User:Isaacl/Community/Fostering collaborative behaviour

English Wikipedia's community comprises editors from around the world, offering diverse viewpoints, experiences, and knowledge in service of creating a general encyclopedia with broad coverage. Its success relies on editors working collaboratively to add new content and edit each others work to evolve and improve its articles. To that end, it is essential that the community fosters collaborative behaviour amongst its members. This page seeks to develop and refine ideas to change processes or procedures so they encourage desired behaviour and dissuade poor behaviour.

Ideas for process and procedure changes

There are many cases where poor behaviour works as a strategy to win disputes. It can discourage participation from opponents, leaving the uncollaborative editor to have a disproportionate amount of influence on the outcome. English Wikipedia's processes and procedures should be designed to reward desired behaviour, and discourage poor behaviour, so that selective pressure will increase the amount of desired behaviour in the community.

For example, current content dispute resolution procedures use multi-threaded discussions with little moderation. This gives an advantage to editors who don't collaborate well with others and swamp the conversation with responses, or who try to drive away opponents with aggressive behaviour. A more structured form of dispute resolution with some form of moderation could help mitigate the advantages of poor behaviour, thereby restoring an incentive for desirable behaviour. The goal is to make the need for enforcement of behavioural policies moot, as the process itself makes it a losing strategy to behave poorly.

Ideas

Instructions: Please add a bullet item to the list below with your idea for modifying an English Wikipedia process or procedure so that poor behaviour is a neutral or losing strategy, thereby reducing the need to enforce behavioural policies. If you have specifics in mind, great; the idea can also be fairly high-level during this initial phase. Please hold off on commenting on anyone else's ideas for now; for the sake of argument, assume that everyone's ideas has major flaws at this point that will have to be addressed.

• structured content dispute resolution: follow a structure that discourages repetitive discussion and mitigates the effect of aggressive behaviour. Link to separate page with proposal to come. isaacl (talk) 19:50, 13 August 2019 (UTC)
• active mentorship for new editors: redirect new editors who are not adopting community behavioural standards to a mentor, who would actively work with the editor on improvement and evaluate if the editor is a good fit for English Wikipedia. There are many challenges to this approach, not the least of which is the enormous time investment this would require. isaacl (talk) 20:09, 13 August 2019 (UTC)
• Introduce warnings and maybe a monitoring system for edit warring. Currently vandals get as many as four warnings before we eventually block them, but edit warrers simply get blocked, despite edit warrers typically being goodfaith editors who have got a little over enthusiastic and can usually return to good editing, whilst vandals very very rarely become good editors. The usual argument for this is that we don't have a monitoring system for edit warring in the way that we do for vandalism warnings, but it should be possible to modify twinkle etc. It would also be good to have the option of editor level page protection, so individual pages can be protected against individual editors (just please don't call it page level blocking) ϢereSpielChequers 21:26, 13 August 2019 (UTC)
• When we block someone for a fixed period of time we are expecting them to return to the community. In such circumstances it is important that the block message is clear as to the reason for the block so that the blocked editor knows what they have to change when they return. Fixed term blocks for unclear or undisclosed reasons should be considered a form of toxic behaviour and an abuse of the admin toolset. ϢereSpielChequers 21:26, 13 August 2019 (UTC)
• Seeking to change policy by sanctioning those who follow that policy or are enforcing it should be considered a toxic behaviour and deprecated, even at RFA. If you want to change a policy there are routes such as RFC. ϢereSpielChequers 21:26, 13 August 2019 (UTC)
• Changes in policy should not be retroactive, and conduct should not be judged or punished by standards introduced later. This goes to what other participants have alluded to: the opportunity to change, and also fairness.--Wehwalt (talk) 12:24, 14 August 2019 (UTC)
• Reform all warning templates with the assistance of behavioural psychologists. Ban untemplated warnings. They may escalate conflict unintentionally. This meta-analysis for tobacco warnings "proposes that threat-induced defensive processing does not interfere with the effectiveness of fear-arousing communications but actually contributes to it". --[E.3][chat2][me] 13:36, 14 August 2019 (UTC)
  • "If fear appeals are the wrong instrument to achieve behaviour change, which behaviour change method (or technique) should health promoters use? In a metaanalysis of Internet interventions, the five most powerful behaviour change methods were stress management, general communication skills training, modelling or demonstration of the behaviour, relapse prevention/coping planning and facilitation of social comparison."
  • From motivational interviewing techniques.

