This page uses content from Wikipedia and is licensed under CC BY-SA.

Wikipedia:Village pump (idea lab)

 Policy Technical Proposals Idea lab Miscellaneous 
The idea lab section of the village pump is a place where new ideas or suggestions on general Wikipedia issues can be incubated, for later submission for consensus discussion at Village pump (proposals). Try to be creative and positive when commenting on ideas.
Before creating a new section, please note:

Before commenting, note:

  • This page is not for consensus polling. Stalwart "Oppose" and "Support" comments generally have no place here. Instead, discuss ideas and suggest variations on them.
  • Wondering whether someone already had this idea? Search the archives below, and look through Wikipedia:Perennial proposals.
« Older discussions, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25


Problem behaviors

Wikipedia is generally a wonderful place to write and connect. Problem solving techniques such as Rfc's and third-party really do work pretty well most of the time. But there are some things--some people--that do seem to fall through the cracks fairly dependably. I would like to see a 'Be nice-Be respectful' policy that when violated can be reported, and if nothing else, that Wikipedia would keep track of the number of violations the same person gets over and over, so I would like to see a policy for consequences for repeatedly biting, not only newcomers, but anyone that disagrees with them. People just get away with that here because it isn't about consensus on content. I have seen more than one editor driven completely off Wikipedia because of personal attacks, slanders, insults, and various kinds of bad behavior by the same person. Nothing ever happens about it. I think that's wrong. Something should happen. It violates Wikipedia's stated policy and that policy should be better enforced. Jenhawk777 (talk) 22:56, 29 April 2018 (UTC)

  • We do have Wikipedia: Etiquette but I am not sure what happens to Wikipedians who violate the policies listed here. Vorbee (talk) 17:00, 30 April 2018 (UTC)
I don't know about 'solutions'; but I and a few others are actively exploring if we can figure out if we can make it easier for admins and others to 1. find problematic comments (assuming that's part of the challenge): [meta.wikimedia.org] Feedback very welcome. Some more ideas we've experimented with some are at [conversationai.github.io] Iislucas (talk) 18:44, 24 May 2018 (UTC)
Nothing. Nothing happens to them. They move on claiming they do it all for the good of Wikipedia. That's the problem. Instant reverts, threats, insults based on ideology, point of view, differences of opinion, not based on consensus--things where one person is not clearly "right"--but the one person is asserting their "position" by domineering and intimidating. Nothing happens. They get you to leave. That's all that happens--at least that is all I have seen so far. I have been on Wikipedia about a year and a half, I have observed this one editor have just over a dozen of these types of conflicts, and people repeatedly contact Admin. about him and nothing happens. He seems bullet-proof. And that's just wrong. What are the chances more than a dozen people are in the wrong instead of him? Wikipedia needs to do a better job at this. What are the options? I would take any improvement at all. Jenhawk777 (talk) 22:54, 30 April 2018 (UTC)
Behaving like a dick can, and does occasionally get people blocked (you've probably already seen Wikipedia:Civility#Dealing with incivility). If the problematic behaviour has been gross, and you have recent examples of it that you're willing to present succinctly, and with diffs, then you can start a thread at WP:ANI. – Uanfala (talk) 23:11, 30 April 2018 (UTC)
I suppose I can start keeping a record since Wikipedia doesn't, but I am afraid of retribution if nothing comes of it. This person is vindictive. I was looking for a more proactive approach from Wikipedia.Jenhawk777 (talk) 16:00, 1 May 2018 (UTC)
this is an example right now. Not sure what idea can come out of this thread though; I would like to see civility enforced more, however we do already have a policy (two, WP:NPA and WP:CIVILITY) which are in the range of "Be nice-Be respectful' policy". I think it is not as much as there being a lack of policy but people unwilling to enforce for whatever reason. Galobtter (pingó mió) 16:07, 1 May 2018 (UTC)
Thank you for acknowledging the issue is real. Yes, we have perfectly good policy, but what good is policy without enforcement? I completely agree that is the problem. Disputes over content are solvable. Wikipedia cares about its content and has made good provision for multiple pathways toward resolution. Not so with personal abuses and attacks. Wikipedia does not keep a record of how often any individual gets called for a mediation dispute or watch for other signs of problem behavior and as far as I know, there is no special path for reporting that particular kind of problem--and certainly no enforcement of the policies we have. I simply want that person to stop. I personally have no ability to enforce the minimum behavior requirements of the larger group--that we all agree to--onto the group's few misanthropes. I can't see how this can be improved without some kind of policy change. Jenhawk777 (talk) 05:39, 2 May 2018 (UTC)
────────────────────────────────────────────────────────────────────────────────────────────────────
Yes, the issue is very real. Extreme perpetrators can be reported at WP:ANI and its related noticeboards, where administrators will decide what to do about it. But usually, rudeness is tolerated and you have to be grossly insulting or disruptive to be sanctioned. The reporting process is also quite bureaucratic and needs good understanding of it to be effective: where else would someone reporting abuse be told, "You haven't reported it properly, so we will ignore you"? I believe that this sad state of affairs is one of the biggest reasons why our community struggles to keep high-quality editors. You and I are not the only ones to have brought it up in one forum or another before now, but as long as the majority of the more vocal editors (especially administrators) are prepared to tolerate it, nothing will be done.
@Jenhawk777: However, from what you say it sounds like you have an abuser who may be overstepping the mark. If you drop me a private email with their username (let us know here if you need help with that), I will take a look and see if anything can be done — Cheers, Steelpillow (Talk) 14:38, 3 May 2018 (UTC) [updated 14:49, 3 May 2018 (UTC)]
DID overstep--not happening right now; we are no longer working on the same article. I am so grateful for the offer, but it just isn't enough to protect myself in one instance. Everyone should be protected. I have watched this person run several people off of Wikipedia. If it hadn't been for one of the good ones stepping in and saying, 'come work with me over here in a corner for awhile', I wouldn't have known Wikipedia could actually be rewarding and fun. I have tried to do that for some of the others, but they are so discouraged from the experience--and the fact it seems to them that no one cares--that they just leave. I know there are people here who care. The responses here are evidence. But Admin needs to do a better job at this. It is a problem for Wikipedia even if they don't recognize it. When people go away in this manner, they say bad things about Wikipedia. And frankly, there isn't an endless supply of people who want to write for free who are willing to put in the time to develop enough experience to actually be any good at it. Even factoring in inevitable losses, this should still be seen as an issue. Thank you again for the kind offer, but what I really need is a suggestion for a workable policy change. Jenhawk777 (talk) 18:42, 3 May 2018 (UTC)
Policy is not at issue. The problem is enforcement, it is too hard to get things done. For reasons I cannot understand, the further up people are in the hierarchy, the less they seem to want to recognise that. This is about winning hearts and minds, not policy. And sadly, I am no good at that. — Cheers, Steelpillow (Talk) 19:17, 3 May 2018 (UTC)
Well, apparently, I am even worse--I can't even get agreement here among people who actually agree with me. :-) Whatever the problem is, those with some actual influence need to act. Jenhawk777 (talk) 07:29, 7 May 2018 (UTC)
Agree to a point. But I don't think enforcement will work, for exactly the reasons you say. See comments below. Andrewa (talk) 21:26, 7 May 2018 (UTC)

Following the threads of this section I came across the Wikipedia:Kindness Campaign to which I've now signed up and which I recommend. Perhaps just promoting this WikiProject is what is needed.

I certainly think there's a problem. We cannot expect to attract and keep new editors, and particularly the sort of editors on which Wikipedia depends, in the current environment.

And I think something specifically focused on restoring no personal attacks to general acceptance might be the key here. See User:Andrewa/gentle editor for some of my ideas on that, and comments on its talk page or (far better) here (or even both) of course very welcome!

There has been no consensus to abandon (or even modify) NPA, but it seems to have happened anyway. I guess the other possibility is an RfC to modify or abandon the policy, and regard these behaviors as acceptable, but I myself believe that would be the beginning of the end for Wikipedia. I could be wrong. Andrewa (talk) 05:51, 9 May 2018 (UTC) Andrewa (talk) 21:26, 7 May 2018 (UTC)

Okay, so if enforcement isn't the answer--what is? I am a member of Wikipedia:Kindness Campaign and practice that--even with the person who kept attacking me--they criticized me for my "excessive civility" too! I don't think joining up is high on their list. You know, I need to add here that most of the people on Wikipedia are great--helpful, patient, kind--but when there is one whose behavior is so egregious, for so long, it can color everything. I'm trying not to let that happen. That's why I'm here. We have a lot of different kinds of people here with lots of different views and need to treat everyone with respect even if they have the audacity not to think like we do! Perhaps this is a personality thing that can't be fixed. IDK. I admit I'm feeling a little discouraged about all this. Jenhawk777 (talk) 20:33, 9 May 2018 (UTC)
I think we need to demonstrate (and perhaps first build) consensus among Wikipedians in general and admins in particular and perhaps even within ARBCOM that NPA is important. Enforcement might follow in some cases, but demonstrating that consensus might be enough without enforcement being necessary, and without demonstrating that consensus enforcement will not help, IMO, and won't happen anyway. Interested in other ideas, and ideas as to how to best do this. I've linked above to my own best attempts so far.
The only other possibility I can think of is an appeal to the founder. I'm almost concerned enough to give that a go.
Hang in there. If enough people give up on NPA, IMO it's the end of Wikipedia. Hard to imagine? Where did Kodak go?
If it did happen, the world would not end, and thanks to copyleft neither would most of our work so far. Citizendium (which ruthlessly enforces NPA) or another fork (well, currently it's not strictly a fork, but might become one again) would take over. But far better to fix Wikipedia IMO. Andrewa (talk) 22:54, 9 May 2018 (UTC)
We've been trying to build said consensus pretty much continuously for the 5 years I've been around, with no success. We are self-governed, and those participating in the self-governance are a self-selected few who are not representative of the whole. The way we decide things is not likely to change in the foreseeable future.
Over the years I have watched the ongoing debate and given the problem much thought, and I've come to believe that (1) it is intractable in the current environment, and (2) the only hope is through gradual attrition and evolution. As stated, the policy is already in place, and it is routinely ignored with various rationales for ignoring it. Any new initiative to stop ignoring it would fail for the same reasons as the many others that have come before; nothing has changed sufficiently to make the difference. More at the essay WP:DISRESPECT.
The only other possibility I can think of is an appeal to the founder. Good luck. The founder no doubt has his opinions, but they don't carry any more weight than those of any other editor. Those opposing stricter enforcement of behavior policy are not going to withdraw their opposition because of a statement by him.
I promise you that this thread is a dead end and a waste of time. ―Mandruss  23:26, 9 May 2018 (UTC)
Cancelled process mini.svg
I guess we sometimes need to dismiss what's not possible, but the purpose of this page is to incubate and encourage new ideas rather than assess them.
Jimbo is the only member of the founder user group, and has some other privileges as well. He's been understandably reluctant to use them but they are still there for the moment at least.
Agree that Those opposing stricter enforcement of behavior policy are not going to withdraw their opposition because of a statement by him. I'm one of them, so I should know! I'm hoping we can find a more effective way.
In fact one of the key problems I see is the common assumption (which I think you may be making too) that the only way to encourage adherence to NPA is by stricter enforcement. My hope for this thread is that we can brainstorm some other ways.
The other key mistake is to assume that violations of NPA are also violations of civility. In fact NPA is far, far broader then that. That's probably where I think ARBCOM and WP/ANI (on which I lurk from time to time) have gone wrong... once we give up on NPA and fall back to mere civility, we tend to fall back from encouragement to enforcement too. Andrewa (talk) 02:02, 10 May 2018 (UTC)
Mandruss--ouch. I hope you're wrong--but I'm afraid you're right. This could be a waste of time, but as I see it, we can't know till we've spent it. I have to try. I love Wikipedia--but using a colorful but descriptive metaphor--I think it keeps stepping on its own foreskin.
I went and read WP:DISRESPECT since I had never seen it before. Forgive my straightforwardness at this point, but in my POV, that is not a good or helpful article. It's not hard to identify disrespect--anyone on the receiving end of it can do so. I was recently reading an article on the neuroscience of making friends; basically, treating people with respect boils down to being as cooperative as possible and as pleasant as possible: correct others as you would like to receive correction. That's pretty much it. It's not difficult to understand, and there is no value that I can see in trying to make it more complicated than it is. If someone feels disrespected that should be addressed; period. It should be that simple.
Perhaps I simply haven't been around long enough and I don't understand how complicated this problem can become. Even if we used the same steps for personal attacks that we have for content disputes--what would be the end goal? To force an apology? No, that would not only never work--it would be disrespectful! But if there is no recognition of violation of existing policy, and there is no clear consequence--something like the three revert rule--three attacks in a row and you're blocked--then in my view this is not a real policy--this is just hypocrisy. We either stand up for what we claim to believe in or we don't. If we don't, let's take down the policy and admit it's a free for all here. Jenhawk777 (talk) 04:11, 10 May 2018 (UTC)
Perhaps I simply haven't been around long enough and I don't understand how complicated this problem can become. Perhaps. ―Mandruss  05:10, 10 May 2018 (UTC)
Well, I've been around for a while, and I think you're hitting some nails squarely on the head. But yes, it's complicated. We're not going to make Wikipedia perfect here, but we can make progress IMO.
WP:DISRESPECT is an essay not an article, and it's not obvious to me how much of it is the opinion of the person citing it but they're one of three contributors and mentioned by the creator. I don't find it helpful either, but one trap to avoid is assuming that if you treat others the way you want to be treated, they'll be happy. They may not want to be treated that way just because you do! See this off-wiki essay of mine. So that essay is well worth reading if only to understand the other mindset.
And that's the problem with having something like the 3RR for personal attacks. What's good vigorous discussion for one person can be offensive to another. That's one reason that NPA is so sweeping. Any idiot can understand it, and most of them do. (;->
You might also find User:Andrewa/Rules, rules, rules helpful. It's another essay, quite recent, trying to point out how radical and interrelated some of our rules are. Or wp:creed is another of mine, older but a favourite.
Hang in there! Andrewa (talk) 06:27, 10 May 2018 (UTC)
I wrote the essay two years ago and it has been on my user page since then. Recently others decided it should be in WP space. No, it's not my opinion, it's my understanding of the prevailing consensus position, with which I disagree. It ends with "Do you buy it?" Just thought I should clarify that, since it's not clear to me that it's clear to you. ―Mandruss  07:34, 10 May 2018 (UTC)
It's still there on your user page I see, and that explains its history. It would have been helpful IMO if a talk page entry had been created when the essay was copy-and-pasted from your user page. As it is, it's arguably a copyvio... the enigmatic reference to your user page in the edit summary doesn't satisfy the copyleft requirements. I'll fix it.
Fixed. Andrewa (talk) 15:50, 10 May 2018 (UTC)
It still doesn't seem any help to us in incubating ideas on this page. But let us move on. Andrewa (talk) 11:22, 10 May 2018 (UTC)
Mandruss, thank you for explaining--and for not taking offense! I apologize if my statement did offend in any way. I agree with you that treating others the way I want to be treated doesn't always make people happy. My assertion is that they should be treated the way they want to be treated--as long as it is within reason. But let's be real. Some people are just bad-tempered. That's just the way it is. Some people won't apologize or admit to an error if their lives depended on it. So what to do about the uncooperative? Is there anything we can do?
Andrewa, vigorous discussion is not the issue. Just today I saw a discussion where one User was attempting to ask the editor I had the problem with to be patient with newcomers, that teaching is a better response than ridicule, that it's easy to forget what it was like when you were new, that threatening and belittling someone with 190 edits for what they don't know yet is counterproductive, that what appears a point of view in a newcomer is often just an interest (Amen!) and more in that vein--it was wonderful--absolutely respectful and kind. His response was "I'm not talking about this" and he deleted the discussion.
This kind of thing happens with him about every four to six weeks--someone has a problem with him, calls for some kind of arbitration with him--it's a pattern. It's easy to see why: his first response to anything he doesn't like is a mass revert without explanation or discussion. If you ask why, you get insulted. If you had a brain you would know you're a crap writer. If you try to adapt it to what you think the problem is and put it back, you get threatened. You can ask for compromise till you're blue in the face. Mostly you'll get ignored. There is no discussion--vigorous or otherwise. I can't tell you how many times I tried to discuss. I ended up with an Rfc where every single vote was in my favor--and it made him so angry he put his point of view in long, long "notes" to counter that. Consensus was against him--he didn't care. He's been doing this for years apparently and is basically bulletproof because of longevity. And because Wikipedia makes no effort to keep track of how many conflicts an editor is involved in or how frequently the same editors are involved in them. That's what I have seen. Jenhawk777 (talk) 19:24, 10 May 2018 (UTC)
The NYRM2016 fiasco had some similar issues. Several of those determined to prevent the move made repeated personal attacks on me and others. (And succeeded somehow in getting a no consensus decision despite the policy and facts both being clear and undisputed. The only change when NYRM2017 succeeded a year later was that we'd had an RfC that clarified that NYS wasn't the primary topic, which surely was clear before the RfC, but the 2016 closers firmly refused to confirm or deny this. I doubt the full story will ever be told. But what concerns us here is just the behaviour.) I reported these personal attacks twice at ANI, with diffs. The first time several non-admins agreed it was clearly a personal attack, but there was no evidence any admin even looked at it, and it was auto-archived through lack of further discussion. The second time, nobody even commented.
If that's not busted I don't know what is.
The "idea" we're supposed to be discussing is to have a policy prohibiting personal attacks. There seems to be no question that we already do have one! Andrewa (talk) 03:16, 13 May 2018 (UTC)
I don't see this discussion as limited to policy only, it has also included discussion of some kind of follow through and/or increased enforcement of the policy we already have. Though I do have to say that any policy without any enforcement is a policy in name only-- in my opinion.
Oh man! Been there! You have all my empathy! Any suggestions for monitoring/enforcement/policy changes/fire-bombing--anything? Jenhawk777 (talk) 20:03, 17 May 2018 (UTC)

