Wikipedia:Village pump (idea lab)

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:

Discussions of technical issues belong at Village pump (technical).
Discussions of policy belong at Village pump (policy).
If you're ready to make a concrete proposal and determine whether it has consensus, go to the Village pump (proposals). Proposals worked out here can be brought there.

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.

« Archives, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61

Encouragement to proofread

As someone who has spent years correcting typos on Wikipedia, I have found many errors that could have been avoided by the original author proofreading his creation. Can we not do something more to encourage contributors to proofread? Or even automatically suggest corrections as WP:AWB or Grammarly do? Even a pop-up window upon submission of a new article or a notice to a talk page would likely be helpful. --LilHelpa (talk) 11:32, 12 July 2017 (UTC)[reply]

Hi LilHelpa I might be one of the people you are talking about and I agree somewhat. However for folks like me who are translating a page across wikis and sometimes simultaneously improving that article, the folks like you who do those little cleanups are an amazing help!. The cognitive load for some tasks like these requires frequent breaks. When I do return afresh to do some proofreading, some saint has already helped out. So that makes my job more efficient. As long as some kind of prompt to proofread didn't compete for my attention while I was doing step-wise translations (i.e. one sentence/one paragraph per edit at a time) then this would be great. Dr.khatmando (talk) 04:45, 13 July 2017 (UTC)[reply]

Thanks, Dr.khatmando as you make some valid points. I don't see much value to nagging every edit step. I do understand the difficult task of translating which you describe... and thank you for accomplishing that. Copy editing is also a task that needs to be efficient. To that end, many do not seek to cooperate in the slightest. --LilHelpa (talk) 12:12, 13 July 2017 (UTC)[reply]

LilHelpa Indeed. I know we value the same things have the same goals in mind. Dr.khatmando (talk) 12:32, 13 July 2017 (UTC)[reply]

Not everyone can proof read well, most of us have problems spotting our own mistakes, and I suspect many of our contributors are doing so in a second language and part of their motivation is getting people like LilHelpa and myself to fix their mistakes. The point about crowdsourcing is that people contribute content and others improve that, I don't see any benefit in making it harder for the content contributors, or rather if there was one change I'd make to contributors it would be to prompt for a sources where they appear to be adding unsourced content. Ϣere SpielChequers 11:10, 13 July 2017 (UTC)[reply]

While sourcing is certainly more important (and more difficult) than proofreading, it is because of the ease of proofreading that encouraging proofreading may have a better return. Even conscientious article authors often do not carefully review their work. Nobody knows better than I that people make mistakes... and I agree that even a careful proofreading may miss some. However, it is those "contributions" that ignore capitalization and spacing (and IMHO could be considered vandalism) that are most trying. Somewhere in between are lazy editors who need a small push and encouraged to take more pride in their work. It doesn't have to be harder for the contributors (as with a pop-up), it just needs to be more portrayed as valuable. Cleanup needs to be efficient, too.--LilHelpa (talk) 12:12, 13 July 2017 (UTC)[reply]

It would be good to get a study done on this, but my experience is that many individuals do pay attention to what happens to their contributions and learn from the correction of their mistakes. Typos that I patrol will usually shift genres as people editing in one subject area take note of the corrections I have made and new people start. We could of course make preview a mandatory phase before saving, but I suspect the WMF would very sensibly veto that as an extreme measure that would undermine open editing and make people's first edit a bit more difficult and complex. Ϣere SpielChequers 06:19, 14 July 2017 (UTC)[reply]

Typos are one of those things, like dab links, which appear very significant when you work through a massive database report that lists thousands of these errors, but which are really one of the last things we should worry about. Not that fixing them isn't important – I hugely appreciate (and have benefited from) the efforts of the editors who fix them. But when you're spent ages researching a topic and then drafting and redrafting your text, and when you've then gone over it alongside the sources to double-check you've represented them fairly, and when you've then edited your text for readability and made sure all the wikilinks go where they're supposed to and that all the ref anchors work, after all this two things are certain: 2) it's very ease to become incapable of noticing your own typos, and 2) the last thing you need at this stage is having another hurdle to stop you from publishing your work. And as for the editors who consistently make gross errors with spacing and capitalisation, well, I think these errors aren't going to be among the most problematic aspects of their contributions. Text that looks really bad should either be removed or left as it is, the character of its spelling and punctuation being a self-referential indicator of the quality and reliability of its content. – Uanfala 18:10, 22 July 2017 (UTC)[reply]

Dispute resolution RfC

Hello. You are invited to comment on this RfC, which seeks to reform certain aspects of Wikipedia's dispute resolution processes. Biblio (talk) 15:47, 23 July 2017 (UTC)[reply]