1. Seeking to understand the person's frame of reference, particularly via reflective listening
2. Expressing acceptance and affirmation
3. Eliciting and selectively reinforcing the client's own self motivational statements and expressions of problem recognition, concern, desire and intention to change, and ability to change
4. Monitoring the client's degree of readiness to change, and ensuring that resistance is not generated by jumping ahead of the client.
5. Affirming the client's freedom of choice and self-direction

I'm not a psychologist but I think behavioural psychologists and organisational psychologists could have much more to offer. Reform won't come solely from !votes and consensus, it just entrenches pre-existing models and patterns of behaviour IMHO. Many processes but esp WP:AfD need fundamental reform. --[E.3][chat2][me] 13:58, 14 August 2019 (UTC)

• Edit conflicts are one of the main forms of conflict that goodfaith newbies can be expected to encounter. Not everyone learns how to resolve them, I'm not even sure they can be resolved on a smartphone or a screenreader, and I'm pretty sure that not all newbies realise that their edit conflict is really with the computer rather than with the editor they ostensibly have an edit conflict with. Many edit conflicts could be simply and easily resolved by making # and * additional pseudo section delimiters as far as resolving edit conflicts is concerned. So if two people !vote at the same time on a page the software could simply append the second !vote as an additional sentence starting # rather than accept one !vote and reject the other. ϢereSpielChequers 15:01, 14 August 2019 (UTC)
  • Alas, the CAP theorem (which has been formally proven) says that the above solution cannot work in all situations. It is one of the common methods that are used to try to get around CAP theorem, and it does work some of the time. Basically you make the unit of "that which the user is changing" much smaller than it is now (a section). The problem occurs when Editor A changes his comment a bit higher up in the list, at the same time editor B adds a comment that refers to A's old comment, and the software has to reconcile the two. We already have decided to not satisfy C when an edit refers to something in another section, and to not satisfy the "receives a non-error response" part of A in any situation that triggers an edit conflict error. All we can do is decide to not satisfy C less often by not satisfying A more often. Or we can decide to not satisfy A less often by not satisfying C more often. We can't do anything about P, BTW. P is built into the internet. the CAP theorem says that when one have a network that is partitioned (which we always do thanks to the Internet), one has to choose between consistency and availability. --Guy Macon (talk) 03:09, 15 August 2019 (UTC)
    • True you can't resolve all edit conflicts, but you can resolve more than we do now. Including where two people both add a vote at the same time. ϢereSpielChequers 11:10, 15 August 2019 (UTC)
      • As I mentioned in the instructions, let's defer further discussion of the individual ideas for now, to maintain focus on gathering ideas (particularly ones within the designated scope). isaacl (talk) 11:42, 15 August 2019 (UTC)
• Put your ideas down and walk away. Now is not the time to SELL your idea or expand on the initial statement of your idea. That comes later. This is called Brainstorming. ―Buster7 ☎ 20:04, 14 August 2019 (UTC)
• Simple things: more thank you's, apologies, expressions of gratitude for editing demeanor, ignoring of perceived annoyances. ―Buster7 ☎ 20:14, 14 August 2019 (UTC)

• "Try not to annoy people. Try not to be easily annoyed." --Guy Macon (talk) 03:09, 15 August 2019 (UTC)

• Many newbies, especially in the USA are only used to one version of English, and have issues when they find a site such as Wikipedia that has articles in several different versions of English. "Correcting" typos to the wrong version of English is a common newbie behaviour that gets them bitten. There is a better way we could change EN Wiki to make language version a user display option. ϢereSpielChequers 11:05, 15 August 2019 (UTC)

It seems to me this could be automated. It should not be hard to detect edits that consist of changing spelling from on English variant to another and then warn the editor.--agr (talk) 18:06, 15 August 2019 (UTC)