────────────────────────────────────────────────────────────────────────────────────────────────────

Rudeness

This is kind of a weird idea, but I would like it if some of you considered improving the article Rudeness, with a particular emphasis on instrumental rudeness and the difficulty of determining what counts as "rude". I think that a clearer understanding of the incentives and the complexities would help us all. WhatamIdoing (talk) 02:00, 25 May 2018 (UTC)

I agree. That's kind of a weird idea. Face-smile.svgMandruss  02:06, 25 May 2018 (UTC)
Ha ha! The line between humor and rudeness is a little smudged isn't it? Thanx for the invite. I will take a look there, but I think I am probably done here. Wikipedia is, apparently, mostly happy with the status quo. The policy should read "Wikipedia is not for the faint of heart. Edit here at your own risk." It would at least be honest. Jenhawk777 (talk) 02:46, 25 May 2018 (UTC)
More accurately, a majority of the tiny fraction of editors who determine practice concerning these matters are happy with the status quo. They are self-selected, not elected representatives, and are therefore not "Wikipedia". The distinction is crucial. ―Mandruss  02:53, 25 May 2018 (UTC)
It's a rather weird situation. WP:NPA is a policy, and if any other policy is openly flouted, say at WP:RM, then many editors will descend in enthusiastic defense of the need to comply with say the WP:AT policy just because it is policy, because it reflects wider community consensus, etc.. NPA is extreme: Personal attacks harm the Wikipedia community and the collegial atmosphere needed to create a good encyclopedia... Insulting or disparaging an editor is a personal attack regardless of the manner in which it is done. (emphasis as in the original, omitted text indicated by ellipsis) It presumably represents consensus. None of the editors (and sysops) who regularly violate it have raised an RfC or even a discussion to have it weakened. How has it come to be so widely ignored? Andrewa (talk) 07:06, 25 May 2018 (UTC)
How has it come to be so widely ignored? The venues where practice is actually determined, such as WP:ANI, are downright nasty environments, and they are widely avoided by editors with milder dispositions who don't care to be around that unpleasantness. That leaves the controlling group highly skewed in the direction of editors who are either (1) combative and hostile, or (2) relatively unbothered by combativeness and hostility—so we have (quite naturally) ended up with a culture that tends to tolerate and excuse combativeness and hostility. That means widely ignoring written behavior policy. My question, not that it matters at this point, is how that managed to become written policy in the first place.
If we had a representative system of government with wide participation in elections, the "milder majority" would for the first time be fairly represented in decisions regarding editor behavior. I have little doubt that the resulting culture would be quite different, and Wikipedia would be a very different place at which to volunteer one's time. But the odds of that happening in our lifetime approach zero, as we would never reach the clear consensus required for such a change. Hence, intractable. ―Mandruss  07:40, 25 May 2018 (UTC)
I'm not sure the article is all that relevant. The lead there reads Rudeness (also called effrontery) is a display of disrespect by not complying with the social norms or etiquette of a group or culture. But that's too general... what is rudeness in our particular group or culture ie English Wikipedia? To make the article relevant, we'd need to find sources that described Wikipedia culture in particular, and cite these. Our own opinions as to what should be considered rude don't belong in the article namespace. They do however belong in the project namespace (here) and the project talk namespace (eg WT:NPA). Andrewa (talk) 07:06, 25 May 2018 (UTC)
I don't think that editing anything in main space can help. For a start it can be challenged over a lack of reliable sourcing, and then it irrelevant if you don't think you are being rude, with "I am just speaking plainly - and anyway they deserve it," kind of self-excuses. Nor do I think that a direct appeal to our founder can go anywhere. I once disagreed with him and he responded with exactly the kind of deliberately offensive insult we are complaining about here. Wikipedia needs a change of its corporate culture and that is extremely difficult if the head honcho is blind to their own failings and therefore themself part of the problem. But I am not wholly dispirited, see the next subsection. — Cheers, Steelpillow (Talk) 10:27, 25 May 2018 (UTC)
I think that a lack of understanding is a problem in these discussions, and when editors don't know anything about a policy-related issue, they often look at relevant articles. (See, e.g., Review article, which is linked in more than 3,000 pages outside the mainspace – 12 links in messages to editors for each link in the mainspace. People use that article to understand Wikipedia's rules.)
I don't think that we can deal with the civility problem when most editors conceptualize the source of rudeness as "Poor guy, he lost his temper" instead of "Hey, that guy chose to be rude. Why would a rational person do that? Oh, I get it: editors choose to be rude because being rude helps you win disputes in this community".
The smaller problem is the difficulty of defining rudeness: it's not just how the recipient feels, it's not just what the speaker intended, and it's definitely not what the speaker later claims to have intended when someone complains about it later. Think about the wikilawyering we see with NPA – the guy who claims that "You're stupid" is a personal attack, but "Everything you've posted on this page is stupid" is not a personal attack. Guess what? They're both personal attacks. They're both uncivil. They're both rude. They're both the kind of thing that we don't want in this community. But until we understand the difficulty of defining this, we won't get very far. And, yes, I do believe that reading some high-quality sources that discuss the subject of rudeness directly will help improve these discussions. (And if you're going to consult some sources, you might as well improve the article while you're at it...  ;-) ) WhatamIdoing (talk) 05:27, 29 May 2018 (UTC)

Best practice

I think the only way ahead is to look at best practice in the wider world and see if there is anything there we can learn from. I have some experience of civility codes in both commercial and public organisations, all in the UK. Here are some of my observations:

  • It is becoming increasingly accepted that rudeness and disrespect are in the mind of the recipient; if they feel insulted by you then you have insulted them, whether you intended to or not. WP:IUC needs updating accordingly.
  • Rudeness is about more than just words. Aggressive behaviours can be equally rude, not only in active aggression such as reversions but also in passive aggression such as refusal to acknowledge or discuss an issue or to admit any personal failing. WP:IUC could make this clearer.
  • Overly-detailed prescriptive guidelines are the wrong way to implement policy. An enlightened moderator is absolutely essential in dealing with incidents that escalate. As it stands today, WP:IUC is a classic example of how not to do it and does nothing but provide ammunition for logic-chopping excuses and "I have nothing to apologise for" attitudes. If it is simplified and refocused on perceived intent, that should help the moderating Admins to make better decisions.
  • Apologising for unintended harm, such as a perceived insult, is increasingly becoming mandatory. It is in this respect analogous to a fine for a parking offence, where the parking itself is only a civil offence but failing to pay the fine is a criminal one. Such a forced apology may well be mealy-mouthed and insincere, but it has been seen to be made and that is the crucial thing. Once somebody has been forced to cough up several such, they will begin to get the message. WP:CIVIL is grossly behind the times in this respect. It also needs a shortcut such as WP:APOLOGY (which currently redirects to an essay) to help raise awareness of its critical importance.
  • To be effective and deal with expert wrigglers, moderators also need a generic getout clause allowing, "we just find it unacceptably disrespectful overall" judgement even though specifics may be vague. An example would be an unjustified demand for an apology, where the demand is really just a cynical revenge manoeuvre. I don't know to what extent our Admins have this already.
  • Logging and tracking of escalated incidents is the norm. "You have been called here on three separate occasions already this year" type information should be available to moderators at the click of a button. Typically, the data is time-limited to prevent lifelong black marks. I don't know of our Admins have such tools, but they should.

Phew! I had no idea this list was going to be so long. I just want to re-emphasise that all this is established best practice that I have seen working well in the outside world, it is not my personal rant. No wonder we Wikipedians are in such a pickle! — Cheers, Steelpillow (Talk) 10:27, 25 May 2018 (UTC)

It is becoming increasingly accepted that rudeness and disrespect are in the mind of the recipient; if they feel insulted by you then you have insulted them, whether you intended to or not. I feel insulted by that assertion!

Actually, I find that assertion problematic because such a subjective criterion makes it very easy for someone to claim insult and demand apology as a method to derail the discussion or to harass an opponent. Or, worse, for someone to find insult in an accusation that they have insulted someone, which could then repeat //ad infinitum//. You mention this, saying An example would be an unjustified demand for an apology, where the demand is really just a cynical revenge manoeuvre, but that directly contradicts the assertion that the insult is in the ear of the hearer rather than in the intent of the speaker or the judgment of a neutral party. Anomie 16:44, 25 May 2018 (UTC)

Haha, cute. But the ear of the hearer is not the same as their scheming. You have provided an excellent illustration of why moderators need to be free to exercise their common sense, thank you. — Cheers, Steelpillow (Talk) 19:00, 25 May 2018 (UTC)
I think that "the ear of the hearer" isn't the whole story. That approach suggests that if you sweetly smile while cussing at someone in a language that they don't understand, then you've not been rude. I cannot agree with this. On the other side, according to this model, if you do something that is widely accepted as being polite or even deferential, such as a strong young man holding open a heavy door for an elderly woman, and she says that anyone who holds a door open for an elderly woman is either sexist or ableist or both, then he's being rude.
That's not how it works. Cussing at someone who doesn't understand your disrespect is still rude, and holding a door open for someone who might need the help is still civil, even if the targets of these behaviors don't see it that way. WhatamIdoing (talk) 05:40, 29 May 2018 (UTC)
Such a forced apology may well be mealy-mouthed and insincere, but it has been seen to be made and that is the crucial thing. It seems not everyone agrees that forced apologies accomplish anything useful.[1][2][3][4] We even have an article about the non-apology apology. Anomie 16:44, 25 May 2018 (UTC)
Yes, and utterly ineffective it has all turned out to be. The real world has discovered that this approach does not work, maybe it's time Wikipedia grokked that too. — Cheers, Steelpillow (Talk) 19:00, 25 May 2018 (UTC)
If Wikipedia had the wisdom to take lessons from the real world, we wouldn't have self-selected self-governance, which is the root of most of its problems in my view. ―Mandruss  03:08, 26 May 2018 (UTC)
@Steelpillow: I find your list brilliant, and just what I was looking for when I first came here, and I agree with each of your suggestions--especially logging and tracking. The objections are addressable.
Anomie In my experience subjectivism is already present in this issue. Recognizing that won't make it worse and might make it better. WhatamIdoing If there is a misunderstanding of intent, it is easy enough to say so. I did just this past week. "I meant no disrespect" generally works. Steelpillow You could be right that forced apologies may not be the best approach, but the real question is whether it would be better than what we have. If Admins had that logging feature, compliance could be considered to demonstrate good faith overall. Accepting that everyone screws up occasionally, it is the repetition of negative behaviors that demonstrate a pattern and without evidence of remorse that could all be weighed to determine overall good faith. Sort of a systemic approach. Mandruss I disagree with your conclusion.
I personally think Steelpillow is on to something. The suggestions are specific and doable. Jenhawk777 (talk) 18:30, 30 May 2018 (UTC)
"I meant no disrespect" is meant to change the mind of the target. The workflow you're talking about is this:
Do something respectful (e.g., use the word "sir" when addressing a man 40 years older than yourself) > Target felt insulted > Try to change the target's mind > If target insists that the respectful behavior was rude, then you actually were rude?
That's not reasonable. Respectful behavior does not become insulting just because someone is feeling grumpy about getting old or being addressed in a formal fashion by a stranger. IMO a more accurate and civilized flow looks much closer to this:
Do something respectful = You were being polite, even if the other person has a problem with the culture that both of you live in. WhatamIdoing (talk) 01:45, 31 May 2018 (UTC)
Agree (with Jenhawk777). All well worth a try. I still think it would be less trouble for more effect to just reinstate wp:NPA, but there seems no immediate prospect of doing that. (I'd love to be proven wrong on that.) Andrewa (talk) 01:50, 31 May 2018 (UTC)
WhatamIdoing > "I meant no disrespect" is not intended to change the mind of the target--at least not when I say it. It is merely meant to inform. People do jump to conclusions about intent, but they can't actually read minds and know for certain what my intent was--unless I inform them. I have found it helps generally.
I also find acknowledging the other point of view is sometimes all it takes. For example, your scenario is not wrong even though it is not the same as mine. It is a perfectly legitimate approach that gets to basically the same place I do--(accept differences)--without all the steps in between. Perfectly okay, (but I like my steps).
If they still insist I was rude, then in their minds, that is their reality, so for them the answer to "was I actually rude?" is yes. They certainly have as much right to define their reality as I do to define mine. In my mind, my only legitimate approach at that point is to say I am sorry they have been distressed--because I care about other people's feelings. It isn't about one point of view being right--my intent was still my intent--so much as it is not assuming that just because my intent was to be polite that it actually came off that way to the recipient. An apology in this scenario simply acknowledges the legitimacy of other points of view.
At that point, if they are still upset, I would say there is reason to believe there are other issues going on. That is when we need some way to get Admin or something involved. So what do you think about Steelpillows suggestions? What about a logging program that keeps track of the number of conflicts an editor regularly gets into? What about inventing a conflict resolution protocol from scratch? That's one extreme to the other, I know, but throwing all the possibilities out there seems legitimate here. I would love to hear your ideas. Jenhawk777 (talk) 09:06, 31 May 2018 (UTC)
If I may add to that, how often have I heard the phrase, "I was trying to be polite" after some misunderstanding arose. Trying to be polite and managing to be polite are different things. For example in some cultures it is polite to stick your tongue out in greeting, in others it is rude. Get it wrong and you have committed a deadly insult, whether you intended to or not. So yes, it is very possible to be rude without intending to be. Furthermore, telling the offended party that they need to change their ways only rubs salt in the wound. "I am so sorry, I meant no disrespect, please can you forgive me", is a far more constructive response. — Cheers, Steelpillow (Talk) 10:06, 31 May 2018 (UTC)
Jenhawk777, when someone misinterprets your intention, why do you want to inform that person of your intention? Could it be that you are hoping to change his mind, away from his erroneous conclusion about your intention and towards an accurate understanding of your intention? WhatamIdoing (talk) 15:23, 31 May 2018 (UTC)
I am hoping to cast oil upon troubled waters, to soothe, and calm the storm of offended feeling. They may continue to think my behavior was rude, but they may also feel less inclined to pursue beating me up for it because intent--motives--matter, generally as much as actual behaviors for most people.Jenhawk777 (talk) 15:36, 31 May 2018 (UTC)
@Steelpillow: Is there some way to make the recommendations you suggested to Wikipedia? Jenhawk777 (talk) 19:19, 3 June 2018 (UTC)
I don't know the best way. People at Wikipedia talk:Policy have been saying that there is no formal process. Perhaps Wikipedia:Village pump (policy) would be a good place to take them next and get some more focused feedback. Or maybe it is better to post a link there back to this discussion? — Cheers, Steelpillow (Talk) 20:39, 3 June 2018 (UTC)

──────────────────────────────────────────────────────────────────────────────────────────────────── I don't think it's appropriate for me to copy your ideas and post them there for you, but if you do decide to do that, please tell me and I will go there and participate in discussion there too. Thank you! Jenhawk777 (talk) 14:46, 4 June 2018 (UTC)

OK. I have started a thread there by asking much the same question as in my last post. — Cheers, Steelpillow (Talk) 16:31, 4 June 2018 (UTC)
Awesome. Jenhawk777 (talk) 01:52, 5 June 2018 (UTC)

It seems that this discussion discussion can be summarized with one question - how are policies enforced?Vorbee (talk) 11:21, 16 June 2018 (UTC)

They aren't. Or at least, where behaviour is concerned, not at all consistently. Andrewa (talk) 01:54, 29 June 2018 (UTC)

Policy discussion

See this now closed thread for some interesting discussion, expecially these suggestions. Andrewa (talk) 02:03, 29 June 2018 (UTC)

Expert review

Moved from WP:Village pump (miscellaneous): --Pipetricker (talk) 08:06, 1 June 2018 (UTC)

I'd appreciate feedback on an idea. I'm thinking of establishing a free expert review service for Wikipedia featured articles on academic topics - topics well-covered by reliable journals, such as medicine and astronomy. Once an article meets the FA criteria, the world's leading experts on the topic would fact-check it and tell you what they think of its comprehensiveness and neutrality.