Seeking feedback on addition to What Wikipedia is not: Wikipedia is not the final arbiter of truth

I'm looking for feedback on my initial draft of a proposal for a new section added to the policy What Wikipedia is not. It distills the long established practice of not taking sides in disputes, and giving due weight to significant dissent. I've noticed that sometimes IPs and SPAs are on a mission to change an article to take unresolved questions and resolve them. Marketers often want articles to say definitively that their company's product is the leader in its category. Quite a few want ownership of inventions assigned unequivocally to one nation, and no other. Many are partisans, many are just using Wikipedia for its most popular purpose: answering questions. When it's for a trivia quiz or to settle a bet, many readers won't accept anything but an either/or answer, when the sources are telling us is "it depends".

Wikipedia is not the final arbiter of truth

Wikipedia is not written to settle bar bets.
The policies requiring a neutral point of view and verifiability, and prohibiting publishing original thought, mean that Wikipedia editors, and the articles they write, cannot judge who is right in an entrenched disagreement. Articles focus mainly on facts that are most widely accepted and established, giving little space to on fringe ideas, but nonetheless, significant dissent from mainstream views is given proportionate attention. When views are divided, Wikipedia struggles to accurately portray that division, not oversimplify it.
Wikipedia cannot resolve questions that established experts have not themselves fully resolved. The ambiguities and contradictions of real life cannot be artificially made simple and tidy by Wikipedia. Articles can strive to give simple explanations for complicated concepts, but they cannot do away with complexity itself, nor make a roundabout series of events into a straightforward narrative. This does not mean Wikipedia should present facts as if they were opinions, only that Wikipedia does not add weight to the judgement of reliable sources.
An encyclopedia article is not the place to find the definitive answer to the question of whether the motorcycle is a German, French, or American invention. Wikipedia can verify that for a long time, most mainstream authorities agreed it was the Daimler Reitwagen, but the muddy and complicated picture made by reputable dissenters who bring up earlier steam motorcycles cannot be neatly cleaned up by Wikipedia editors. The International Astronomical Union's mission may include announcing, once and for all, that Pluto is not a planet, but Wikipedia's mission is only to describe the IAU's statements, and the significance and influence the IAU represents, but not to give or withhold a Wikipedia seal of approval. An encyclopedia of unlimited size can give a plot summary and production details of the Friends episode that featured the practice of going commando, but Wikipedia cannot change from opinion to fact the assertion that Friends is almost solely responsible for popularizing the slang term for not wearing underwear.
Changing an article, or requesting that it be changed, to give answers unambiguous enough to settle your bar bet is contrary to Wikipedia's core principles.

I threw in a few colorful illustrations from my own editing experience. I'm sure there are some issues of this type that I'm not familiar with. I'd like to draft a proposal that doesn't get bogged down in choosing the perfect wording, and can simply gauge consensus for or against the basic idea of this kind of addition to WP:NOT. Thanks in advance! --Dennis Bratland (talk) 06:27, 24 July 2017 (UTC)[reply]

Dennis, if you don't mind, what distinguishes this from Wikipedia:NOTTRUTH ? ☆ Bri (talk) 22:24, 24 July 2017 (UTC)[reply]

NOTTRUTH is an essay, this is a proposed amendment to a policy. Jo-Jo Eumerus (talk, contributions) 07:36, 25 July 2017 (UTC)[reply]

I think it's fair to say it's related to the essay Wikipedia:Verifiability, not truth. But I'm totally serious when I say this is about bar bets, or similar trivial hairsplitting. People like to win contests, or pick winners, over what is the first or best or which country invented the whatever. And they want Wikipedia to settle those questions. They want answers, not shrugs. And they want short answers, not long, complicated explanations that raise more questions than they answer.

Just like people turn to Snopes to debunk internet memes. Or PolitiFact or FactCheck.org to adjudicate public statements. Perhaps I should adjust it to include "Wikipedia is not your Snopes, or PoliticFact or FactCheck.org." It's not that we don't dabble in that, for example at List of common misconceptions. The central point is that we don't feel obligated to issue a ruling on every question. Many of these "lie checkers" have a bias toward giving some kind of answer, even if the data doesn't fully support a definitive conclusion. So the driving force behind my proposal is the No original research policy. Fudging, or erring, or ignoring contrary evidence, ignoring dissenting points of view, because we want articles to offer clear solutions or conclusions or judgements not fully supported by the sources.

