36 (talk) 04:53, 24 July 2018 (UTC)
This AFD back in 2012 raised the question of whether entries consisting of only phone numbers of non-notable organisations should be kept. Consensus was to keep due to the notability of most of the entries; however, it now seems that the vast majority are just numbers along with URLs to various sites. It seems these should be removed due to WP:NOTDIR. Another problematic example is the "Current storm information" list of three websites in the "Current storm information" for tropical storms such as at 2018 Pacific hurricane season#Hurricane Bud which likewise serves no encyclopedic purpose. Is there some sort of exception that I'm missing or is this rule just going to be ignored? 93 (talk) 19:55, 12 June 2018 (UTC) 93 (talk) 19:58, 12 June 2018 (UTC)
There are a lot of policies, style guides, etc. that are ignored in various articles, until someone takes on the task of trying to enforce them. Cleaning up just one article can at times turn into a time-consuming and aggravating task, and I may hesitate before even mentioning a minor problem in an article. It is an issue of rationing my time and energy to concentrate on things that I think are most likely to improve the encyclopedia. I am not saying we should ignore all rules (although there is WP:IAR), but I have accepted that I can't fix everything in Wikipedia. - Donald Albury21:08, 12 June 2018 (UTC)
When WP definition diverges from dictionary definition …
that divergence* should be clearly indicated. This goes beyond inserting an 'other uses' template linking to Wiktionary (or other dictionary). (* I.e., changing, adding, deleting from especially the generally accepted primary dictionary definition)
In such cases, the definition used in the article is not a distinct definition; it's effectively a change to an existing dictionary definition.
One solution: The lede for each such article should begin with an 'As used here, …' template that clearly indicates the divergence from the generally accept dictionary definition.
Clarify whether a well-annotated gallery illustrating the subject of an article fits the WP:NOTGALLERY
Here's an example of a gallery of images in an article about a typhoon. The images have detailed captions and serve to illustrate the different types of damage caused by the typhoon, for example broken windows, pavements, crushed cars and so on.
According to WP:NOTGALLERY, "Photographs or media files with no accompanying text" are not welcome. It further clarifies that inserting a single image with no text is not recommended. It does not appear to be explicitly referring to the "gallery" tag; rather, it uses the word "gallery" metaphorically, as in "Wikipedia is not your gallery".
Therefore it appears that using the gallery tag with detailed annotations to each image to illustrate different types of damage caused by a typhoon should not be against WP:GALLERY.
Just what number of images in an article crosses the "not a gallery" line is a judgment call, but I do think that thirteen images of damage from one typhoon in a Gallery section has crossed that line. What does the thirteenth, or the twelfth or even the third image tell us about the damage caused by the typhoon that cannot be conveyed by one or two well-chosen images? - Donald Albury11:51, 17 September 2018 (UTC)
I would say that this article doesnt need a galleries section as Donald Albury suggested one or two well placed images within the article is all that is needed. A reader wanting to see more images can use the link to Commons. MilborneOne (talk) 12:53, 17 September 2018 (UTC)
WP:CRYSTAL and flags in knockout tournament brackets
In ongoing knockout sporting events, especially in tennis and chess, when two people from the same country are paired against each other, for some reason a lot of wikipedians think it is appropriate to enter this flag into the draw for the next round, even though the identity of the participant is not known yet. This is a recent example - someone has entered the flag of China in the quarter finals. This to me is clear extrapolation, per point 3 of WP:CRYSTAL. How do we know both players won't withdraw? How do we know there won't be a double DQ? To me entering the flag in anticipation is no more appropriate than (say) entering "Williams" into the bracket for the next round if Venus and Serena are drawn against each other.