I can only offer this service if I'm allowed to put two prominent links at the top of the current article version, one linking the reader to the version that passed review, and the other linking to a simple diff between the current version and the fact-checked version.

The world's topic experts aren't going to review an article if the version they endorse disappears into the article's history in a day or a month. They will if we link to the reviewed version. And the "simple diff" is a service to the reader: it shows them at a glance how the article (and topic) has evolved since the expert-review.

I've thought deeply on this for a long time. I asked BMJ, the publisher of The BMJ, to recruit experts to review Parkinsons disease, and they obliged. One of the reviewers was a main author of the current PD diagnostic criteria and another is the most-published author on the illness. This was a very high quality review. That's the standard of review I intend to maintain.

Do you think rigorous independent expert-review of featured articles is a good thing, and would you support prominently linking to the reviewed version and the diff? --Anthonyhcole (talk · contribs · email) 13:34, 30 May 2018 (UTC)

  • I'd be in favour of this, for what would inevitably be a rather limited number of articles, with some kind of simple control/approval process. In line with WP:MEDRS principles, I think there should a time-limit of up to 5 years set on the links, unless there's some kind of re-review. Johnbod (talk) 13:58, 30 May 2018 (UTC)
  • Not sure who will recall, but there were 2 similar proposals offered back in 2016: User:Atsme/WikiProject_Accuracy, which was presented at meta:Grants:IdeaLab, and a similar project was presented at the same time by another editor: Academic Reviewers. There's also Proposal:Expert review which is along the same lines. Atsme📞📧 14:08, 30 May 2018 (UTC)
How do you feel about those two prominent links at the top of the current version, Atsme? --Anthonyhcole (talk · contribs · email) 16:48, 30 May 2018 (UTC)
I'm fine with it. When I was researching for Project Accuracy, I spoke to quite a few academics (various teaching levels) and explained the significance of the GA & FA symbols on articles. Their responses are what inspired me to design the Seal. I still believe that once an FA goes through the drill of expert/academic review, they should be afforded some protection which makes that "seal" worth something. Atsme📞📧 16:59, 30 May 2018 (UTC)
  • Thank you, Atsme. I'm not sure where I stand on universal automatic protection for articles that have passed expert review. I think I'm against it but need to do more thinking about it. --Anthonyhcole (talk · contribs · email) 17:14, 30 May 2018 (UTC)
  • You're quite welcome, AHC - and if I may briefly explain why I feel some level of protection is needed...once an article has been reviewed to top level, any additions that follow will not have been reviewed; therefore, any newly introduced inaccuracies may be read/cited before the err is caught. The onus will fall on the promoting reviewers (presumably whose links are at the top). At least with some level of protection, it will allow the time needed to review & clear the new material. It is not that we are changing "the encyclopedia anyone can edit", it's simply a brief delay from time of edit to time of publication, but only for those promoted articles. Atsme📞📧 19:45, 30 May 2018 (UTC)
  • I think Dengue fever has been semi-protected since it was published in a peer-reviewed journal in 2014 and that hasn't ruffled any feathers. I see your point about protection. It may further encourage expert collaboration, too. As I say, I'm still making up my mind on this. But it's something for later, anyway. It's by no means a deal-breaker for me. Before I start spending my time and money on this, though, I need to know whether the community will let me do it. --Anthonyhcole (talk · contribs · email) 02:53, 31 May 2018 (UTC)
  • Hmm, the prominent links could be a hatnote, like the one we have linking to introduction articles, e.g like on General relativity. I think that at-least would be accepted by the community, and having a reviewed version does seem good; I'm just thinking - if we're not using the reviewed version as the default, we're sort of un-endorsing it; at the same time the fact that the reviewed versions would get out of date +general principles means we can't keep articles fixed on that. Not precisely related to this, but looking at the Project Accuracy pitch; most readers are not really critically looking at Wikipedia, and thus I don't think having reviewed versions would somehow make Wikipedia more reliable in the eyes of people Galobtter (pingó mió) 17:29, 30 May 2018 (UTC)
Was also here to say this, we already have Wikijournal where people can send FAs to get peer reviewed by professionals. Having one more similar process would drain the reviewer man-power. FunkMonk (talk) 14:16, 31 May 2018 (UTC)
It doesn't have to be separate - it can be coordinated in those topic areas. There's more to WP than just meds and science. Atsme📞📧 14:24, 31 May 2018 (UTC)
I've added more info lower down, but I thought I'd note here that I like the idea of coordinated mechanisms. I think that scholarly journals are an efficient way of incentivising expert engagement (whether WikiJournals or other journals), but I think that multiple mechanisms can work. E.g. an article gets written via WikiEdu, then undergoes GA, then expert review, then journal publication, then FA... etc. NB, There is also a WikiJournal of Humanities in the works. T.Shafee(Evo&Evo)talk 03:20, 15 June 2018 (UTC)

Jens, Mike, Dank and FunkMonk, I've been watching the development of Wikijournal of Medicine since its inception. My model differs in several important ways from Wikijournal of Medicine. The quality of the reviewers I'm offering is the highest possible. That's not the case with Wikijournal of Medicine. I won't be using the same pool of reviewers as Wikijournal of Medicine so I won't be draining that resource. I'm proposing we offer the Wikipedia reader a link to a simple diff showing them clearly the difference between the last reviewed version and the current version; Wikijournal of Medicine doesn't have anything like this in its model. My proposal includes a prominent link to the "reliable" version. The Wikijournal of Medicine model uses a tiny, essentially meaningless little book icon that no readers will understand without clicking and few will click. The names of all my reviewers will be published and prominently displayed on the reliable version; in the WikiJournal model the reviewers may remain anonymous. I'm not proposing to start a new journal to host the reliable version - the reliable version of an article that has passed review simply sits in the history of the article, available to readers who click the prominent link. There are other important differences too but this list should make it plain these are not the same product.

Let me emphasise this important distinction: The traditional academic publishing model relies on the reputation of the publisher, whom the reader trusts to run a high quality review by anonymous peers/experts, and the reputation of the authors, whose names are all disclosed. Both elements - the reputations of the publisher and the authors - are essential to rigorous science publishing. Wikipedia permits authors to remain anonymous and Wikijournal of medicine allows the reviewers to remain anonymous - leaving only the reputation of the publisher as a guarantee of reliability. That's not enough. WikiJournal has no reputation to speak of yet; in my model we use highly esteemed journal editorial boards with an already-established strong reputation for reliability to select only the very best reviewers. But even if WikiJournal were to develop a reputation rivalling Lancet and BMJ the WikiJournal model would still be inadequate. Humans - with careers and personal reputations and egos to protect - need to put their name behind the article. In my model the experts stake their reputations on the reliability of the reviewed article. I can't stress enough how important this particular difference between the two models is. (Although several of the other differences are very significant too, in terms of epistemology.) ---Anthonyhcole (talk · contribs · email) 15:45, 31 May 2018 (UTC)

I'm not saying that WikiJournal of Science meets all our needs and we don't have to consider other journals. I'm saying that there's already precedent for putting a special symbol (the book symbol) at the top of an article to notify readers that there's been some external form of review, and the symbol serves to send them to that paper. - Dank (push to talk) 16:28, 31 May 2018 (UTC)

Would this idea only be for featured articles?Vorbee (talk) 10:15, 2 June 2018 (UTC)

Yep. I only want to submit our very best work to experts (they're busy people) and the FA process is the best system we have for assessing quality. --Anthonyhcole (talk · contribs · email) 11:39, 2 June 2018 (UTC)
  • Support Both the concept of the expert review process as well as the inclusion of a link to the reviewed version. I see a suggestion that it can be done with a hatnote. I agree that it should be something akin to a hatnote but there may be a legitimate argument for making it look a little different than a hatnote as the concepts aren't exactly the same. (Generally, a hatnote is going to direct you to a completely different article, and if I'm looking at an article and it seems to be the right subject, I might not pay attention to a hat note, even though in this case it might be the one I'd prefer to see.)--S Philbrick(Talk) 20:11, 2 June 2018 (UTC)
  • Question – How many articles would this actually affect, in practice? I count 52 FAs in the Health and medicine category, which isn't that many when you think about it. I haven't counted FAs in other "academic" categories like astronomy, but let's be generous and say there are 250 of them. Is it really worth it to create a whole new system for the benefit of 300 articles, many of which figure to be in less need of expert help than B- and C-class articles? It's not like the medical WikiProject is cranking out FAs like crazy; most of the time we don't see any medical articles coming through FAC. Never mind that the vast majority of readers aren't going to bother clicking on a small icon that goes to a potentially years-out-of-date version of an article (they don't do it often for the talk page links to the version that passed a given process), or that experts might propose edits that would damage an article (not knowing our norms for a given topic). Without enough work on relevant articles, I fear that such an idea wouldn't be worth the effort. There's no point in pushing for experts to sign up for a review service if they won't have anything to do. Giants2008 (Talk) 00:49, 4 June 2018 (UTC)
  • Thanks, Giants2008, for your thoughtful response. As Johnbod suggested above, there'll be very few - at least to begin with. There are relatively few medical FAs and very few new ones rolling out.
  • There's no system or infrastructure required to implement this service - it piggy-backs on the peer-review process already in place at all the top journals. It will take a bit of my time to commission each review, and I'll supervise the review to make sure the reviewers aren't proposing off-policy changes (like adding dose information to drug articles). Scan the right hand column of this review, where you'll see this happening. I'm more than willing to put that time in.
  • You mention "... the vast majority of readers aren't going to click on a small icon ...". That's the point of this thread. It won't be a small icon. It'll be a prominent link of some kind. Galobtter, above, suggested a hatnote and Sphilbrck supported a hatnote or something like that. I'm not wedded to any particular format for the links, as long as it's not ugly, fits our style and is obvious to the reader. The hatnote (or whatever it ends up being) will only go up if a version of the article has passed the experts' review. --Anthonyhcole (talk · contribs · email) 10:08, 9 June 2018 (UTC)
  • Mixed Support encouraging such reviews, oppose fossilizing the reviewed version with a diff anywhere, especially in the article space itself. There should be no marker on the article text, and I'm leery of even a talk page notice, which would tend to encourage WP:OWN-type issues and fossilization of articles. I like that experts want to help review our articles, I don't like that someone will use this to prevent future improvement. --Jayron32 01:50, 8 June 2018 (UTC)
  • We're already linking readers to a peer-reviewed version of our articles, where one exists, Jayron23. See Dengue fever. Click the little book icon in the top right hand corner. This proposal is to make that link prominent and obvious when named experts perform the review and the review is managed by an established, highly regarded publisher with a strong reputation for reliability (a publisher who publishes highest quality reliable sources).
  • Experts of the calibre I'm talking about won't be interested in reviewing an article if the reviewed version is going to disappear forever into the article's history the moment another editor saves a revision.
  • Regarding "someone will use this to prevent future improvement", experience does not bear this out. Take Dengue fever, for example. It passed peer-review in 2014. I made this simple diff in 2016. The topic evolved over that time and the article kept up. None of those editors seem to have been remotely concerned about offending the reviewers or messing with a sacred cow. Even if someone does feel that way, Wikipedia's policies and guidelines prevail.
  • As for the reviewed version getting stale: above, Johnbod recommended a time limit on how long we should leave the link up, and I agree with him. --Anthonyhcole (talk · contribs · email) 10:54, 9 June 2018 (UTC)
If they're not interested in reviewing articles that will later get improved, that's what Wikipedia is'. If you want to have some permanent, unchanging encyclopedia written by experts, find somewhere else to do it. --Jayron32 11:55, 10 June 2018 (UTC)
Could you expand on that a little for me please Jayron32? I'm not proposing an unchanging encyclopedia. Editors will still edit the public-facing article. It will evolve just as Dengue fever did after its review. I can see you're strongly opposed to this but can't yet see what your objection is. What's the down side you're seeing that I'm not? --Anthonyhcole (talk · contribs · email) 02:42, 11 June 2018 (UTC)

──────────────────────────────────────────────────────────────────────────────────────────────────── Support: I think that there is room for several models of trying to engage expert review of articles. I agree that attracting high-quality reviewers is essential and collaborating with an established journal such as BMJ is a good way to achieve this. I think a hatnote and category would work well, or possibly a note at the top of the references section (e.g. Rotavirus#References). I agree that just the symbol alone is insufficient (most readers are similarly unaware of the FA star). As well as simple diffs in markup, the visual diffs viewer is pretty good these days or even lust a link to the version after review (same as done with GAs and FAs).

