Jump to content

Wikipedia talk:Requests for comment/Harassment solutions

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

How can we best address harassment and abuse of editors?

Harassment problems

[edit]

Taxonomy of controversies

[edit]

These kinds of controversies can lead to harassment issues:

  • Caste
  • Disability rights
I am not allowed to talk about this. EllenCT (talk) 07:01, 25 August 2020 (UTC)[reply]
The following discussion has been closed. Please do not modify it.
  • Economic
  • Ethnic
  • Gender
  • Medical
  • National
  • Political
  • Personal
  • Philosophical
  • Religious
  • Scientific

Manifestations of harassment on Wikipedia

[edit]

include but are not limited to:

Specific instances: request for examples

[edit]

Harassment issues which you feel have not been adequately addressed or are ongoing issues affecting accuracy:

  • Example: those affiliated with different sides of a controversy have been causing inaccurate articles on ______.
  • Those trying to promote a particular product or service are harassing editors trying to maintain WP:NPOV.
  • ____
  • ____
Positive examples

We should probably give examples of harassment that have been dealt with OK. An example from WP:AN/I is https://en.wikipedia.org/w/index.php?title=User_talk:Persia&diff=prev&oldid=969240089 where the remedy is for the recipient of the message is to remove the post from their talk page. After removal, I don't know if everyone is happy with the result. (Particularly offensive postings could be rev-delled or even oversighted). Graeme Bartlett (talk) 06:15, 25 July 2020 (UTC)[reply]

Excellent! Other examples of really good resolutions to harassment issues? EllenCT (talk) 14:50, 29 July 2020 (UTC)[reply]
  • ____
  • ____

RFC on solutions

[edit]

{{rfc}}

How can we best address harassment and abuse of editors? Since this RfC has sub-proposals, I ask that it run for 45 days. EllenCT (talk) 18:55, 7 August 2020 (UTC)[reply]

Brainstorming constructive solutions

[edit]

Proposals in this section are not put forward for consideration, but only alternative proposals and constructive criticism (e.g., "yes, and....") is allowed. Please boldly add comments to a proposal at any stage of development, but if you can't find anything nice to say, then try to make an additional, new, better solution. Please allow days to weeks for individual proposals to be developed. EllenCT (talk) 19:43, 22 July 2020 (UTC)[reply]

Requiring that the Terms of Use include a Code of Conduct with friendly space provisions

[edit]

We need some sort of definition of "friendly space" before we can reasonably comment here. If it means that nobody should ever tell anyone else that they are wrong then I would be vehemently opposed. Phil Bridger (talk) 19:49, 22 July 2020 (UTC)[reply]

I agree. I know there is a lot of work on drafting these sorts of things, but every time I've asked people for a draft in the previous month and before, they've asked me to wait. I'm generally optimistic, because I can't wait for the hegemony. I'm going to draft the friendliest friendly space policy known to emotively spatial denizens.
Example existing code of conduct provisions
The following discussion has been closed. Please do not modify it.


EllenCT (talk) 19:46, 24 July 2020 (UTC)[reply]
@Phil Bridger: how do you feel about the prohibition in meta:Celtic Knot Conference 2020/Friendly Space Policy#Behaviour that will not be tolerated against "unwanted photography or recording"? EllenCT (talk) 22:35, 23 July 2020 (UTC)[reply]
I was unaware that the possibility existed of anyone photographing me or recording me over the Internet while I am editing Wikipedia. Phil Bridger (talk) 07:48, 24 July 2020 (UTC)[reply]
Ha! I plan to make a table of prohibitions, so that the various friendly space policies can be compared. EllenCT (talk) 18:33, 24 July 2020 (UTC)[reply]
You might also be interested in User:Lightbreather/Kaffeeklatsch, which was an attempt at a safe space for female-identifying editors. It largely failed when the creator was sitebanned (unrelated), and also due to rampant misogyny, but maybe that's improved in 2020? (don't @ me, I know it hasn't) It had been deleted at the creator's request before she was banned, but it was restored not long afterwards at someone else's request. I was about to tag it with {{historical}} since it hasn't been substantively edited since 20 March 2015, but in the spirit of the original space it would be better for a female-identifying editor to take care of that. Ivanvector (Talk/Edits) 18:58, 24 July 2020 (UTC)[reply]
I'm hoping to propose general solutions transcending all the different kinds of harassment issues. A general monitoring system to illustrate apparent strategy and tactics of any set of editors suspected of perpetuating bias might be a toolset we could build later on in this area. I'm reluctant because something like that could be used to exaggerate evidence and could turn into a weapon in arms races, so it's probably better to keep humans in that loop. Maybe I can wait for the hegemony. EllenCT (talk) 20:58, 24 July 2020 (UTC)[reply]
Foundation interview videos
[edit]

[1],

[2],

[3]. EllenCT (talk) 07:29, 25 August 2020 (UTC)[reply]

Requiring that the Foundation adhere to the projects' terms
[edit]

Here is meta:Talk:Strategy/Wikimedia movement/2018-20/Recommendations/Iteration 1/Diversity/9#Terms of Use for the WMF by de:Benutzer:Tinz which was approved around 45-0:

I agree that it has to be a two-way street, and I think this says that well. EllenCT (talk) 16:05, 31 July 2020 (UTC)[reply]

Petition for remedy of repeated paid advocacy: report card on enforcement progress? Partnerships with freelance brokerages?