Or because we want to define or categorize things. I guess you could call me one of the editors who thinks Wikipedia:Categorization is broken. It is mostly wonderful, but there is pressure to pigeonhole things into one category or another, but not both, not neither. "A place for everything". What if nature doesn't give us things that have a place? What if our taxonomy is flawed? Often, the taxonomist isn't a reliable source, it's a Wikipedia editor. Creating a taxonomy out of whole cloth violates WP:NOR. What if the sources don't tell us what category something belongs in? The intent of this WP:NOT addition is to offer support for editors who resist forcing an article into a category for the sake of sticking it somewhere rather than nowhere. I don't have a proposal that would fix the categorization policy, but I think this small addition to WP:NOT could help editors resolve some disputes, and help educate the general public about what we really do. --Dennis Bratland (talk) 01:23, 27 July 2017 (UTC)[reply]

Is this even necessary? Most of this seems already covered by WP:NOTSOAPBOX. —Farix (t | c) 01:41, 29 July 2017 (UTC)[reply]

Yes, this bears a relationship with WP:NOTSOAPBOX, as it does with WP:NOTTRUTH. Thanks for bringing that up.

The difference is that soapboxing or promotion means one supports one point of view, and opposes others. Editors seeking clarity and simplicity, or answers that can let them score trivia night, don't necessarily care which side 'wins'. They aren't even that concerned with 'truth'; as long as they are given one and only one answer to their question, giving that answer the Wikipedia seal of approval. Similarly, editors who definitively categorize a topic when the sources don't support such certainty aren't using Wikipedia as a soapbox or means of promotion. They just want things to be neat and tidy, and perhaps they think Wikipedia's policies encourage neat and tidy versions of facts, and discourage uncertainty.

There are are often overlong, and redundant, discussions that usually reach the same conclusion, since the NOR, RS, and V policies don't support pat answers unless sources to justify it. These redundant discussions might be avoided if WP:NOT said explicitly that Wikipedia is not here to give easy answers, and policy doesn't favor either certainty or uncertainty, only faithfulness to sources. --Dennis Bratland (talk) 19:26, 29 July 2017 (UTC)[reply]

Aside from whether you agree this proposal should be adopted, does it have much chance of success written in this form? Is it too long? Do the examples (motorcycles, going commando) help or not? Would it be better to make two proposals, one seeking support for adding some kind of 'not your Snopes' addition, and a second one if that succeeds to settle on the wording? Or does it work better to just ask for up or down consensus on the proposal as written? Thanks! --Dennis Bratland (talk) 19:26, 29 July 2017 (UTC)[reply]

Reading List

I often find myself stumbling upon Wikipedia articles that I would like to read, even though I don't have time to. To resolve this issue I would suggest the implementation of a basic "read list" feature. This would just be a simple (ordered by priority or date added) list that is tied to ones account. It would be added to and then removed from once the user has finished with each respective page. When the user then views a page again, they can see whether they've marked it as read. This kind of idea could then also be extended in a variety of ways. I also fail to see any technical constraints that may be involved. --Sherbet-head (talk) 14:54, 24 July 2017 (UTC)[reply]

This can be requested at Phabricator, see Wikipedia:Bug reports and feature requests for how to request it. עוד מישהו Od Mishehu 07:00, 25 July 2017 (UTC)[reply]

Such a thing already exists (at least in the android app), with the same name as the heading here. See:

It did exist on the desktop as mw:Extension:Gather but was killed due to wikidrama. It will probably go next to the mobile web. 11:19, 29 July 2017 (UTC)

Do you mean having a summary or abstract at the top of each article? It is a good idea and I rather like it. Vorbee (talk) 17:52, 29 July 2017 (UTC)[reply]

Another one of Anna's jugheaded ideas

Is anyhow familiar with the Kim Jong-un non-free image thing?

The community has now wasted hundreds of hours on reads and keystrokes over this matter.

Do you remember Wikipedia:Donated artwork?

Bottom line:

It listed BLP articles with huge hits needing any image. Artists make a one-hour drawing to get credit and half a million views a year. No money offered. No takers. Artists are poor. Asked at WMF and it was all hoops to jump through to loosen the purse strings.