I think that locking or any sort should be handled in the same way as it would for any FA. For example, Circular_permutation_in_proteins has undergone several changes since its publication in PLOS CompBiol. Of course, thew ideal in my point of view would be for BMJ to publish the article if it passes their peer review standards, but that would of course rely on them being happy to publish CC-BY-SA and comfortable with large group authorship attribution, which is uncommon in many journals.[1]

Some possibly relevant links:

I don't think that there will be any great need to lock the pages after expert review (or at least, no more than for FAR). — Preceding unsigned comment added by Evolution and evolvability (talkcontribs) 03:13, 15 June 2018 (UTC)

  • Oppose, with an emphasis on my conflict of interest as creator of Wikiversity:WikiJournal of Medicine, where the WikiJournals can be regarded as competitors to this idea. I do think the WikiJournals serve the main purpose of this proposal already. It is already aiming at having the quality of the reviewers to be "the highest possible". If the problem is that the WikiJournal review symbol is too tiny, I think a better solution would be to make that one more prominent. As for having a latest version and a last reviewed version, I think we already have this mechanism in the form of Wikipedia:Pending changes. And as for a system for reviewers to clearly mark their contributions to articles, we already have Template:External peer review (its usage can be seen on its WhatLinksHere). Regardless, I support having experts review Wikipedia articles, I just don't think we need yet another system for it. Mikael Häggström (talk) 04:45, 16 June 2018 (UTC)
    Mike, I won't support putting a prominent link at the top of a Wikipedia medicine article, linking the reader to a version that has passed review organised by WikiJournal. As discussed recently, here, WikiJournal is not a reliable publisher and its journals are not reliable sources by the standards of WP:MEDRS. --Anthonyhcole (talk · contribs · email) 10:01, 3 July 2018 (UTC)
I accept that, and by the same reason I don't see why this proposal should be allowed to have even more prominent links at the top of mainspace articles. Is there anything making these reviews more reliable than those of WikiJournal? Mikael Häggström (talk) 10:10, 3 July 2018 (UTC)
Here, I see support for prominent links to the reviewed version when the review is conducted under the model described above. If you want to put prominent links at the top of articles, linking to WikiJournal, open a discussion like this one and get input from others. I'll elaborate on my opposition there, if you like, Mikael. --Anthonyhcole (talk · contribs · email) 09:13, 9 July 2018 (UTC)
  • Support: Expert peer review in Wikipedia is currently a marginal practice, and it would be good to test various mechanisms. If there were several mechanisms, they would probably not compete with one another, but rather help one another by making the practice more mainstream. And I do not see what harm would be done by linking to the reviewed version and the diff. This said, I fail to understand precisely what is the aim of the proposal. If it is to improve the quality of science articles in Wikipedia, why start with the ones that a priori need it the least, i.e. the featured articles? (In contrast, the rationale for WikiJournals is straightforward: incite academics to contribute more to Wikipedia, by making Wikipedia-style articles count in publication lists. This is why I am participating in WikiJSci.) Sylvain Ribault (talk) 19:41, 16 June 2018
    A key element of this proposed service is the involvement of the editorial boards of the most prestigious science and medicine journals in reviewer recruitment, and the selection of field leaders and other recognised experts as reviewers.
    They simply won't take on the review of an article that needs a lot of work, any more than they would accept such an article for review and publication in their journals. --Anthonyhcole (talk · contribs · email) 03:36, 18 June 2018 (UTC)
  • Hi Anthony, I have two questions. First, is this proposal for science articles only? Second (the perennial problem), how are you going to persuade the reviewers to do it? SarahSV (talk) 04:03, 18 June 2018 (UTC)
    Personally, Sarah, I'm not taking this any further than medicine. Medicine is a fairly "hard" (as opposed to soft) topic that's underpinned by a great deal of rigorous publishing and robust systems for consensus-building. If it works and is useful in medicine, then it might work and be useful in the hard sciences, too, but I won't be taking it there. I don't know if it will work in softer topics like the social sciences, history and literature, and it won't work for the vast majority of Wikipedia topics that aren't well-supported by academic publishing.
    It's the editor-in-chief of the relevant journal/s who needs persuading. Then their managing editor goes through her Rolodex and offers the gig to the relevant experts.
    The editorial teams at the top relevant journals have the expertise, experience and relationships to do this well.
    IF we begin with and stick with only the most highly-regarded journals, and IF they consistently come through with stellar review panels, I hope experts will soon become proud to be asked to review Wikipedia articles, and it will be something they'll put in their resume. IF that's how it unfolds then, as the reputation of expert-reviewed Wikipedia articles grows, the managing editors may find it easier to recruit the best reviewers. We'll see. It's early days. --Anthonyhcole (talk · contribs · email) 05:08, 18 June 2018 (UTC)
    Anthony, what do you see as the benefit for the editors of the journals? They will have to reward the reviewers in some way. It used to be easier to get academics to do reviews. But the more we grew, and the more money the Foundation became associated with, the less eager reviewers have been to volunteer their time. I can imagine that someone might pay reviewers (e.g. some charitable medical foundation), although there's a risk that the payer would interfere editorially, and we would face the unfairness of reviewers being paid while writers are not. SarahSV (talk) 01:29, 23 June 2018 (UTC)
Sarah, many of the top journals are published by scholarly societies and professional associations that have, as a part of their mission, education and the dissemination of knowledge about their specialism. If we present the reader with a prominent link to the fact-checked version, it'll be fairly easy to convince those journals it is worth the effort to manage a review, I think. The reward for the reviewers is (1) altruism, like you and me, and (2) prestige, per my last paragraph above. This latter isn't an afterthought. It's a key element of the model. I expect a Wikipedia medical article reviewed under this model to be regarded as the most reliable source on the topic, period. And I expect scholars and experts to see an invitation to review as a very visible public acknowledgement of their standing in their field.
As for money:
  • There is a role for a very experienced Wikipedian in each review, liaising between the experts and the writers: (1) ensuring the expert suggestions are compliant with our policies, (2) finding reliable sources that support proposed changes, (3) updating the article in collaboration with other editors in response to the review, (4) re-presenting the updated article to the reviewers for endorsement and (5) formatting the "reliable version" with relevant templates, etc. This is pretty onerous, tedious, exacting work and I can see it becoming a paid role at some point.
  • I would be very disappointed and a bit surprised if it turned out the reviewers needed paying. I think, I'm pretty confident actually, it can be avoided.
  • Any money supporting this effort can't come from the WMF because of perceived (at least) conflict of interest. There are a number of non-profits out there with education in their remit. Anthonyhcole (talk · contribs · email) 09:39, 23 June 2018 (UTC)

I'm thinking about the consequences of Google linking searchers directly to the fact-checked version. That could cost us a lot of editors, like me, who love clicking publish. I'd lose interest in editing if my edits took months or years to appear. A few years ago, Google committed to privileging reliability over popularity in its search results. If this journal-driven expert review service becomes a thing, I'll make sure Google knows the harm direct-linking would do. --Anthonyhcole (talk · contribs · email) 12:49, 25 June 2018 (UTC)

  • Support. Only expert reviewers who spend many hours every week doing research in their field of expertise, who read vast numbers of papers can have a good judgment to weed out subtle problems with review articles written by non-experts. A review article written by scientists is always based on a huge amount of literature research, this is obviously not the way we go about writing articles here. We cruise on autopilot by summarizing the contents of review articles, but this can lead to inaccuracies. The main problem here is caused by our reliance on the top journals like e.g. the BMJ, the Lancet etc., while these journals only publish a small percentage of the research results. A lot more is going on behind the scenes, the vast majority of the relevant research for any particular topic is published in technical journals that we cannot possibly keep track of. While our articles will end up presenting most of the relevant information, things tend to go wrong with giving the right weight to different ideas. Different ideas that are not considered to be equally likely to be true, tend to get presented in a more equal way in the top journals, because one wants to see what the most rigorous evidence tells us and not be biased based on the previously known less rigorous evidence. That's a good way to eventually get to the rigorously verified truth, but it may sometimes mislead lay people about how researchers in the field think about the different ideas. Count Iblis (talk) 15:42, 25 June 2018 (UTC)

References

  1. ^ a b "Wikipedia-integrated publishing: a comparison of successful models". 26 (2). Health inform. doi:10.13140/RG.2.2.27470.77129. 
  2. ^ Mietchen, Daniel; Wodak, Shoshana; Wasik, Szymon; Szostak, Natalia; Dessimoz, Christophe (2018-05-31). "Submit a Topic Page to PLOS Computational Biology and Wikipedia". PLOS Computational Biology. 14 (5): e1006137. doi:10.1371/journal.pcbi.1006137. ISSN 1553-7358. PMID 29851950. 
  3. ^ Maskalyk, James (2014-10-02). "Modern medicine comes online:How putting Wikipedia articles through a medical journal's traditional process can put free, reliable information into as many hands as possible". Open Medicine. 8 (4): e116–e119. ISSN 1911-2092. PMC 4242788Freely accessible. PMID 25426179. 
  4. ^ Shafee, Thomas; Das, Diptanshu; Masukume, Gwinyai; Häggström, Mikael (2017-01-15). "WikiJournal of Medicine, the first Wikipedia-integrated academic journal". WikiJournal of Medicine. 4 (1). doi:10.15347/wjm/2017.001. ISSN 2002-4436. 
  5. ^ Shafee, Thomas (2018-06-01). "The aims and scope of WikiJournal of Science". WikiJournal of Science. 1 (1): 1. doi:10.15347/wjs/2018.001. ISSN 2470-6345. 
  6. ^ Butler, Declan (2008-12-16). "Publish in Wikipedia or perish". Nature. doi:10.1038/news.2008.1312. ISSN 0028-0836. 
  7. ^ Su, Andrew I.; Good, Benjamin M.; van Wijnen, Andre J. (2013). "Gene Wiki Reviews: Marrying crowdsourcing with traditional peer review". Gene. 531 (2): 125. doi:10.1016/j.gene.2013.08.093. ISSN 0378-1119. PMID 24012870. 

Make "thank log" more visible

Dear all on WP:Village pump,

First time here, if this was the wrong place to post such idea please advice me to reroute.

To promote the atmosphere of appreciation and encourage positive collaborating on the WP environment, I'd like to suggest we make thank log from and towards a user more visible, for example, make it a direct link from Tools, or somewhere default in a user page. Also, we can make it more visible that some users are generally more appreciating. The idea is still very early stage, I'd like to hear what people thinks about that.

Xinbenlv (talk) 18:26, 5 June 2018 (UTC)

I have no problem with this. I'd also like a little "thank" to appear after my signature when I sign a comment. --Anthonyhcole (talk · contribs · email) 01:10, 10 June 2018 (UTC)
I'd rather make thanks hidden as thanking for some edits which may be controversial upsets other editors who oppose those edits Atlantic306 (talk) 20:56, 13 June 2018 (UTC)
The thanks log Special:Log/thanks does not reveal which edit was thanked or which page was edited. It only shows the two users and the time of the thanks. PrimeHunter (talk) 21:18, 13 June 2018 (UTC)
True, although it is often not difficult to make an educated guess about the edit when a dispute is going on. WhatamIdoing (talk) 19:40, 21 June 2018 (UTC)
  • There is already a red coloured heart icon on User page, to click and send Wikilove as a THANK YOU. I feel the current Facebook style notification for Thank you is quite excellent in doing the intended job. any more visibility i believe will be intrusive and Editors would then ask for a button to turn these thank you notifications off.--DBigXray 21:08, 13 June 2018 (UTC)
    @DBigXray: that button is in here. — xaosflux Talk 00:25, 14 June 2018 (UTC)

Standardizing revert reasons and process

I'd be interested in people's views on the following idea, to help address a problem I've noticed many encounter

The Problem. Reverting other people's edits is required to deal with vandalism and other issues. But problems arise when this is abused - e.g. when users delete content they disagree with, generating disputes, incivility and edit wars, potentially driving away editors, particularly new ones. Currently there are few guidelines for reverting edits, leaving it mostly to the discretion of editors when to delete other people’s work, contributing to disputes. Editors are merely requested to type their own reason for reverts, some don’t even do that

One Possible Solution. One way to help address this would be to agree upon specific, valid reasons for reverts, and then list these on Wikipedia Diff Pages, for users to click to always indicate their reason for Undo (these are illustrative only):

 Undo(?): Vandalism  NPOV  Verifiability  Copyright  Redundant  Other  

E.g. clicking NPOV would automatically add “Undid: NPOV” in Edit Summary, to indicate NPOV rule violation was reason for Undo, thus saving typing. Note: Clicking (?) would take user to WP page that explains these agreed-upon revert reasons, including what users should do before, or instead of resorting to undo, where appropriate. For reverts via regular edit, could add radio buttons below Edit Summary box, to indicate delete reason, e.g:

 If you deleted other people’s edits, indicate a reason for delete(?):
 o Vandalism  o NPOV  o Reliable Sources  o Copyright  o Redundant o Other 

Clicking NPOV radio would insert “Deleted: NPOV” in Edit Summary, to indicate NPOV violation as delete reason. There are many possible variations, and slightly different approach could be used for History page Undo and Rollback (latter needs to be single-click)

Benefits. Specifying, reminding and requiring users to give valid revert reasons could reduce abusive reverts, and let people know why their edits have been deleted, thus reducing revert-related disputes, incivility and edit wars. This could also aid in retaining editors, particularly novices, women, etc (Note: above Diff page Undo-reasons would not require any extra clicks, and also save on typing reason. Regular-edit reverts would require 1-extra reason click, but save on typing reason - so almost always would be less effort than currently)--Thhhommmasss (talk) 18:41, 10 June 2018 (UTC)