I expect my removal of this flag to be reverted - for some reason a lot of people are very insistent that it's appropriate, and will justify it in the name of wikipedia tradition. I think it would be a good idea to spell out in the policy that this is a violation of WP:CRYSTAL. I would not expect to see "anticipatory flags" in any professional sports publication. MaxBrowne2 (talk) 00:28, 8 November 2018 (UTC)
It's premature but not likely a massive violation. A couple points is that the matter will be resolved in a few days or so, it seems; it would be different if we were talking months or more out. Second, how many times have both players in such an event have withdrawn? There's a very slim chance, but more than likely in the current case, a match to played by two Chinese players will result in one Chinese person moving on. It's definitely not a change to edit war over, but I would recommend caution in adding it too soon. --Masem (t) 00:46, 8 November 2018 (UTC)
Interesting. I agree with Masem that it's not a huge violation or something to war over, but on the other hand, it seems very clearly WP:CRYSTAL. What's the big hurry about entering data about the future? I can't help thinking about some of the old bbs discussion boards, where a few editors would hasten to respond "First!" in reply to new posts on popular boards without any actual content in their message other than that one word, apparently merely to establish some kind of bragging rights of primacy, or increase their board visibility. A minor violation there, also. Here, it makes me wonder if it's the same people back again. Mathglot (talk) 00:17, 15 November 2018 (UTC)
Those same people now do exactly the same in the comments of youtube videos to the extent that there exist optional filters that prevent those comments from being posted (and possibly block them from commenting at all, I'm not sure). Thryduulf (talk) 09:25, 15 November 2018 (UTC)
I'd like to get a feel for whether the wording of NOTMEMORIAL needs to be more explicit. The first part seems to me to be very clear that it is saying you cannot create articles as memorials: "Subjects of encyclopedia articles must satisfy Wikipedia's notability requirements." Subjects of encyclopedia articles must be notable. Obviously. The perennial confusion over article content vs article subjects, explicitly clarified at WP:NNC, means that notability is about making articles about things, not what goes inside articles which are themselves notable.
Now the second part seems to be getting interpreted as going beyond article subject notability: "Wikipedia is not the place to memorialize deceased friends, relatives, acquaintances, or others who do not meet such requirements." "Wikipedia is not the place" seems to be interpreted as a global, across-the-board policy. The phrase "others who do not meet such requirements." is being interpreted as applying "such requirements", that is, notability, to any listing or mention of the names of anyone who fails WP:GNG or Wikipedia:Notability (people).
You should read the comments at Talk:USS Fitzgerald and MV ACX Crystal collision#WP:NOTMEMORIAL edit -- the editors who believe this applies to article content, not just subjects, have eloquently expressed why they hold that position. They could re-state it here, but for now I only want to see if anyone feels strongly one way or the other about the need for changes in the wording of NOTMEMORIAL.
Is it necessary to clarify this wording so that it is consistent with WP:NNC, and isn't interpreted as going beyond article subjects? Or if it truly does forbid mentioning the name of non-notable people -- well, I find the implications absurd -- but if that's what it really means, then should we reword it to say so more explicitly? --Dennis Bratland (talk) 21:45, 14 November 2018 (UTC)
"There are some common exceptions to the typical notability requirement: ...In a few cases, such as lists of board members or academics holding notable positions, the names of non-notable people may be included in a list that is largely made up of notable people, for the sake of completeness."WP:NOTMEMORIAL is Policy; when someone says, "no can do, that violates NOTMEMORIAL" they're saying it's way out of bounds. Full stop. End of discussion. Pretty much the only straw you have to grasp is WP:IAR if you want to violate policy. Listing non-notable people doesn't violate policy, and it doesn't even violate the WP:LISTPEOPLEguideline, because the guideline itself carves out several exceptions.
I don't mean to generally encourage making extensive lists of non-notable people or other things, only to establish clearly whether or not WP:NOTMEMORIAL has any direct relevance. If the guideline were to totally forbid non-notable names, we couldn't have such large numbers of GAs, FAs, and FLs with such lists, or mentions in prose. --Dennis Bratland (talk) 22:55, 14 November 2018 (UTC)
Questions that have haunted Man through the ages: What does Wikipedia is not a Memorial actually mean?
I've been reading through the talk archives here. I hadn't realized how many times this same exact question has been debated, and how many times even the same arguments on both sides have shot past each other, going back to at least 2006. The wording of the not-memorial section has changed somewhat over the last 14 years, but never has it actually been changed in a way that addresses what is obviously a perennial source of disagreement. It seems to go like this:
As with WP:NNC, WP:NOTMEMORIAL applies to stand alone articles, not the mention of names within articles about otherwise notable topics, or on lists that are otherwise acceptable lists.