[edit]
Many if not most may see me topic banned from this request. EllenCT (talk) 07:03, 25 August 2020 (UTC)[reply]
The following discussion has been closed. Please do not modify it.

I'd like to see a thermometer-style meter template describing how many organized advocacy attempts have been uncovered in the previous month, perhaps in comparison to whether they are still active. EllenCT (talk) 21:52, 23 July 2020 (UTC)[reply]

Like I said below, I think this proposal is out of scope for a discussion on solutions to harassment. Unless the point is to prevent harassment of the paid editors? To the point, though: this would be difficult to measure. Do we go by how many editors are blocked for undisclosed paid editing? Maybe we count WP:G11 deletions? And I also don't know how we would track whether or not the operations are still active: in the present era it seems the large firms have been supplanted by individual freelancers working through sites like Fiverr and Upwork, though it may also be that the firms are just outsourcing the work. Ivanvector (Talk/Edits) 19:02, 24 July 2020 (UTC)[reply]
@Ivanvector: Are you saying that corporations can't harass individuals? I know someone who works at Upwork.[4] EllenCT (talk) 19:55, 24 July 2020 (UTC)[reply]
Not at all. I'm saying I don't understand how the issue with paid advocacy ties in with harassment of editors on Wikipedia. In my mind, someone (be it a person or a group or a corporation) who comes here seeking to create a promotional article in exchange for payment from a third party may be violating our terms of service, but they're not inherently harassing anyone. Harassment by our policy's definition is "a pattern of repeated offensive behaviour that appears to a reasonable observer to intentionally target a specific person or persons [...] to make the target feel threatened or intimidated." I don't see how paid advocacy, on its own, meets this definition. Perhaps there is a phenomenon of paid editors targeting specific Wikipedia editors to intimidate them that I'm just not familiar with? Ivanvector (Talk/Edits) 17:10, 27 July 2020 (UTC)[reply]
It absolutely happens. We missed the tobacco splurge, but have you seen the history of our articles on fracking, or climate change in general?
I am likely topic banned from this observation. EllenCT (talk) 07:06, 25 August 2020 (UTC)[reply]
The following discussion has been closed. Please do not modify it.
We've had admitted paid advocacy factions openly battling on each each side for vaping, too. There are so many more examples, but I've asked people to list their own in the "request for examples" above. EllenCT (talk) 14:39, 29 July 2020 (UTC)[reply]

WikiProject-specific academic partnerships for studies of bias

[edit]

Many of these already occur, thanks to researchers being members of the WikiProjects they are interested in. However, in some topics the factions dominant on Wikipedia are not the factions dominant in the field. More often than not, Wikipedia has made the correct choice in such cases. This is because astroturf is a thing, and monied interests are capable of biasing the literature in many cases. EllenCT (talk) 21:55, 23 July 2020 (UTC)[reply]

The problem of bias is very different to harassment. Every editor has some kind of bias, so we cannot count editing with bias in itself as harassment. However a biased editor may be harassing others, eg by inappropriately nominating a biased class of articles for deletion. Extremely zealous campaigners are more likely to be at risk of harassment. I think that harassment solutions will be covering different kinds of solutions to dealing with systematic bias, otherwise the scope here will be too big. Graeme Bartlett (talk) 23:57, 24 July 2020 (UTC)[reply]
Bias harasses the readers, which reflects poorly on the editors and thus harasses them indirectly. I can see no bright line distinction between the sorts of organized advocacy to mislead for political and economic purposes. EllenCT (talk) 14:36, 29 July 2020 (UTC)[reply]

Ask Foundation to commission independent studies of bias on controversies resulting in harassment

[edit]

In the few of these I know about (and I don't think there have been many more, but I've lost track of everything at all the chapters) it's not clear the Foundation intentionally paid for the initial work or simply added funds to expand what otherwise would have been an insufficiently sized study, covering total costs in one case. I'm still studying these, with the caveat that I'm not exactly sure how to define them because some activities may qualify marginally.