Wikipedia:BOLD, revert, discuss cycle is a highly influential code of conduct. It states that you can, and should, revert when you don't think that the preceding edit was an improvement. It has worked for us terrifically well. Somewhat paradoxically, it's only by making reverting easy that we can make implementing changes easy.
As for the standard edit summaries, there are tools you can use for this. – Finnusertop (talkcontribs) 06:18, 14 June 2018 (UTC)
That’s fine, but the problem arises when this is abused. See for example the above comment, referring to “instant reverts, threats, insults...”, with the effect that “I have seen more than one editor driven completely off Wikipedia”. Wikipedia, when it works well, as it often does, works because of specific rules (NPOV, Verifiability, etc.) – i.e. people can’t merely exercise their discretion (e.g. “do whatever you think will improve the article” is generally not the Wikipedia way). Yet here, for something that clearly ticks off people – seeing their good-faith edits reverted – a lot of discretion seems to be provided, i.e. revert anything you think does not improve the article, and discuss it out. This can then lead to some of the additional abuse mentioned in the above comment – insults and incivility on talk pages, etc. There are specific, listed criteria for speedy page deletion. This is a similar issue of deleting someone else’s content, thus I think listing comparable reasons for reverts would also be helpful. And I think one can still be bold on reverts, while listing valid reasons, just as listing page delete criteria still allows for speedy page deletion. Also, the above suggestion would in nearly all cases save effort, that is save on typing a revert reason--Thhhommmasss (talk) 19:47, 15 June 2018 (UTC)
"Do whatever you think will improve the article" actually is the Wikipedia way. This principle is enshrined in one of the oldest policies here: Wikipedia:Ignore all rules.
We do have people (new and otherwise) whose idea of improving an article isn't widely shared, with the result that their additions or reversions don't – in the opinion of the rest of us – actually improve the article. But generally, the idea is that you should do your best, and that others should also do their best, and in the end that usually works out. WhatamIdoing (talk) 19:47, 21 June 2018 (UTC)
I still think what makes Wikipedia is its rules, even if occasionally they must be broken. Take away the rules, and you’d have a mess, i.e. people throwing random opinions together. On revert, I’ve seen people come up with their own revert reasons, which I’m sure 99.9% of the people would consider bogus, and then you’re stuck days debating with them over this bogus reason. I think that’s what ticks people off, arbitrariness. I’ve seen this in particular on another language Wikipedia, where people have given up, and the result is every couple of years media articles appear on how biased this language Wikipedia is. Now it’s true that even defined, valid revert criteria, could be misused or misapplied - someone might claim something is a valid delete, e.g. due to lack of Reliable Sources, when it’s not. However Reliable Sources are quite well defined, and it is more difficult to just wing there. In any case, with well-defined, valid revert reasons, the discussion would be much more constrained. The delete reason tag in Edit Summary could link to a description of the delete reason, for everyone to instantly refer to the reason description, use it as an arbiter, instead of having to engage on nonsense, made-up reasons, which take people away from valuable tasks, like improving articles, or as others have noted, leads them to quit--Thhhommmasss (talk) 01:12, 26 June 2018 (UTC)
Editors already widely abuse concepts like vandalism, NPOV, and BLP, as seen in their edit summaries. Formalizing this abuse would accomplish exactly nothing. ―Mandruss  20:23, 21 June 2018 (UTC)
In the limit that could be an argument for no rules, since all rules are likely to be abused. There is a lot of research that people can be "nudged" in the right direction with subtle clues. What no doubt happens in abuse is people get upset by ideas that are contrary to theirs, and instinctively act (i.e. delete first, and rationalize later). Perhaps something that requires them to consider valid delete reasons for just a millisecond before they can delete, can prompt slightly more rational behavior. Ultimately such principles can be empirically tested via A/B testing and other strategies. Often the winning approach can be what we might consider as counter intuitive Thhhommmasss (talk) 21:15, 24 June 2018 (UTC)
The folks willing to be nudged tend to learn from people pointing out to them a particular revert was not proper, or from reading the guidelines, or from seeing what other long-term editors use revert for even without having to think for a millisecond as result of being prompted for a standardized response. The folks who revert things not meant to be reverted while not being open to nudging/correcting are frequently perfectly willing to be deceptive (e.g. claiming they are reverting vandalism—which means they do consider what are valid delete reasons for a millisecond—when what they're reverting is by no means vandalism nor even remotely resembles it).
A similar idea was deployed on the Wikipedia App for general edits, the so-said "canned edit summaries". They are so often used deceptively or dishonestly that I—and probably most vandal-fighters—see use of them by new or IP editors as at best neutral/not effecting the suspiciousness of an edit and at worst an actual indication it might be a good idea to check said edit. If it didn't resolve the issue there, I see little reason to believe it *would* resolve reverting-related issues. AddWittyNameHere (talk) 23:32, 24 June 2018 (UTC)
Nudges work, they work subconsciously, instantaneously, even a small thing like changing button color can impact the likelihood people click it. Here we want to nudge people from automatic, emotional behavior (i.e. delete in anger, rationalize later), to have them first provide a quick, rational delete reason. Hopefully, that can kick them into a more rational mode (e.g. can I defend this delete reason, if not, should I delete?). Studies of negotiation under anger show that for rational reasoning to kick-in, this must be done early in the cycle. Here a delete reason is selected before deleting, which may be more successful than delete first, specify a reason later (which is what canned edit summaries sound like), because once a delete is made, people are more likely to defend it, justified or not. That’s the hypothesis, with some experimental backing (i.e. nudges work, early interventions work), although admittedly those cases are not exactly like this proposal. Thus, I do not know for sure in advance that it will work, just like you can't know for sure that it will not. That is why there are things like A/B testing, so people don't just debate which color they think will encourage most button clicks, etc. Running experiments is also the scientific way – i.e. scientists do not decide merely by debate, or majority vote, which hypothesis is correct--Thhhommmasss (talk) 22:54, 25 June 2018 (UTC)
Yes, running experiments is the scientific way. However, scientists don't generally go from a hypothesis to proposing laws to experiment whether the addition of that law would help solve the problem. Which is the real world equivalent of your proposal.
I'm not saying nudges don't work at all. As you say, there's some research backing that they may work—when someone is acting from automatic, emotional behaviour. However, that is not the premise I am in disagreement with. What I am in disagreement with is the to this experiment fundamental analysis that a large, if not the largest, portion of problematic reverts is the result of people acting from automatic, emotional behaviour to the degree they either lack rational awareness of their actions and/or have said rational awareness overruled by their emotions. (That is not to say that I believe no reverts are done in such circumstances, but rather that from my observation I suspect they account for a far smaller proportion of problematic reverts than you suspect they do)
Rather, I believe that most people who commit problematic reverts are, in no particular order
  1. acting from a misunderstanding of the rules, in which case asking them to provide a rational reason will hardly work, simply because they already are in a rational mindset and genuinely believe they are acting in line with guidelines&policy (good faith problematic reverts);
  2. or acting from a genuine belief that even if this exact situation isn't coded into the rules, it's common sense and obvious that an exception applies/should apply to this particular situation (good faith problematic reverts/misapplication of IAR);
  3. or perfectly well-aware that what they're doing is not in line with the rules, but are either holding disregard for rules in general or are in such disagreement with a specific rule (or rules in a specific context) that they are unwilling to hold themselves to said rule/to rules in said situation (bad faith problematic reverts; this one is partially emotion-based but nonetheless not lacking the rational awareness that is the entire premise of nudges working);
  4. or acting especially because what they're doing is not in line with what should be done. (trolls, vandals, LTAers etc.)
While, as I said above, I don't doubt that some cases of predominantly-emotional problematic reverts occur by people genuinely lacking rational awareness in that moment, even of those it is not a given that they are committed by people who would, if only they were thinking rationally, both realize that what they are doing is not what they should be doing and consider this sufficient reason to thus not do it. Some will be, of course, but I suspect that they make up such a small portion of the problematic reverts that the additional hassle for everyone else using revert in a non-problematic manner is simply not worth it. AddWittyNameHere (talk) 03:39, 26 June 2018 (UTC)
Agree, hypotheses should be questioned, but on the rest I’d disagree. Studies show edit wars are most frequent on subjects that generate strong emotional response – politics, religion, history, gender, race, etc. You probably don’t see heated edit wars on Medford, Oregon or 19th century ballerinas. Psychologists, like Daniel Kahneman, have shown that our much older, automatic, emotional response systems are much more powerful than our recently developed rational systems, and that these emotionally-based systems often lead us astray, with all sorts of biases, etc. Others, like Jonathan Haidt, have shown that automatic, emotional responses are particularly strong on issues like politics and religion, which often go to people’s identity, thus driving bad behavior – incivility, polarization, etc. Cognitive behavior therapies are most successful in dealing with bad behaviors (e.g. anger management) and these seek to put in circuit-breakers, so people aren’t controlled by negative emotions. Of course, that requires individuals taking action themselves, which is unlikely in edit wars, trolling, etc, thus external circuit-breakers are needed. Wikipedia rules (NPOV, Verifiability, etc), are all intended to promote more rational behavior, so people aren’t merely guided by their emotional biases, and these rules have been constantly extended and refined over time --Thhhommmasss (talk) 01:24, 27 June 2018 (UTC)
(edit conflict)Interesting study, though from what little the link you gave shows, it doesn't show that edit wars are most frequent on those topics, but that the highest volume of reverts of any type occurs on those subjects, a fairly significant difference. As these subjects are often also the frequent target of vandalism and test edits, it would be interesting to see what percentage of those reverts is in fact reverting vandalism or otherwise non-problematic reverts. Not every revert is an edit war; not every seeming edit war is a content dispute. (Some of our more prolific LTAers are perfectly happy to keep dumping the same vandalism into articles over and over again until they're blocked. And then again as soon as they can get a new IP)
I'd disagree that one might not see heated edit wars on, say, Medford, Oregon. We've had one at Chicken, Alaska. We've had one over the use of an exclamation mark at Berwick-upon-Tweed(!) Similarly, we've had edit wars on ellipses; on diacritics; on capitalization; on taxonomy; on vernacular names of species; repeatedly on capitalization of vernacular names of species; on whether to call a . a period or a full stop; and on basically every other subject one would think wouldn't be the target of edit wars.
Certainly emotion drives strong impulses and responses, and some subjects are more likely to trigger strong emotions. We disagree on the proportion of problematic reverts that occur from a purely-emotion-driven state, though we both agree it does occur, and probably disproportionally often in case of edit wars as opposed to other types of problematic reverts. As I've already explained in my prior response why I disagree with you on this part, I will not bore both of us by repeating myself. Instead, let's consider those folks that are acting from an emotional state.
As you say, identity drives emotion drives bias. If someone, from a deeply emotional response and lack of recognition of their own bias, is convinced they are dealing with a NPOV-violating edit or article, no nudge is going to make them consider that perhaps the article is NPOV and it's their view that isn't, because if anything, considering such a thing would be a far larger and more direct challenge to their identity than the edit/article involved. They'll remain mostly in their mindset ("wrong, wrong, wrong") and treat the nudge in a similarly-automatic manner as they do the revert itself.
A nudge works on behaviour, by putting people into a rational enough mindset that they see an action they were about to commit is not appropriate for the circumstances without challenging their identity. However, reverts are not by default an inappropriate response to a NPOV-violation. If anything, they fairly frequently are the appropriate response. What is inappropriate is in their identification of NPOV-vio, which a nudge is for the reasons I mentioned above is unlikely to change.AddWittyNameHere (talk) 02:42, 27 June 2018 (UTC)
Psychologists like Jonathan Haidt write that nearly all bad behavior on volatile issues, like politics and religion, is driven by our automatic, impulsive emotional systems, and would likely say that nearly all bad behaviors (e.g. arguing over capitalization), are also driven by our automatic, impulsive systems. Cognitive behavioral therapy is intended to deal with bad impulses, regardless of what sparked them (religion, bad punctuation, etc.), and as mentioned often works by seeking to put in circuit-breakers, so people aren’t driven by negative impulses/emotions. By contrast, an act-first (i.e. revert), and-rationalize-later (put reason in edit summary) approach facilitates impulsive behavior. No process, rule or law will entirely prevent bad behavior, not CBT, not nudges. Yet CBT is effectively applied to deal with highly compulsive behavior – gambling, addiction, violent impulses, etc. Whether above proposed nudge can reduce some bad revert behaviors is unknowable without testing - Google and Facebook constantly do thousands of tests to nudge behavior, precisely because they do not know in advance what will or will not work. Btw, I agree valid NPOV issues should be reverted, but valid, rationally-based reverts would of course still be encouraged, made even a little easier, since part of reason is placed in edit summary via single click, instead of having to type it--Thhhommmasss (talk) 01:31, 29 June 2018 (UTC)
Happy edits are all alike; every problematic edit is problematic in its own way. Having a canned "revert: not compliant with WP:NPOV" edit summary is almost negligibly more useful than no edit summary at all, and it relieves the editor making the revert of the obligation of providing an even remotely situation-specific bit of reasoning. (Hint: virtually every contentious edit – revert or not – is made by an editor who believes they are defending WP:NPOV.) TenOfAllTrades(talk) 01:47, 27 June 2018 (UTC)
As I mentioned to Whatamidoing, I’ve seen editors revert and give ridiculous reasons, and then people have to waste time arguing over ridiculous reasons. If instead people used valid revert reasons - NPOV, Verifiability, Reliable Sources, etc. - these are fairly well-documented, harder to misuse than entirely made-up reasons, and I’m always happy to discuss valid reasons. I’ve also seen where users revert, but not give a WP rule, and then after some back-and-forth, finally give a valid WP Rule (e.g. Verifiability) – again wasted initial effort, compared to always having a revert reason/rule everyone can refer to, without needless back-and-forth. With automated revert-reason tags, users could still be asked to provide further, instance-specific elaboration in Edit Summary--Thhhommmasss (talk) 01:50, 29 June 2018 (UTC)
But providing canned edit summaries wouldn't stop people from making reversions that you disagree with. It would, however, increase the problem of edit summaries that have nothing to do with the actual problem.
I think that you might be interested in [www.wnycstudios.org] (partial description) about what happened to Facebook abuse-claims team, on the day after half the world got smartphones for Christmas. They came back to work and discovered huge numbers of photos being reported as hate speech, threats, and nudity. The photos were things like people wearing matching sweaters, or puppies.
It turned out that their friends and family had uploaded a bunch of photos, and people were using the abuse-reporting feature because they were embarrassed by the photos. There wasn't (at that time) another option, so the puppies got tagged as hate speech. The people tagging these weren't deterred by puppies not being hate speech; the system gave them about four options, none of which were "This photo is embarrassing", so they clicked whatever buttons were available, basically at random, and reported the photos that way.
I believe that exactly the same thing would happen with your proposed system: If someone decides to revert an edit, and you force them to choose one of six options to be able to complete the reversion, then they will pick one of the options, even if those options have nothing at all to do with their real reasons.
If you want to deter reversions, you need to be reaching people before they click the undo button, or you need to give them an alternative that is (to their way of thinking) more effective. Your proposal won't do that. WhatamIdoing (talk) 22:07, 29 June 2018 (UTC)
As mentioned in above proposal, users would have to select a valid Undo reason, before undo happens, with the goal of nudging user into a more rational reasoning mode, as opposed to doing impulsive emotional undos, and then typing reasons/rationalizations later (now you can even undo, without typing any undo reason). To your other point, one cold randomly select 100-200 existing undos, where undo reasons have been specified in Edit Summary, to see if these can be categorized into 7 or 8 valid reasons that account for 80-90% of reverts (it may be possible to have more, e.g. 11 canned reasons, but fewer would be better). In any case, it’d be good to make decisions based on such systematic analyses of actual data, instead of what often happens in these discussions – my anecdotes/impressions vs. your anecdotes/impressions. Then for exceptions to canned reasons, one could add an Other reason, with requirement that this be further elaborated in Edit Summary, and that it be used for valid exceptions, not to invent reasons. Further refinements could be done (e.g. later analyze reasons people are putting under Other – if there are repetitive, valid reasons, add these to canned list, but if they’re making up stuff to abuse reverts, give these as examples of what is unacceptable, etc.) Thhhommmasss (talk) 18:40, 4 July 2018 (UTC)
  1. By the time I have asked the software to present me with your list of pre-approved "valid Undo reasons", I have already decided that I want to undo it. As I said, "If you want to deter reversions, you need to be reaching people before they click the undo button." That's "the undo button" that you find in the page history and at the top of the diff, not the "Publish changes" button that you click when you're finished.
  2. What's going to stop the would-be reverter from selecting any of the pre-loaded "valid Undo reasons" at random? Nothing? Then it won't work. Your process is (a) click the Undo button, (b) choose from among a list of pre-approved, guaranteed-to-be-valid reasons, (c) save your change. I'm telling you that nobody who really wants to revert a change is going to be deterred by needing to click on a pre-approved "valid Undo reason". It's barely even going to slow them down. On the other hand, the use of random and irrelevant "valid Undo reasons" is going to be very irritating to people who are trying to review these edits later. No information about the edit is better than wrong information. WhatamIdoing (talk) 02:34, 6 July 2018 (UTC)
  1. As indicated in proposal, on diff page, user would not click Undo, but instead in order to Undo, they would have to click one of given valid reasons listed next Undo, so user would always have to give a reason before undoing. On History page, without getting into design specifics, similar could be done (e.g. as soon as user hovers over Undo, a list of valid Undo reasons pops up, and they must select 1 to do undo), so again would need to select valid Undo Reason before Undo
  2. If user gives random response, then other person can come back with “You’re wrong”. Since people usually do not enjoy hearing they’re wrong, this is disincentive for random responses. Now users can entirely make up revert reasons, and people have to engage with them to discuss these bogus reasons. As to whether a revert is done with no explanation, or with a random canned reason, editors still need to check what was done, and then revert accordingly, so I do not see any difference in effort here. In fact, now when no revert reason is given, I suspect people often engage in additional needless effort to ask why was revert made, or spend time to comment that no revert reason was given, even though it should have been — Preceding unsigned comment added by Thhhommmasss (talkcontribs) 23:25, 8 July 2018 (UTC)
    I assume then - and not entirely facetiously - that one of the pre-approved valid Undo reasons is "You're wrong"? —Cryptic 23:48, 8 July 2018 (UTC)
Also "Test edit".
I just don't see how providing half a dozen Undo buttons (each with its own custom edit summary) is supposed to discourage people from clicking an Undo button. In fact, it would probably result in slightly more reversions, due to misclicks and reminding people of all the many reasons why they could reject someone else's work.
Thhhommmasss, I think this is a bad idea. But in the interests of fairness, if you still think it is a good idea, then your next step is to write this in Javascript and run it as a user script on your own account, to see how it works for you. Alternatively, you could start using WP:Huggle, which already provides this. However, I feel compelled to tell you that previous academic research about Wikipedia indicates that Huggle (and other software like it), which has made it so easy to revert other editors without even needing to hand-write an acceptable reason for doing so, is significantly responsible for the decline in the number of volunteers editing here. WhatamIdoing (talk) 00:45, 9 July 2018 (UTC)
  1. I’m suggesting after user clicks an Undo reason, it puts this reason in Edit Summary, and user can be asked to add more explanation, but at least it fills in part of reason, so there is some saving on typing. User must still click Publish Changes, same as now, for Undo to take effect. So user has chance to look at their Edit Summary, see which canned reason they entered, expand on this, or change canned reason, before they click Publish Changes. So it’s very unlikely they’ll press Publish Changes with wrong canned reason, and should also avoid issues you mention with WP:Huggle
  2. You say Huggle led to increased editor attrition. This goes to core of proposal, confirming that slight changes in UI (i.e. 1-click Revert, vs. having to confirm via at least 1 additional click on Publish Changes), can produce substantial changes in behavior - thus it is not the case that those who behave badly, will behave bad regardless, and UI changes have no impact
  3. On issue of revert reasons perhaps serving as reminder to revert, one simple way to take care of this is to put the Reasons in a drop-down menu, which user sees only when they click Undo, so they are less obvious. Then when they select Reason, it appears both in Undo and in Edit Summary, so there’d be even less possibility for wrong selection, plus this saves space, and is smaller change compared to current UI
  4. Regarding reverting a Test edit, I assume user is reverting own edit. For this there could be a canned Revert reason, e.g. “My”, where they can revert themselves all they want. For Wrong I’d put something like Incorrect (sounds more neutral, means same), and then in documenting valid revert reasons say that for Incorrect they owe more explanation in Edit Summary and/or on Talk page. I’d be glad to debate Incorrect (as opposed to entirely bogus reasons I got on revert, which did not claim my citations were incorrect)
  5. I agree with SMcCandlish, that a good first step would be to document valid revert reasons, similar to how valid page delete reasons were documented, since this is needed regardless of any other possible actions. Is there some way to get 200-300 randomly selected Undos? This would help determine what people give as Undo reasons, to see if these can be grouped into some sensible number of valid reasons. Or have studies been done of Undo reasons? --Thhhommmasss (talk) 02:17, 10 July 2018 (UTC)
This proposal would be very limiting. I would rather see WP:Editing policy be updated with clarity on what are and usually are not valid revert rationales. Many editors (including experienced ones) believe they have a right to revert, no matter what, and to keep doing so (within the limits of WP:EDITWAR), even if they cannot or will not articulate a policy-based reason. They're wrong.  — SMcCandlish ¢ 😼  02:10, 3 July 2018 (UTC)
Agree documenting valid revert reasons, similar to existing effort to document valid page delete reasons, would be good regardless of specific approach. In addition, putting a question mark next to the Undo command that links to these reasons would be subtle, constant reminder of valid reasons. Also “Undid” that now automatically appears in Edit Summary, could link to these same reasons, so everyone, including new users, would have instant access. If revert was done via regular edit, “Reverted” could be automatically placed in Edit Summary with link to same explanatory page (there are many possible tweaks/variations). As to other parts of proposal, I still maintain subtle changes can influence impulsive behavior substantially. That’s why Amazon has 1-click-buying to promote impulsive behavior, since they know just adding one extra click can start people thinking “maybe I should’ve bought the other item”, or “do I really need this?”. While Amazon wants to facilitate impulsive purchases, Wikipedia would benefit by not facilitating impulsive behaviors (i.e. revert-first, rationalize-later), and instead try to nudge people into more rational patterns. Just yesterday i saw someone did a revert on another page from an IP address, giving no reason, because current workflows enable that, and someone else had to go and fix that, spend extra time noting they did not give a reason, and this is no doubt repeated ad infinitum. As to which specific subtle UI nudge is more or less successful, this can only be determined via A/B experiments --Thhhommmasss (talk) 18:40, 4 July 2018 (UTC)