But we can't have lists of hundreds of thousands, or millions, of soldiers killed in WWI, etc. There has to be something to discourage that
Occasionally a slight change in wording, but nothing that clarifies the basic ambiguity. Both sides seem certain the wording isn't ambiguous at all, yet they read it to mean two different things.
It's always sad when people die, but Wikipedia is not the place to honor them. We're trying to build a encyclopedia. Of course, you're free to write articles about dead celebrities or other people with notable achievements. (9/11 victims should be entered on the 9/11 memorial wiki).
Leaving aside the ebb and flow of rules for the non-article namespace, the current verison isn't much different in essence:
Subjects of encyclopedia articles must satisfy Wikipedia's notability requirements. Wikipedia is not the place to memorialize deceased friends, relatives, acquaintances, or others who do not meet such requirements.
There hasn't ever been a version of the text that says outright what the two debating sides have been claiming it says:
This applies to both the creation/retention of articles and lists, as well as to the content inside them.
This only applies to the creation/retention of articles and lists, but not to the content inside them.
Instead of changing the text to say one or the other, the debates peter out or end in a deadlock, or compromise on wording that sidesteps the real issue. We're left with what is clearly an ambiguous policy, in spite of the certainty by so many editors that the problem isn't ambiguity, it's that the other sides is wrong. If only someone would make them stop being wrong, everything would be fine.
We really need to add one of the two above sentences. I favor the second one, because I think we can address the real problem by emphasizing that we already discourage lists of 400,000 war victims. WP:LISTPEOPLE discourages it, without forbidding it altogether. It allows exceptions where we may decided to name non-notable casualties, while keeping the principle that we don't intend to make dozens of gargantuan lists. We have many other rules against indiscriminate content, collections of facts that don't really mean anything. Meaning we could have a list of casualties, or articles that name them in whatever format, list or prose, but it still has to have encyclopedic merit. And local consensus might decide against it simply because their editorial judgement doesn't favor an extensive roll call of casualties.
I'm still testing the waters here. History indicates this is a fools errand. Many others before have tried to sort this out and they haven't done it, so it might not even be worth it to make an proposal. --Dennis Bratland (talk) 23:13, 15 November 2018 (UTC)
Its not, and I'll tell you why: no one cares. And do you know how I know that? Because I once flew too close to the sun myself, and when the wiki-heat melted my wings I crashed to the earth, as was foretold. But when I went to try again I was banished for preaching heresy and told my contributions were worth neither my time nor effort because they "didn't belong here". So as one former adventure turned bitter but wiser editor to another still adventurous editor, I'll say this: stop trying, before they crucify you too for the wiki-heresy of thinking that everything belongs here and be content with the small consolations they dole out for those who thought about thinking about being courageous. TomStar81 (Talk) 23:50, 15 November 2018 (UTC)
I can live with adopting the first version, rather than my preferred text. I've gotten pretty used to losing myself. All I'm really hoping for is a solid consensus finally choosing one or the other, and sticking it in policy page so it stops being a battleground. I'm not so sure I'm an inclusionist -- I've spent years in WikiProject Motorcycling deleting any mention of which colors a model came in, unless notable, and demanding any article about model of motorcycle demonstrate it is notable by the same standards as an article about a flavor of soda or brand of soap. --Dennis Bratland (talk) 00:45, 16 November 2018 (UTC)
A list of victims/casualties is not the same as a memorial (="something that keeps remembrance alive" as per Merriam-Webster). A list of victims may not be added to keep the memory of victims but merely for completeness. Within the context of an article's topic, their inclusion is not indiscriminate. -- P 1 9 9✉18:50, 16 November 2018 (UTC)
Interestingly it seems the editors of articles about airline accidents/incidents (plane crashes) dislike listing non-Wikipedia notable victims, but some articles about school shootings and massacres do have victim lists. I wonder if the relative projects should be contacted for this discussion. Perhaps Wikipedia:WikiProject Aviation? WhisperToMe (talk) 08:17, 17 November 2018 (UTC)
Uh I think this discussion is in the wrong place? If there's an article about an event where let's say eight people died... editors may disagree about whether to give their names, but neither side is talking about a memorial in most cases. One side is "Well, this is just trivia and clutters the article" and the other side is like "Well it's data strongly connected with the incident which some readers might want, so it's worth including". (To be honest, it often comes down to "Hey I like to put in victims names" vs. "Well I don't".)