My thoughts on this are, let's figure out how to draft the request for proposals and then figure out who can administer and pay for it most effectively. Maybe a grad student needs a Ph.D. somewhere and some other foundation would pay their stipend and tuition if they took it up instead of the Foundation. EllenCT (talk) 20:53, 24 July 2020 (UTC)[reply]

Explore different kinds of partnerships with for-profit, non-profit, non-governmental, and governmental entities

[edit]

Instead of asking the Foundation for help with bias issues, what if there was a way to get for-profit companies to sponsor a fund that pays for such studies? EllenCT (talk) 22:12, 23 July 2020 (UTC)[reply]

Pilot projects to automate peer review-like processes to monitor bias on controversial topics

[edit]

Can we leverage the data in WP:BACKLOG for this? EllenCT (talk) 22:12, 23 July 2020 (UTC)[reply]

Perhaps this could show up tag bombing against particular users' pages. But there will still be a distinction between harassment, over enthusiasm and identifying valid problems where a human would have to assess. Most of this backlog would not be in controversial topics. Controversial topic pages would generate a lot more loss of temper and overreaction, but also would have a lot of editors, so the pages probably won't stay in any backlog for long. Perhaps the pages marked with arbcom sanctions could somehow be monitored with some automation. But how is the harassment manifest? Perhaps edit summaries, reverting or undoing a particular editor's work, stalking, or inserting text that is harassment. Graeme Bartlett (talk) 00:33, 25 July 2020 (UTC)[reply]
I was suggesting using the conflict article templates, but you suggest those need to be supplemented with information from the user talk page warning templates, as well. I'm working on it, but ... slowly. EllenCT (talk) 23:45, 25 July 2020 (UTC)[reply]
Examples
[edit]

These are roughly in reverse chronological order for now:

I will need some time to complete this list. EllenCT (talk) 18:38, 5 August 2020 (UTC)[reply]

Sanctions for specific behavior or patterns of behavior which perpetuate bias

[edit]

I added this section because the RfC that inspired this one was being run by Arbcom, but I'm not sure it's necessary. If people just edit normally to improve accuracy while aware of harassment issues, that will likely work even better than sanctions. Living well is the best revenge. EllenCT (talk) 22:02, 23 July 2020 (UTC)[reply]

This strikes me as sort of a "don't feed the trolls" argument, which in context of harassment is a problematic form of victim blaming. Should you not react to harassment? Maybe, but that doesn't forgive our duty as a community to create an environment free of it to the best of our ability. That most certainly includes showing problematic editors the door. Ivanvector (Talk/Edits) 19:06, 24 July 2020 (UTC)[reply]
Agreed; I meant, I doubt whether there is any need for new sanctioning proposals as long as we get the sunlight we need in the right places to figure out who best to sanction under existing rules. EllenCT (talk) 20:50, 24 July 2020 (UTC)[reply]
By introducing more sanctions it is also risking making a more toxic editing environment, that enables more threats to be made, whether justified or not. I think we have to concentrate on stopping the harassment, rather than trying to identify behavior which perpetuate bias. For example an editor has only ever created articles on men. Is this perpetuating systematic bias of Wikipedia? Probably yes, but in it self the work is not bad. Instead we have to make the place more welcome for people that want to edit in other ways that fill in the holes in Wikipedia. Note that Wikipedia will be "biased" due to the presence of bias in reliable published material. I have found that classing some kinds of sources as unreliable (eg magazines for women) results in a deficiency of certain topics here. But I think it is outside the scope for harassment solutions. Graeme Bartlett (talk) 00:06, 25 July 2020 (UTC)[reply]
I think, in a roundabout way, it's not out of scope. A very significant problem for Wikipedia is systemic bias, which is unintentionally perpetuated because (as WP:SYSTEMICBIAS says) we tend to attract an average editor who fits a specific demographic, and that demographic tends to be socially privileged, and that creates a sort of power imbalance in the editing environment. SYSTEMICBIAS also cites reliable statistics that female-identifying editors tend to leave Wikipedia much earlier and much more frequently than male editors, and of course they do: we're quite a hostile environment to anyone who doesn't have the average editor's experience of oppressive power structures, and the average editor is nearly always on top in those imbalances. Many anti-harassment discussions I've participated in have had a significant proportion of (male) editors opposing changes for reasons similar to "you have to expect some abuse on the internet", and that's just so wrong if we don't want this to be the White American Male Wikipedia. "Getting sunlight in the right places" is a very good analogy, I'll use civility enforcement as an example: all too often the community accepts gross incivility from editors deemed to be "doing the right thing" (defending articles from POV pushers and paid editors, usually) but this attitude of accepting the "right rudeness" just perpetuates the toxic environment that keeps potentially valuable editors away, and editor retention is really our most serious existential threat. I know a number of female-identifying academics who I've approached over the years for input on articles in their areas of expertise, and nearly every one has declined the invitation specifically because of the personal and professional risk of exposing themselves to Wikipedia's toxic editing community. The other related problem is that we let too many editors get away with too many things until their behaviour gets so bad we have no choice but to ban them, and they take a bunch of people with them because those discussions are so dramatic and factionalized. This was a lot of words I know, but to come back around to your point: we can't make the place more welcome for people who want to fill in the holes if we continue shrugging at these serious problems. Ivanvector (Talk/Edits) 17:48, 27 July 2020 (UTC)[reply]

For consideration

[edit]

Discussions may include critiques of a not necessarily constructive nature (e.g., "2nd choice," etc.)