Article 11 Copyright Reform and URLs

Is this article [5] accurate about Europe's newly planned copyright reforms, and the problems they mean for Wikipedia? If I'm reading this correctly, URLs will now be subject to copyright, meaning wikipedia may have to remove URLs from its articles wherever we cite news sites (i.e. everywhere). If correct, that is a big problem, but I may be misinterpreting it. Has the Wikimedia Foundation said anything about this proposal? If it does pass, how would we go about adapting to this? My first thought is a bot could be written to remove URLs, and a message could pop up whenever a user adds a reference saying not to include the URL, but article histories would still have them. --HighFlyingFish (talk) 18:06, 18 June 2018 (UTC)

It puts to the member states how to handle the link tax, and only applies to organizations operating in the EU/member state. WMF is an US organization so it may not affect WMF. (I should note that the WMF knows that Article 13 of the same will potentially have a more harmful impact, see [6]). --Masem (t) 18:21, 18 June 2018 (UTC)

Phasing out human editors in favor of bots

"If you think I'm going to improve 250 GA articles this evening at breakneck speed, you need to put down the Elon Musk and stop watching Hollywood blockbusters."

Progress in AI technology has led to AI systems that have good language comprehension and skills. Take e.g. a recent IBM test:

[www.cnet.com]

"We saw computers beat humans at chess in 1997, beat humans at Jeopardy in 2011 and vanquish the world's best human players of the ancient game of Go in 2017. On Monday, a computer edged out a victory over people in a far more nuanced competition: debate."

"To formulate its argument, it had at its disposal a collection of 300 million news articles and scholarly papers, previously indexed for quick search results. But it had to find the information, package it persuasively, listen to its opponents' arguments and formulate a rebuttal."

IBM could probably create a new Wikipedia from scratch that's edited by a similar AI system today. But we would still have an advantage over any such encyclopedia, based on almost 2 decades of editing experience. However, on the long term we'll end up being replaced by autonomous, self-editing encyclopedias unless, of course, we start using AI ourselves. Count Iblis (talk) 16:58, 19 June 2018 (UTC)

  • I, for one, am ready to transcend. --Izno (talk) 17:17, 19 June 2018 (UTC)
Arm yourselves. Tin foil do-rags and SCAR-Ls for everyeditor and their cousin. We will not be oppressed by the lizard men/cabal/Alphabet/Obama/Mnangagwa with their plans to quash the canaille of peasant editors will not be stood for. [FBDB] cinco de L3X1 ◊distænt write◊ 20:54, 19 June 2018 (UTC)
  • I can think of some editors whom I'd like to see phased-out. --Tryptofish (talk) 21:39, 19 June 2018 (UTC)
  • At present, AI systems suck up biased data from the Internet and inherit the bias they find there. White male American chauvinism tends to creep in. And what would an AI make of Elsevier's loony but profitable alternative medicine journals "peer" reviewed by the same community of loony but profiteering academics? General intelligence technology has a high wall to climb before it can relieve me of this editing chore. >sigh< — Cheers, Steelpillow (Talk) 19:02, 20 June 2018 (UTC)
  • Is this suggestion serious? Vorbee (talk) 08:43, 21 June 2018 (UTC)
Is this question serious? EEng 14:31, 21 June 2018 (UTC)
  • Yes - we already have a proposal at Wikipedia: Village pump (perennial proposals) saying that we should have a bot to welcome new users, and a message about why this proposal has been rejected. Vorbee (talk) 15:54, 22 June 2018 (UTC)
  • But I've been activated since ages already! My grammar module seems to be stuck in 'Yoda' mode, though. TP   16:04, 22 June 2018 (UTC)
Surely you mean In Yoda mode stuck my grammar module seems to be? EEng 18:26, 22 June 2018 (UTC)

Why isn't there a bot that adds the Wikispecies template when there's a name match for species articles?

For reasons even I don't know, I do random page patrolling. A lot of times I add the {{Wikispecies}} tab to random species stubs if the link leads to an article. Why isn't there a bot doing this if there's an article name match at Wikispecies for random species stubs here?--occono (talk) 23:18, 19 June 2018 (UTC)

@Occono: that link should already be appearing in the sidebar under "In other projects" if the information is populated to an infobox. The template usage notes (at Template:Wikispecies) say when this is used it should be in the external links section. If you think this should be on every article, some wider discussion is a good idea. I suggest you link in editors from Template talk:Wikispecies and maybe some of the larger species-related projects such as Wikipedia:WikiProject Mammals and Wikipedia:WikiProject Amphibians and Reptiles. If there is a broad consensus for doing this, and how to do it, you can request a bot operator build a bot for it at WP:BOTREQ. — xaosflux Talk 02:11, 20 June 2018 (UTC)

Enwiki's Featured Pictures process

At WP:FPC, editors can nominate Featured Picture candidates. If, at the end of 10 days, there are at least 5 in support with at least a 2/3 majority in support, it is promoted to FP. Whenever I've popped in over the last year or so, it seems like few images are being nominated but also -- and more concerning -- few are receiving the necessary participation. It's not uncommon to see a nomination archived with 100% of participants in support but still not promoted, for example.

On the talk page, a few users raised this problem. Three possible remedies were suggested: advertise FPC with the POTD, lower the threshold/standards, and elongate the nomination period. Pinging talk page participants: @Charlesjsharp, TSP, MichaelMaggs, The Herald, Paul 012, and Kaldari:.

As for advertising it, I'd add that another more inward-facing technique may help too/instead, like a watchlist notice.

In terms of the threshold, this could be taken to mean lowering the number of supports needed (for example, if at least 3 or 4 in support and none opposed, then it can be promoted, otherwise the existing threshold stands).

As for the nomination period, it's unclear what a good extension would be. There should certainly be some time limit. IMO it would be better to encourage people to weigh in one way or the other rather than abstaining (I tend to do this myself).

As many of you know, Commons has its own FPC process. It is similar to enwiki's except it's for 9 rather than 10 days, requires 7 rather than 5 support votes, and weighs technical quality and a "wow" factor at least as much as encyclopedic value whereas enwiki priorizies the latter.

  1. It may be useful to first reaffirm that it makes sense for enwiki to have its own FPC process. Our media is otherwise handles by Commons and nearly all of the material that is on enwiki but not Commons would not qualify for FP.
  2. Assuming it's something we want to continue, how could FPC participation be improved?
  3. Does it make sense to lower the standards for promotion?
  4. Would elongating the nomination period be helpful?

Rhododendrites talk \\ 16:26, 20 June 2018 (UTC)

  • I and some other editors believe that Featured anything (and our "Just decent" articles as well are a waste of time ans serve little purpose either ot the good of the encyclopedia as a whole or to society, and a even greater number think the Commons is nothing to be emulated. Except for your first question, such posts are orthangonal to your purpose, which is why this is small. My opinion for your three questions are We don't, not to me, and probably yes. cinco de L3X1 ◊distænt write◊ 22:48, 20 June 2018 (UTC)
  • Methinks that since English Wikipedia is not picture-hosting site, then all this Featured picture process should be completely abolished and divert the attention given to them to other areas needing such attention. Wikimedia Commons is the dedicated project meant for handling media needs of hundreds of WMF projects and it is working excellently. If any editor is inclined to vote for featured picture it is just few clicks to land on Commons and do so. Wikipedia first and foremost essence is to write.–Ammarpad (talk) 07:54, 21 June 2018 (UTC)
  • Two changes that I would support would be advertising it more widely (maybe a Signpost article would help) and slightly lowering the resolution standard. Over the years (back when FPC was flooded with candidates), we gradually increased the resolution standards from "1000 pixels in width or height" to "1500 pixels in width and height". Could we relax that to "1000 pixels in width and height"? I don't think that reduces the encyclopedic value and it would open up a lot more candidates. I don't think changing the support vote threshold or the length of time is going to have any effect on the number of people participating, which is what we need to improve. Kaldari (talk) 15:05, 22 June 2018 (UTC)
  • @Kaldari: changing that "or" to an "and" is huge, the current minimum would be 1500px2 but that would make it become 1000000px2.xaosflux Talk 19:21, 22 June 2018 (UTC)- scratch that, I see it is already an "and" - which also seems a bit arbitrary - why couldn't a long or wide image be great? — xaosflux Talk 19:24, 22 June 2018 (UTC)
    • @Xaosflux: I would also be fine with changing it to "1500 pixels in width or height". Basically I would support any reasonable reduction in the resolution requirement. Kaldari (talk) 19:42, 22 June 2018 (UTC)
      • Why would decreasing the necessary technical quality attract more participants to vote on images? I'm not necessarily against the measure (though in a time when every cheap phone has a camera capable of 6-10+ megapixels those minimums should be for special cases rather than a standard... I also think something like the QIC requirement of 2 megapixels makes more sense than a particular height/width minimum), but that seems like it would, if anything, make the problem in this thread worse by creating more nominations that our minimal participant pool would be reviewing. Could the number of people who would be active but stay away because they don't like the resolution requirement really be more than one or two? I don't think there's really a shortage of images/photographers used/active on enwiki that could create nominations. The reason they don't, I think, is because it's pretty inactive and/or they don't know about it and/or it's not sufficiently distinct from Commons' FPC process. — Rhododendrites talk \\ 20:16, 22 June 2018 (UTC)

To articulate some concrete possibilities for ways forward:

  1. Better publicize FPC to get more participants.
    • Watchlist notice?
    • Signpost?
  2. Change the promotion requirements.
    • Fewer support votes necessary?
    • Change to a consensus-based system with a minimum quorum of, say, 3 people, and determination of consensus upon closure?
  3. Change enwiki's FPC process
    • Get rid of it entirely, deferring to Commons for POTD, etc.?
    • Better distinguish it from Commons?
      • Rewrite the guidelines to add emphasis to encyclopedic value over technical quality (including a reduction in the required resolution, for example)? Redefine (or otherwise flesh out) EV?
      • Factor in other stuff like pageviews?

Rhododendrites talk \\ 20:16, 22 June 2018 (UTC)

Comment I am against lowering the requirements. There is no lack of potential featured images, just lack of people participating. So the solution is spreading the information. Regards, Yann (talk) 20:55, 22 June 2018 (UTC)
  • Comment agree with Yann and Strongly Oppose lowering the requirements -- Lowering image size just encourages folk to upload heavily downsized-for-the-web images. In 2017 we ideally want images that could be displayed on a 4K TV or monitor, or printed high quality in a magazine. Remember that Wikipedia is a "free content" project, not just a "free to view" project, so the content being re-usable is part of our mission. Many people, who see an image here, will click on it to look at the larger version, and appreciate the high quality of a full sized image. For number-of-votes, if you lower it then it just devalues what the project is worth. The current 5 support votes is pathetically small, and if anything the Commons 7 vote threshold probably should be increased to 10, and the participation there is strong enough to achieve that IMO. 5 people really does not represent the Wikipedia community.
Wrt participation, the problem is that a project like this needs a community who encourage each other and set consensus standards for what is reasonable. Once lost, that's hard to regain. I'm not seeing any strong reason for me to participate here vs Commons. It's like a restricted version of Commons FP, where one has the potentially short-lasting goal of ensuring the image is a lead for some article. I got bored with WP FP being just a repeat of a subset of Commons FPs that happen to be on some Wikipedia page. There's a slightly different emphasis on encyclopaedic value but not enough to make the project worth my time. -- Colin°Talk 11:38, 24 June 2018 (UTC)
Comment I'm of the opinion that Wikipedia's FP is redundant and irrelevant to the project. Commons is full of experienced photographs and Wikipedia is full of experienced writers. Reducing the quality requirements or the reviewers threshold makes no sense to me as it would further decrease the quality of this subproject. Commons' FP already seem to take into account the encyclopedic value, ie everything is categorized and a simple good picture of a not well documented species seems to get approval, maybe this encyclopedic value criteria could be emphasized further in case the technical quality of an image is flawless but it's ever rejected based on lack of wow, though I see plenty of boring-to-me pictures get promoted based on their encyclopedic value already. For anything valuable to the encyclopedic scope but of lesser quality then there is the Commons valued image subproject. --Trougnouf (talk) 21:22, 25 June 2018 (UTC)
Commons POTD has selected too many "beautiful young women" portraits (e.g., File:Veiled_in_Red.jpg, but all the usual tropes: nude young woman, 'exotic' young woman, celebrity young woman, etc.) for me to believe that educational value is always a significant factor. (Ping me if you find an equivalent "handsome young man" portrait among Commons' POTD. I've never seen one.) WhatamIdoing (talk) 21:57, 25 June 2018 (UTC)
I agree that it's not directly redundant/equivalent. Landscape panoramas, macro photos on solid backgrounds, church ceilings, bridges, castles, etc. do much better than pictures of people, events, and other encyclopedic subjects that have less visual impact or that are more difficult to take technically high-quality pictures of. At the same time, EV may be a bigger factor here but it's not different enough that people are really using it. As Charles said in the FPC talk page thread, everything he nominates here is already FP on Commons. For me, I see FPC here as a good place for an image that I think has very high EV but lacks either a touch of technical quality or a "wow" factor that's needed on Commons. My success rate has been pretty mixed, though, so I'm probably not the best judge. :)
As for POTD, I do think that has more to do with the people involved with POTD than the FPC criteria. POTD is typically a single person finding the next available day for an image (a year or two down the line at this point) and putting in a picture they like (that's been featured and hasn't been POTD already). It's the sort of thing that more people could get involved with. Of course, that doesn't change the pool of available FPs to select for POTD, but looking at the pictures featured so far this year, there aren't many portraits. Of pictures of a single individual, there are more men than women and it would be hard to argue any of them (amybe one?) are glamour shots vs. the sort of pictures of women I'd like to see more of on the main page (action sports shots, women in politics, etc.). Regardless, using FPC on Commons doesn't mean we need to use the same POTD. (Though to be clear, I'm not at this time actually advocating to shut down FPC on enwiki -- I'd much prefer it be reenergized/reworked). — Rhododendrites talk \\ 22:54, 25 June 2018 (UTC)
  • I'm strongly against reducing standards. If we continue with the project (and I'm not sure it is necessary) then we should mandate that an image has to be FP on Commons before it can be submitted to En Wiki. Then the technical issues will have been debated and the ONLY issue is then encyclopaedic worth. I've had many images rejected at En Wiki by a couple of votes on purely technical grounds which is just crazy when it is already FP on Commons. I agree with Colin that Commons vote requirements should be increased to say 10. Charlesjsharp (talk) 21:44, 27 June 2018 (UTC)
  • If they're already FP on Commons, enwiki has no real need for a system. We can just have a POTD system that looks for pictures that have EV without going through a second FPC. The only potential usefulness of it is if it's sufficiently distinguished from the Commons process such that one is not just a subset of the other but overlap with some that would qualify on Commons but not enwiki, and some that would qualify on enwiki but not Commons. The more I think about it, the less I think trying to promote it and get more participation is likely to be a long-lasting fix, since in the end it still feels redundant to Commons. I think the decision is whether to shut it down on enwiki or to revamp it to better distinguish it. IMO that would probably look more like (but not exactly like) a jazzed up VIC than Commons FPC. That's vague, I know, but it seems doable to decrease redundancy and increase the size of the non-overlapping parts. — Rhododendrites talk \\ 00:05, 28 June 2018 (UTC)
  • Yes, feed the WP process from the Commons one, or broaden it to included failed candidates. Study the pictures that come out of there, which have already been selected by judges of what a good photo is. Select among them the ones having most illustrative value, for reasons including rarity. Presumably this will be a smaller number, and it will be a smaller job. Alas, that's likely to mean only a small fraction will be of lovely and little-known young women, such subjects being already abundantly illustrated, but be that as it may.