We certainly do include non bluelinked people in article text I think, when it is appropriate. First resident of a town or whatever. In this article I listed four people who helped establish the company; none of them are bluelinked. Is this a "memorial"? I sure didn't intend it as a memorial. Should I have included those names, or is it trivia? I dunno, but either way it has nothing to do with "memorial". The names are written into a paragraph sentence rather than a list. I could have made it a list. Does this format question matter? Why? I mean if the problem is the list format, I guess most of these could be recast as paragraphs. That's be silly IMO, but if that gruntles everyone, that's fine.
It's possible that we don't need to micromanage this. I guess we could have a rule that any two or more people that died at the same time shouldn't be in article. Maybe we should allow ten but not eleven or whatever. My preference is to let the people writing the articles decide.
Anyway there're a lot rules here where people really just read the title and go on that. This happens a lot with NOT NEWS. If people are going to be of the mind "Well, writing in the names memorializes these people, at least to some degree, right? And NOT MEMORIAL is a thing. So the names should go" you can't stop them I guess. All you do it talk it out.
As always, the only question that matters is what best serves the reader. It's reasonable to hold that a list of 1000 doesn't really much serve the reader, I would say. It's too long, too much clutter, unbalances the article to the point where it's maybe harder to read and suss. What about a list of ten? I dunno, but maybe that's different? If somebody's gone to the trouble of creating and ref'ing a paragraph (or list) of that, then isn't the burden on the person who is saying "Well, we have this information, but let's not tell the reader" to make the case of how this improves the reader's experience. Herostratus (talk) 09:24, 17 November 2018 (UTC)
It seems it would make sense to have a limited selection of sayings for people who are particularly well known for their sayings, such as Oscar Wilde or Mark Twain or Dorothy Parker, especially if they're often quoted. In that case, though, I'd expect the article to indicate this to be the case, at least in a lead sentence in the Selected Sayings section, to the effect of "Many remarks made by X, a selection of which appear below, are frequently repeated." This should then be footnoted to sources affirming the oft-quotedness of the person. Otherwise, it seems more like their inclusion reflects admiration by the editor who added them for the person or for the views promoted in the quotations (for example, the religious sentiments expressed by an imam) than that they reflect a general recognition of the person as the source of well-known quotations. Largoplazo (talk) 19:43, 15 November 2018 (UTC)
I agree with both of the above. If someone is particularly noted for their sayings then we should have a sourced statement to this effect in the article and they should have a page at Wikiquote listing them that we should link to. Possibly there should be a short list of examples in the article, see for example Yogi Berra#Legacy which has good presentation but I think we only need at most half that number of examples. The most egregious examples of quotefarms though are the reactions sections of articles (or indeed whole articles) that quote verbatim every tweet by a world leader, semi-significant government official or "celebrity" no matter how relevant to the event the person is or how formulaic their words. Thryduulf (talk) 22:10, 15 November 2018 (UTC)
I suppose exporting the sourced quotations to Wikiquote, removing the unsourced ones and only leaving a link to WQ here should then be fine, right? --HyperGaruda (talk) 19:36, 16 November 2018 (UTC)
If you are referring to Hasan al-Askari#Selected sayings then I would support that. If there are any particularly notable ones that have sourceable commentary then they could be integreated into prose elsewhere in the article as well, but all of the sourced ones should be at Wikiquote. Thryduulf (talk) 17:33, 17 November 2018 (UTC)
A Discussion Concerning Victims Lists Has Been Opened at The Village Pump | Archive