Proposal A: _____

[edit]
Discussion
[edit]
[Support/Oppose/etc.]
[edit]

Proposal B: _____

[edit]
Discussion
[edit]
[Support/Oppose/etc.]
[edit]

General discussion

[edit]

Sorry for creating a new section that might break your structure; please feel free to move my comment to whichever section you think is most appropriate, but I did not see an obvious place for general comments. It seems to me that many of the items under "constructive solutions" are not addressing harassment directly or at all. For example, a code of conduct in the terms of use is a good thing to discuss in terms of addressing ongoing issues with harassment, I'm excited to see that discussion. However, a number of these go into detail on studying and perpetuating bias, and while I can see a link between "Wikipedia suffers systemic bias" and "Wikipedia users suffer harassment", it's a very weak and tangential link, at least to me. We don't solve harassment by solving bias; while some conflicts over perceived or systemic bias may lead to harassment, we could say the same for any sort of conflict. Then we get to the points on remedying paid editing and partnerships with for-profit entities, which just aren't related to harassment at all. They're issues for sure, but we're not trying to solve everything here. Maybe I just can't tell where you're going with those proposals but it will become apparent after further discussion? Ivanvector (Talk/Edits) 19:43, 22 July 2020 (UTC)[reply]

Yes, I do intend to add other proposals I've been thinking of and specify the ones that I've already listed. Thank you for making this section. EllenCT (talk) 01:39, 23 July 2020 (UTC)[reply]
  • Invalid RfC. The statement, whilst brief, tells us absolutely nothing about the issue and merely requests a change to the normal duration. This is how it looks to the outside world. Please observe WP:RFCBRIEF. --Redrose64 🌹 (talk) 07:35, 23 July 2020 (UTC)[reply]
  • Invalid RfC(Invited by the bot) I agree with the above comments that this is so confused and is missing so much that it is currently invalid. Further, I think that it is overall mis-cast. Abuse of editors is certainly a huge problem in Wikipedia which needs fixing. Casting is as harassment of editors due to the listed factors is a misfire. Suggesting that you start over. And rename it for the actual problem which is abuse of editors. North8000 (talk) 11:08, 23 July 2020 (UTC)[reply]
Good idea; thank you! EllenCT (talk) 21:48, 23 July 2020 (UTC)[reply]
 Done Thank you. EllenCT (talk) 21:48, 23 July 2020 (UTC)[reply]

Delisting pending resolution of WP:AHRFC

[edit]
 Done Thanks! I like your style. EllenCT (talk) 19:10, 24 July 2020 (UTC)[reply]
@L235: Kevin, are there any reasons this should remain delisted pending closure of the private evidence RfC? EllenCT (talk) 03:08, 4 August 2020 (UTC)[reply]
@EllenCT: Hi Ellen, thanks for the ping. There are a couple reasons why I think this should be delisted pending resolution of the RfC:
  • The outcome of the ArbCom RfC might change the baseline of what participants perceive as the current state of policy, which will affect what kinds of proposals they will support.
  • The outcome of the ArbCom RfC might spur further community action (such as a further clarification of policy) that should be part of this RfC, which means we should wait until those ideas are collected before launching this RfC.
  • Quite aside from the outcome of the ArbCom RfC, I worry that this RfC has too much going on – that it has too many independent parts that aren't strongly related to each other or to harassment. Surely all of them have some relationship with harassment, but there are hundreds of proposals that are more related to harassment than e.g. Pilot projects to automate peer review-like processes to monitor bias on controversial topics. I understand that you're trying to start a conversation here and this is just a first step, but I wonder if this is not focused enough to result in a meaningful consensus. I'd be happy to talk more personally about this concern, even over a Zoom call or something, if you'd like.
Best, Kevin (aka L235 · t · c) 06:38, 5 August 2020 (UTC)[reply]
@L235: Thanks, Kevin. I think the first two reasons are completely reasonable, and agree with aspects of the third. I don't want to focus too tightly on either economically or politically motivated harassment to the detriment of treating both, but I {{hat}}-ed the example Codes of Conduct as there is a new committee trying to draft a universal version. I still feel strongly that I should make the table of prohibitions to review whether any of those which may be promulgated could affect non-in person editing with Scunthorpe effects and the like. I have added some examples of review-like processes and am adding more. EllenCT (talk) 19:20, 5 August 2020 (UTC)[reply]