• Upgrade "don't template the regulars" to a strong recommendation and introduce software to warn templaters when they are about to template a regular. The shift from communicating with newbies to templating them is one of the most credible theories for the community going off the boil in 2007, bespoke messaging for regulars and goodfaith editors would be better. ϢereSpielChequers 11:27, 15 August 2019 (UTC)
• Make "thanking" a more public thing. The thanks button was introduced several years ago by the WMF, the intention was presumably to foster collaboration by making it easier to thank people. The problem is that thanking via the thank button is a much more private process than the previous systems of barnstars and talkpage messages that it has somewhat supplanted, so it may have masked or hidden the amount of collaboration on the site. When you go to the talkpage of a heavily thanked editor you may think them less friendly and part of the community than if you looked at their talkpage before the thanks button came in. If we want to foster collaboration rather than hide it we should make the thanking process more visible. This could be done by userboxes and or display options that allow editors to display how many times they have been thanked and by how many different editors or see that, articles in the Signpost on "most thanked" editors and options making it easy for editors to get a list of people who've recently thanked them when updating their Christmas card lists. ϢereSpielChequers 11:33, 17 August 2019 (UTC)
• Have a specific page where users can report uncivilised/intimidating behaviour, and include a link to that page on the welcome message posted on the talk pages of newcomers. This way, the community can look at the reported user's behaviour and determine whether or not the reporter's complaints are warranted. However, we should also have a system in place to ensure that this process is not abused, for instance a systems of blocks for users who constantly make unwarranted complaints. The dog2 (talk) 02:30, 18 August 2019 (UTC)
• When a user makes a report against an old timer, no one may introduce BOOMERANG claims against the reporter other than the reported user. Once the reported user has made such claims other users evaluate them. An exception may be made for suspected sockpuppetry reported at SPI.
• Have a cadre of "Fair Witnesses" monitoring a specific page where any user can report uncivilised/intimidating behaviour (perhaps anonymously) that they observe. A "Fair Witness" can then go to the article and, with the express purpose of being a "peace broker" and, secondarily, a spokesperson for the newbie, facilitate a change in dynamics, easing the dis-ease.
• Consensus should be followed and not only handled by admins. Admins should not be the judge, jury, and executioner of all blocks as everyone has a different point of view and it should be seen that way and a consensus reached without linking to a million articles and essays on what is expected, especially essays which only reflect the view of a select few.

... add your idea here ...

• My idea is that all of the above suggestions should be deployed ASAP!! Atsme 💬 📧 21:22, 13 September 2022 (UTC)

Analysis

Improving response to edit warring

See User talk:Isaacl/Community/Fostering collaborative behaviour § Improve response to edit warring for discussion.

Improving guidance for communicating reasons for blocking a user

See User talk:Isaacl/Community/Fostering collaborative behaviour § Communicating reasons for blocking for discussion.

Increasing visibility of thank you messages

See User talk:Isaacl/Community/Fostering collaborative behaviour § Increasing visibility of thank you messages for discussion.

Structured content dispute resolution

I have written a draft of some techniques to mitigate some of the difficulties with managing mostly unmoderated online discussions in a large group across many time zones: User:Isaacl/Community/Content dispute resolution toolbox. The three ideas I have listed are round-robin discussion phase, pros and cons summary of options, and revisit respite. Please feel free to discuss these techniques on the corresponding talk page. isaacl (talk) 23:07, 16 September 2020 (UTC)

Active mentorship for new editors

Some previous statements on ideas for active membership, including significant barriers:

... perhaps there could be a new user patrol who could redirect editors who show potential but need guidance to active mentors, and weed out those who are going nowhere fast. Because this is an idea I thought of in two minutes, I know it has possibly insurmountable drawbacks such as having to enact additional bureaucracy to monitor the patrolling group and the mentors, and probably having to hire mentors due to the immense amount of time required by active mentorship. Nonetheless, we need to start thinking about new approaches like these in search of a way to minimize as much as possible the incentives for poor behaviour.

...

Should the editor eventually show no promise at adapting [to the Wikipedia community's principles and guidance], the mentor would have to tell the editor that sadly they don't seem suited to contribute to Wikipedia.
— from comment 1 and comment 2 located at Wikipedia:Community response to the Wikimedia Foundation's ban of Fram/Archive 13