Beautiful young women

WhatamIdoing criticises Commons FP for selecting too many "beautiful young women" and bases this on what is recalled to have appeared on POTD. I'm not involved at how POTD is selected but more importantly is how featured pictures are selected. I recently examined the stats. There are 11,316 featured pictures on Wikimedia Commons. Of these, 490 images are classified as "people", about 100 images of people classified as "historical" and another 100 of people classified as "sports". So around 700 photos of people, which is around 6.2% of all pictures that could appear on the main page. Of these, only four are of nudes: A non-sexual image of a Himba woman from Namibia, a partial glamour nude of Michele Merkin, a b&w upper-body and a recumbent nude. Leaving aside the Himba photo, we only have had three sexual female nude photos pass FP in the 14 years the project has been running.

Remember that an unbiased Featured Picture process can only promote images in the ratio that high quality images are uploaded and nominated. While many images at FP come from photographers on the project, many also come from people transferring images that have a free licence or are in the public domain. There is very very little the FP process can do to influence what people take photographs of, or who chooses to make their work free. The definition of "educational" on Commons is simply that there exists a realistic chance that the image could be used for an educational purpose, not that one might look at an image and the first thought you have is how educational it is. Our licence terms discourage the creation of photographs of people that can be used "for any purpose", rather than sold for a profit and where the publication usage can be controlled. So we have a fair number of images of politicians and sports people, and historical photos, but are hugely deficient in photographs of anyone else.

Consider that the vast majority of photographs in the world are of people, yet on commons FP people are outnumbered by other subjects 15:1. Outside of politics and sport, people photography is heavily biased towards women. Fashion, beauty products, glamour, entertainment, advertising all use mostly young beautiful women. Women users outnumber men on Instagram, and both are predominantly young and beautiful. On iStockPhoto, there are 9,184,475 images classified as Women and 5,844,462 images classified as Men. Yet if you look at the People category, the ratio is fairly evenly balanced, and plenty non-beautiful and older people. Take a look at The Featured Picture Log for 2018. I'd say that was an educational collection of images and certainly not the sort of teenage boy's bedroom poster bias suggested above. If anything, Commons photographers seem to spend an awful lot of time in church, or outside photographing the scenery and wildlife. -- Colin°Talk 08:27, 26 June 2018 (UTC)

Colin, I'm not really criticizing Commons' process. I am saying that I'm unable to identify any expected educational purpose is for this portrait of the non-notable woman with a bit of filmy red fabric draped around her face. It's a truly lovely photo. I believe that it's a useful photo. But I am not convinced that it's an educational photo. For example, she's not actually wearing a functional veil, so you can't use it to illustrate Veil. She's wearing cosmetics, but it gives a poor view, so it won't get used in Mascara or similar articles. The fabric might be Organza, but we don't know, so we can't use it to illustrate the article on the type of fabric. I can't think of a Wikipedia article where this could be usefully placed.
I conclude from this sort of image that Commons does not always care very much about whether POTD images are educational. I'm okay with that: I think that some processes on Commons actually should care more about subjective artistic merit than they do (or, at least more than they are willing to admit to doing). My goal was to say that there may be a small gap between Commons' POV and enwiki's POV on that particular point. If we choose to defer to Commons for POTD, we need to understand what we're getting into. It's a trade-off. WhatamIdoing (talk) 20:19, 26 June 2018 (UTC)
WhatamIdoing, you did base your "educational value" argument on the claim that "Commons POTD has selected too many "beautiful young women" portraits" which I find hard to believe since they'd use up all the "beautiful young women" FPs in about a fortnight. The big big difference between Commons and Wikipedia Feature Picture is the difference between "educational value" and "encyclopaedic value". The former weighs the ability for an image to be used in an educational publication. The latter weighs the ability for an image to be used in an encyclopaedia, and for that it must typically illustrate an article topic or an aspect of that topic. As with all our images, accurately identifying the subject(s) is a big help, whether this is the type of cloth used or a species of butterfly. For me (and I think many at Commons FP) there are three aspects to an image to judge. Educational value, technical quality and artistic quality. An image that lacks all three is out-of-scope for Commons and could be deleted. But an image doesn't have to be excellent in all three. Here the technical quality is very good, though the artistic is a bit unimaginative and I agree it isn't particularly strong on educational value. If you look at our log of featured pictures, I hope you agree that most are much more obviously of educational value. For what it is worth, I've just been reading a book on colours (paint, dyes, etc) and if we knew what dye was used for this red fabric then it would be an excellent illustration -- but that comes from the additional (meta) data not from the image itself. Also you will notice her face has no shadows and her eyes have the reflection of a large rectangular softbox. So it can be used to illustrate those techniques of studio beauty portrait photography. Anyway, that's just one image. In my original comment, I noted that I didn't find the difference between educational and encyclopaedic value to be compelling enough to spend time on both projects. From a personal photographer POV, the latter is really quite constraining: we have a number of photographers who only/mainly shoot and upload images that they know will be useful for the lead image in a Wikipedia article. Which is a shame because that eliminates a whole lot of choices of subject and focus, as well as technique and artistic goals. Wikipedia articles are a rather limiting format for images, and they haven't moved on much since the MediaWiki software was created.
I don't think deferring to Commons FP for the Wikipedia POTD would work. If you have a POTD then it needs to be for an English Wikipedia image, used on English Wikipedia in some significant way. I think Wikipedia neglected its FP for many years. For most of Signpost's life, for example, it listed the Featured Pictures and didn't actually display the pictures, just the lead of the article they were used in. For the Main Page, perhaps a POTD isn't relevant, and you'd be better off with two articles, accompanied by thumbnails. After all, if people only see the thumbnail, then the sort of quality standards imposed at FP aren't really relevant. -- Colin°Talk 21:37, 26 June 2018 (UTC)
If we include "illustrating a photographic (or other imaging) technique" as "educational value", then Commons could drop that criterion entirely from their POTD, since every technically strong image will be "educational" in that sense.
And yes, even a small fraction of images having no clear educational content is "too many" for me to believe that they always care about educational value. It's not too many for them to choose; it's only too many for me to believe that educational value is always a significant factor in the decision. Based on the results, I could would only feel comfortable saying that their images "usually" have educational content. WhatamIdoing (talk) 23:46, 26 June 2018 (UTC)
WhatamIdoing, I agree that illustrating a photographic technique (or mistake!) is generally a weak form of EV. However, the EV for that is diluted by the quantity of images that would demonstrate that technique well. Many images show no more technique that arranging the subject in the centre of the frame and pressing the shutter button, so we have millions of them. My own photo File:Bluebells ICM, Ashridge Estate, 2015.jpg failed twice at Commons FP because some argued it lacked EV but actually passed Wikipedia FP without trouble -- just because it is a good illustration of a camera technique (ICM). I had rather hoped Commons would see the EV in the impressionist mood of an English bluebell wood, and some did. We don't actually have that many good quality high resolution studio-lit photos of models in makeup, and yet that is a huge huge domain of photography. I think you are placing too much emphasis on your impression gained from a couple of pretty young women appearing on POTD and not really fairly judging the The Featured Picture Log for 2018.
To counter your complained that Commons EV is unsatisfactory, I would claim that Wikipedia's judge of EV is also unsatisfactory: it requires that an image is in use in an significant way in an article (typically the lead and certainly not in a gallery). So take my very high resolution photo of Tower Bridge. No argument this has high EV and I hope you think it ticks the boxes for technical quality, composition, etc. It was a solid FP on Commons. But it isn't used on the Tower Bridge article and merely appears at the bottom of the City Hall, London article as a view from. At thumbnail size, it competes with other potential views (at night, from the air, historical), and so is rejected because the article can only contain so many thumbs. So according to Wikipedia FP, the EV as an illustration of Tower Bridge is absolutely zero. Which isn't right. -- Colin°Talk 08:12, 27 June 2018 (UTC)
I think y'all are looking for Commons:Village pump. Not sure what bearing this discussion has on English Wikipedia. Kaldari (talk) 22:03, 30 June 2018 (UTC)
We definitely do need more pictures of ugly old men, that's for sure. Oh, wait, isn't that who most of our bios are about?  — SMcCandlish ¢ 😼  02:14, 3 July 2018 (UTC)

Indicating source preponderance

One of the biggest problems I have with Wikipedia is its misuse to 'broadcast' FRINGE, POV, OR, or ESSAY views as 'truth' to the world. Anyone can invent/prefer any fringe view, and select (cherrypick) only the facts and sources that support that idea, so to the reader it understatedly seems to be widely accepted fact. There doesn't seem to be much of a mechanism in place to counter this, as researching the full context/scope of a situation (to determine whether a propos is FRINGE or not takes time and, most often, prior knowledge of the subject at hand. Often such interpretations of reality belong to a fringe 'group' (dogma, worldview, etc.), and sniffing this out (because, where fringe views are concerned, this info is most always excluded) makes things even harder.

Yet one indicator of a fringe view would be the preponderance of, and quality of, references supporting that claim. Is there any sort of source evaluation tool/effort in existence? If there isn't, perhaps there should be.

In any case, I do see SOAPBOXing as one of Wikipedia's biggest headaches and reasons for loss of credibility. TP   16:20, 22 June 2018 (UTC)

Wikipedia:Fringe theories/Noticeboard and Wikipedia:Reliable sources/Noticeboard come to mind. – Finnusertop (talkcontribs) 20:46, 22 June 2018 (UTC)
But that's just it: a case where X claim has X amount of sources whereas a Y counter-claim only has a Y amount of sources (yet the article presents Y as 'the whole truth') doesn't necessarily have to be FRINGE, it's more a... lie through omission. Or, for example, there's SYNTH, where one can concoct any story through a 'cocktail' of individually-verifiable facts. Both of these seem to fall through Wiki-oversight cracks. TP   08:25, 23 June 2018 (UTC)

If an article can only be read as original research, it can go to Wikipedia: Articles for deletion. Vorbee (talk) 10:35, 23 June 2018 (UTC) This can also go for Soapboxing - after all, Wikipedia is not meant to be promotional or advertising. Vorbee (talk) 08:05, 24 June 2018 (UTC)

Take, for example, the State atheism article: only a few of its (hundreds of) references even use the term... its entire purpose is to 'confirm' the term (and the accusation it contains). Yet all of the events mentioned actually happened (so seem referenced correctly); the fact that only the article title (and not the source) classifies those events as State atheism seems to have made it past several RfCs and other inquiries.
So if one sees that sort of thing happening, one should just propose an article deletion? That's a bit drastic. TP   20:50, 24 June 2018 (UTC)
Your concern is fundamentally about WP:DUE weight, and different people sometimes have different, but still legitimate, notions of what's balanced.
The best course of action depends upon the outcome that you want. If you want the article cleaned up, then your best bet is to find some good sources, click the Edit button, and get busy. Adding content and perspectives (rather than removing the 'wrong' POV) is usually easier (but not always). If you have tried that, and improvements are being blocked, then an WP:RFC or a trip to WP:NPOV/N might be the best ways to get advice from others. If you want someone else to fix it, then look for a relevant Wikipedia:WikiProject (almost always listed at the top of the article's talk page) and ask for help. WhatamIdoing (talk) 22:03, 25 June 2018 (UTC)
Thanks, but my concern isn't about (only) DUE, it's more in the line of NOTESSAY or perhaps COATRACK; frankly, I don't know. What I do know is that almost none of the sources cited call whatever they're describing 'State atheism' (the article authors are doing this, and that's frankly OR), and it's a case of wikipedia being misused to trumpet-pin the blame for history's worst atrocities on 'atheism' (which would be SOAPBOX), which is a trait common to apologists and anti-atheists (in a fallacy called 'the atheist atrocities fallacy', and the article mimics it almost perfectly), a POV completely absent from reliable and mainstream historical sources, yet the article pretends that it is 'common knowledge' without any reference to the source (and that it is opinion, not fact).
Going to any one of the 'specialised watchdog' panels will only result in ineffective and partial measures. But if data about the source cited were made available, this would nip any of the above in the bud (as the 'fringe' POV would become clear); but, in the meantime, if you have any further advice what to do about articles like this, I'm all ears. Best, and thanks. TP   17:39, 29 June 2018 (UTC)
PS: and such behaviour isn't limited to topics such as that: I spent ten years attempting to thwart a very persistant contributor (and their off-wiki MEATPUPPETs) attempting to SOAPBOX Paris as a skyscraper-filled city twenty times the size it really is (in glossing over its poorer suburbs, etc.) in 'citing' an obscure demographic statistic tool that almost no-one living in the city has ever heard of... all measures were ineffective, mostly due to administrator unfamiliarity with the subject (as they can't tell reliable sources (for the subject) from non-reliable ones... this would take a lot of research). I've been seeing this sort of thing going on across many topics since years now, and have pretty well given up on admin-side intervention in such matters, so my input here is with the goal of finding a possible client-side solution. TP   17:51, 29 June 2018 (UTC)

Mastodon instance?

Mastodon is a federated social network that emphasizes microblogging. A few large communities & institutions have Mastodon instances of their own, such as MIT, the Chaos Computer Club, and LQDN. I think it would be a good idea to have our own instance, possibly at social.wikimedia.org (or mastodon.wikimedia.org). Thoughts? Enterprisey (talk!) 20:19, 27 June 2018 (UTC)

Could you provide some details regarding what the benefits/impact of doing that might be? --Yair rand (talk) 02:16, 28 June 2018 (UTC)

My first thought was that if one goes to Wikipedia: What Wikipedia is not, one can go to Section 2.5 to see that Wikipedia is not a social networking service. Vorbee (talk) 08:01, 28 June 2018 (UTC)

As the domain name social.wikimedia.org implies, this would not be part of Wikipedia, and instead be another project hosted under wikimedia.org. Enterprisey (talk!) 13:40, 28 June 2018 (UTC)
But why?  — SMcCandlish ¢ 😼  02:15, 3 July 2018 (UTC)
An improved UI and outreach tool for editors and new ones alike to discuss articles would be a worthy endeavor. Talk pages require WikiMarkup knowledge and have very old school forum style. I would be concerned about splitting communication, but the idea of having improved collaboration/communication is a good one. We organize hackathons already to improve content of articles, why not harness that energy online with first class support/search for wiki articles? Something say facebook, twitter, do not sufficiently help with. Shushugah (talk) 12:24, 12 July 2018 (UTC)

Color coding in source editor

In the source editor, it would be neat if references were color-coded. Ie everything between two ref tags is colored red or something. It will make things so much easier. Kurzon (talk) 11:57, 29 June 2018 (UTC)

Much less of a problem if you use short form refs and keep the full citation for a bibliography. This readability in running text issue is a key argument in favour of Harvard-style referencing. Martin of Sheffield (talk) 12:10, 29 June 2018 (UTC)
There are some WP:Syntax highlighters available. --Izno (talk) 12:46, 29 June 2018 (UTC)
Kurzon, look for a "highlighter marker" (Codemirror-icon.png) in the toolbar. It probably does what you want. WhatamIdoing (talk) 22:25, 29 June 2018 (UTC)
Shout-out to WikEd for collapsing references. — Dispenser 13:09, 30 June 2018 (UTC)

New user right request

I would like Wikipedia have a new user right. I want it called Signature manager, but the user right I want to see on Wikipedia is about the ability to modify other people's signatures. I want to have the modification tool at either Special:SignatureManager or Special:UserSignatureManager and have it logged at either Special:Log/signature or Special:Log/signatureother. 209.52.88.255 (talk) 18:52, 4 July 2018 (UTC)

I'm glad that you saw my response to your message and decided to come here to discuss your thoughts. Why do you believe that this user right is needed? What would having this user right developed and available to be granted to users accomplish? ~Oshwah~(talk) (contribs) 18:54, 4 July 2018 (UTC)
This right might be useful because people can fix typos such as typos in CSS or attributes such as letter case in classes, CSS color, such as ref to red (assuming they're using a QWERTY keyboard), or times when they can't decide between one signature or another, or uses like User:Cyberpower678/SignatureColorKey, where the wrong color is added to the signature in a sign. This user right should also always display a preview of the other's signature first, which wouldn't include the timestamp. 209.52.88.255 (talk) 19:09, 4 July 2018 (UTC)
Hi, first only registered users can have permissions so this would not be something you could use directly 209. Generally "signatures" are just wikitext on pages and are not protected. If someone is using a signature that is causing disruption (such as introducing Special:LintErrors) they can be asked to stop or be blocked under general disruption. Being able to forcibly change someone else's preferences (such as signature) is mostly a non-starter, as it could lead to that person making edits they did not expect to make. — xaosflux Talk 21:09, 8 July 2018 (UTC)
I think phab:T178879 and phab:T140606 are much more likely and reasonable than allowing people to change other people's signatures. --Izno (talk) 21:28, 8 July 2018 (UTC)

Main Page suggestion

The French Wikipedia's main page currently hides the "Welcome to Wikipedia" banner in the Timeless skin if the screen/window is narrower than 850px, using the nomobile CSS class in Timeless. Would it be a good idea to hide the English Wikipedia main page "Welcome to Wikipedia" banner at small screen widths (for all desktop skins)?

Alternately, only the portal links could be hidden at some screen width, with the "Welcome to Wikipedia, the free…" text becoming aligned to the centre of the banner. (Both options should be possible with TemplateStyles in about a week and a half.) Jc86035 (talk) 11:56, 10 July 2018 (UTC)

When we have template styles, we can fix this properly much easier. I suggest we wait 2 more weeks for that. —TheDJ (talkcontribs) 12:03, 10 July 2018 (UTC)
@TheDJ: Would the .css page be started from scratch entirely, or something like Edokter's main page CSS (from the 2016 redesign) also be used? Jc86035 (talk) 11:27, 12 July 2018 (UTC)
Whatever we want, though we should probably be conservative considering the main page redesign history. Just move part by part into a stylesheet and then slowly rework with the new possibilities we will have. Here is an example of CSS for the main page of Mediawiki.org. —TheDJ (talkcontribs) 11:34, 12 July 2018 (UTC)

WikiProject participant lists

Rather than have editors self-assign themselves to project participant lists, I think it would be better have an automated tool that listed active editors. Ideally it would be configurable by individual projects (e.g.: the top 50 editors or editors with 10+ edits), but the tool itself would only run monthly, similar to the popular project pages. I have no knowledge of how to make it happen, so I'm here. –Fredddie 01:15, 11 July 2018 (UTC)

@Fredddie: This already exists at e.g. Wikipedia:WikiProject Directory/Description/WikiProject Video games. --Izno (talk) 03:27, 11 July 2018 (UTC)
Wow, I need to get outside of my WikiProject more. Thanks! –Fredddie 05:04, 11 July 2018 (UTC)

Giving schools the ability to access wikipedia with the option of disabling editing

While talking with a friend who works at a school, he told me how they set up a local wiki (which is view only) since they have students who often vandalise Wikipedia pages otherwise: they are too immature to know better and can't be reasoned with (say below 6-7th grade).

I think it would be great if we, on our part, somehow gave the entire school access to it with an option of disabling editing when needed. Plus point is: If the school feels the need to say, teach them how to edit, they can enable it whenever they're ready. I'd imagine not all schools would do what my friend's school did, they probably block the entire site.

I have not much clue on the technical feasibility of this idea. Ugog Nizdast (talk) 10:44, 12 July 2018 (UTC)

A couple of practical points, how does the systems manager distinguish between a staff account and a pupil account?
When I started WP, anyone could edit and that founding principle established the WP as the worlds greatest source of knowledge. The people who rely on it, and enjoy using it are determined to place restrictions and limitations on who may get involved - it is a case of 'look but don't touch'.
Schools also used to be places where thinking was developed- not where conformity was imposed. Is it correct to teach children that what they see in WP is the only truth, and they haven't the right to be consulted?
If we are talking about selection by maturity, ability, political persuasion I am sure the majority of WP wouldn't trust their own government to run a bath (WP:POV).
Kids experiment, but it isn't too onerous to keep looking at our watch lists. Kids experiment but then start makng responsible edits when they find out it is more fun ClemRutter (talk) 11:42, 12 July 2018 (UTC)
Was just watching this topic, but cannot ignore the silliness there. Any systems manager will be easily able to distinguish between staff and pupil accounts. It has happened in every school I have worked in. (And that's a lot.) And nobody teaches children that what they see in WP is the only truth. HiLo48 (talk) 11:47, 12 July 2018 (UTC)
(edit conflict)Entities (such as schools) that have control of their network can effectively block editing or logging in a number of ways (most of which involve implementing an ssl intercepting internet filter). — xaosflux Talk 11:49, 12 July 2018 (UTC)

Better filtering of "own work" image uploads

There's lots of images uploaded to en and/or commons which are tagged as "own work", but clearly are not. See this conversation on my talk page for a recent example. In this case, I believe the uploader honestly didn't understand that what they were doing was a copyright violation. In many other cases (paid spammers), they simply don't care. In either case, we need some better filtering.

In most of the cases I see, the uploaded images have no EXIF data. There can be legitimate reasons for this, but in the overwhelming majority of cases, it's a clue that somebody deliberately stripped out the EXIF copyright statement. Virtually every camera in existence puts in at least some basic EXIF boilerplate automatically. I strip out the copyright EXIF data when I upload images to commons, but that's on purpose; I have my camera set up to automatically insert my copyright, but for stuff I'm putting on commons, I want to disown that. I do, however, leave the rest of the EXIF data, including the Author. For example, File:George_Faile_Gravesite.jpg.

So, I'm thinking we should require images which appear to be photographs to have EXIF data. Or, at least, detect that they don't and require the uploader to enter an explanation of why they don't, and flag those for human review as likely copyvios. -- RoySmith (talk) 12:59, 12 July 2018 (UTC)

If it were required spammers would probably just start adding fake EXIF data (or stop stripping it out). It might be better to have the software warn the user, but not tell them why and not prevent them from uploading the image. Jc86035 (talk) 13:20, 12 July 2018 (UTC)

Concept for a new logo for Wikipedia

Here it is: [ibb.co] Anyway, see you later! Peppa Pig the Second (talk) 14:08, 13 July 2018 (UTC)

Running our own archive service?

Has any thought been giving to running our own archiving service for references? Right now, some people use archive.org, that doesn't seem very reliable. I can't actually remember the last time I got it to work, and I've given up trying. As reference links go dark, we're losing a major part of the value of the encyclopedia. Even if it was reliable, by having our own service, we would be able to have a cleaner integration with the editing tools. -- RoySmith (talk) 18:49, 13 July 2018 (UTC)

@RoySmith: Does Save Page Now not work reliably? I find it doesn't work on some pages but those usually work with archive.is or [webrecorder.io], although I am inexplicably only able to access the former through the Tor network (and through Webrecorder). Webrecorder requires an account and content is attached to that account (which has a limited amount of space), but can capture dynamic content like YouTube videos and flash games. Jc86035 (talk) 20:49, 13 July 2018 (UTC)
I honestly don't remember how I used to access the archiver. What is "Save Page Now"? -- RoySmith (talk) 02:25, 14 July 2018 (UTC)
@RoySmith: If you go to [web.archive.org][your.url.here], you can save most URLs and most of the embedded content used on those pages. You can also access this through the web.archive.org landing page, by typing in a URL. (Or were you referring to accessing saved URLs? I assumed you were referring to archiving pages.) Jc86035 (talk) 05:48, 14 July 2018 (UTC)
Personally I think that archive.is runs a better service. It saves pages more accurately than archive.org and WebCite and has good uptime. It would be a major undertaking for Wikipedia to offer a similar service, so it looks like we are stuck with third party services for the time being.--♦IanMacM♦ (talk to me) 05:57, 14 July 2018 (UTC)
Managing permissions, rights and licenses for archived material looks like a major headache. (Copyright law varies from place to place.) Wikipedia avoids such issues by having a CC-BY-SA licence for everything. This said, archiving cited material would obviously make Wikipedia much more resilient, so the idea should at least be studied. Aerkem (talk) 11:40, 14 July 2018 (UTC)
If we do not run own service we could improve the page at Help:Archiving a source. It seems like you lot have views on which service are better and they could be included in that article. Emir of Wikipedia (talk) 17:04, 15 July 2018 (UTC) P.S.: Please ping me in your reply. Thank you.
Help:Archiving a source was created a few weeks ago. IMO it's a stub fork of the long-standing WP:Link rot which itself needs an overhaul I started on at Wikipedia:Link_rot/new but haven't had time to finish. -- GreenC 14:37, 16 July 2018 (UTC)

This has been raised a few times, but [at least in those threads I've seen] hasn't led to any sort of formal proposal. It's certainly something that would need to involve WMF tech folks. I'd be surprised if they would be game, since we have such a good relationship with the Internet Archive. I definitely do not think we should be relying on the others, like archive.is. That's not because it isn't a good service, but because it's largely a mystery. AFAIK it's the work of a single person who has expressed that he doesn't know how long it will be sustainable. It has also been blacklisted here in the past (not currently) because of spamming and general sketchiness. Most importantly, we don't know anything about these organizations sufficient to place faith in them to prevent link rot when the Internet Archive, by contrast, has a clear organizational structure, large user base, and is generally well-known as an entity beyond just the service itself. There are a bunch of ways to archive pages and access archived pages, and if those aren't working for you that problem is most likely on your end. It's slow sometimes, for sure, but across all of my computers over the years I don't think I've had any trouble. We also have bots that can do this automatically such that there's limited need for other editors to do so. What's lost in the Internet Archive, however, are those things that prevent archiving via robots.txt. Archive.is ignores robots.txt, and so may make sense to use in those cases when it's the only option. If Internet Archive isn't willing to bypass robots.txt, it's unlikely WMF will be. Scattered thoughts, sorry. — Rhododendrites talk \\ 18:00, 15 July 2018 (UTC)

You bring up a good point about the organizations. I think Perma.cc is probably the one I would put the most faith in due to its backing from universities in multiple continents. {{ping|Chetsford|GreenC|Cyberpower678} You three have discussed on the Perma talkpage so you might want to contribute here. Emir of Wikipedia (talk) 21:13, 15 July 2018 (UTC)
@Chetsford, GreenC, and Cyberpower678: fixing the ping format.
OK, I just tried the suggestion above, with [web.archive.org][your.url.here]. It did indeed work. I don't remember exactly what problems I had the last time I tried it. Be that as it may, the process is still pretty rough. I remember when I first tried this, I found my way to ((IIRC) Help:Archiving a source. That page doesn't tell you how to do it. It gives you a choice of five ways to do it. So, the newbie user is immediately faced with having to make decisions.
My suggestion about doing our own was not so much about where the data would live, but about controlling the process so we could build simple interfaces. I'm fine with using one of the other services.
But, we should pick just one, commit to it, document how it's used, and build interfaces to it which are dead simple (perhaps, totally automatic, and integrated into the visual editor). -- RoySmith (talk) 00:09, 16 July 2018 (UTC)
Rhodo, IA as of a year or two ago is ignoring robots.txt. --Izno (talk) 00:20, 16 July 2018 (UTC)
Whoa. If that's the case, we should really start to replace archive.is links sitewide. @Beetstra: (someone I remember being part of previous archive.is threads) is this something you're aware of? — Rhododendrites talk \\ 03:04, 16 July 2018 (UTC)
@Rhododendrites: No, I wasn't. Maybe it is then time to replace the links. I am however also very interested in whether we can have our own archiving system (in a way, WikiSource is such a system, for out of copyright stuff that is). --Dirk Beetstra T C 09:59, 16 July 2018 (UTC)
@Rhododendrites: Wayback also ignores robots, except in certain situations so it would be unfair to target one provider or another as Wabyack itself is not consistent ("not consistent" broadly speaking even within its own policies of what to ignore or not). Robots.txt was not designed to determine if a page should be archived or not it was sort of a hack mechanism. Why do we care if a provider ignores robots.txt out job is WP:V and if a provider has the page better for our readers. -- GreenC 16:05, 16 July 2018 (UTC)
  • I just wanted to note that as part of the Knowledge Integrity program at the WMF, we're aiming to work with the Internet Archive to set up a process of near-instant automated archiving of all citation links added to Wikimedia projects this year. You'll be able to follow our progress on this at phab:T199193, though the phab board is still in the process of being fleshed out more fully. Samwalton9 (WMF) (talk) 10:57, 16 July 2018 (UTC)
    • Thanks for that link. I was previously unaware of that effort. -- RoySmith (talk) 11:56, 16 July 2018 (UTC)

Through my bot WP:WAYBACKMEDIC I've worked with dozens of web archives warts and all. There is no single best archive service, they each have pros and cons. All other things equal, we default to archive.org but they don't have everything. The full list of services is WP:WEBARCHIVES, its around 20 or so with more to be added.. we have a rich and diverse group to draw from. I could list many problems with archive.org and I'm concerned about over-reliance of them, but they have the broadest coverage. I would support Wikipedia having its own archiving service, so it can solve the problems other archive services cant/wont solve, so we can tailor the system to the needs and requirements of Wikipedia, to prevent loss of links by providers outside our control, add new features and services. From a technical standpoint the current system of third parties is incomplete, error prone and difficult to manage. The technology in use at webrecorder.io (open source) is probably the future of archiving, there is no way around it with modern websites moving to JS, video, website aging, etc.. each day that goes by without this technology deployed against Wikipedia is a forever loss of cites, it's sad to see no one doing anything about it. User:Jc86035 wrote a good summary of the situation at T199193. Waiting for a third party provider to solve it for us may or may not be the best thing for Wikipedia, another example of why using third party providers is a problem (keeping up to date with technology). -- GreenC 15:55, 16 July 2018 (UTC)