What if Wikipedia:Donated artwork tendered? An off-wiki crowd-funding/donation thing? Would-be WMF donators sick of the pile of unused loot instead put their cash into something direct. Thoughts? Anna Frodesiak (talk) 00:28, 26 July 2017 (UTC)[reply]

Alright. No interest. I get it. I had to try. Please hat. Thanks for the read. Anna Frodesiak (talk) 00:15, 27 July 2017 (UTC)[reply]

Well give it a couple more days. It's a little off the wall for rapid responses :) --Elmidae (talk · contribs) 00:56, 27 July 2017 (UTC)[reply]

I don't really see how this is a solution to the Kim Jong-un problem. I've never follow the situation but a quick loook suggest to me that AFAICT, there are already multiple free content sketches of him. These have been used on and off in the article since early 2015 or earlier. The primary object to them is not due to questions over whether they are free content, nor to do with the quality, but that some people feel there should be a photo. The only person I've seen objecting to the quality is you. Possibly 1 or 2 other people although I'm not sure about these since their comments were ambigious, I think they just don't want the sketches because they don't think any sketches will do. In others perhaps this will resolve your objection, but it's just going to be a waste of the artist's time since there's a fair chance the problem will still exist and I'm not even sure that this sketch will be more popular than any of the other current ones. Nil Einne (talk) 08:00, 27 July 2017 (UTC)[reply]

I've now looked more carefully at the discussion. I found one other person who seemed to be objecting specifically to the sketch nearly 3 years ago since they said it was cartoonish although this was when there were fewer options and I'm not convinced they would have been happy with artwork. As said, there are other people who made less specific comments like "ugly" but I'm even less convinced these people will be happy with any artwork. One comment which caught my eye was "A talented portrait artist working in a hyper-realistic style could create an original image, or painting, of Kim Jong-un based on study of many photos rather than copying one photo. If that artist donated the image under an acceptable Creative Commons license, then this problem would be solved." I think this comment may have partly inspired this proposal but I'm not convinced it's true. Personally I share Jack Upland's view in that same discussion, and this is IMO supported by previous discussions for this very issue (as well as previous ones I remember seeing). Even hyperrealistic professional art is likely to have problems since people simply object to the idea of art instead of a photo since they either don't understand or don't share our free content goals. (Although it's true very high quality photorealistic art can be very difficult to distinguish from an actual photo so many casual observers won't know. But then again I get the feeling this is going to increase the objections of those who feel there's something wrong with using art instead of a photo.) In any case, even if something like that would be acceptable unlike the other sketches so far, while I'm not an artist I'm pretty sure it's not something that is going to take only an hour but many hours. Nil Einne (talk) 09:12, 27 July 2017 (UTC)[reply]

To put things a different way, most of the discussions have actually been about why there's no photo and whether or not we should allow an NFCC photo and whether our policy on that is a good thing etc or not. While it's often hard to tell due to the lack of clear commentary, when there has been discussion of the sketch it's often been about which one is preferable. When people have objected to all the sketches and suggested no photo is better rather than simply saying we should use a photo, I've seen almost no one saying the current options are too poor but a better sketch would be okay. As said, their reasons aren't always clear but with a few of them I get the feeling they either don't think any sketch will ever be acceptable to the community and so prefer to keep it out to reduce controversy, or they personally feel that all sketches are too poor a subtitute for photos. It's possible some or many of them haven't considered truly photorealistic art, but as said earlier it's also easily possible the objections will remain with that, or even increase. It's true some articles on living people have survived for years with sketches of various kinds, but these have tended to be less high profile and their sketches weren't necessarily any better than what we already have for Kim Jong-un. Note that none of this means that the idea won't be helpful elsewhere, simply that the cited example doesn't seem to help the case. There's a fair good chance even if someone did pay for art 3 years ago, we would have spent the same or more time on discussion at Kim Jong-un. And actually, it does highlight one of the risks of paying for art, there's no guarantee consensus is going to be in favour of including it no matter how good it is. Nil Einne (talk) 09:12, 27 July 2017 (UTC)[reply]

Maybe people don't understand the proposal? I'm a native speaker of American English and have never in my life seen the verb tender as in "offer payment". I wouldn't donate to that cause, but maybe some people would. Ntsimp (talk) 14:31, 29 July 2017 (UTC)[reply]

Then it must be a AmE/BrE thing. "to tender" is perfectly standard English, as in the phrase "legal tender". I must confess I haven't regularly used buses for a couple of decades, but on one man operated buses you often saw the notice "please tender exact fare". Martin of Sheffield (talk) 15:11, 29 July 2017 (UTC)[reply]

The fixed phrases "legal tender" and "to tender one's resignation" do survive in American English, but note that Anna used "tender" intransitively. Is that standard in British English? Ntsimp (talk) 02:24, 30 July 2017 (UTC)[reply]

The transitive use you queried 'the verb tender as in "offer payment"' is fairly common, not just in fixed phrases (see my comment about bus notices). Anna's precise use is more akin the the tender process whereby contracts are offered and accepted. "Acme Engineering tendered to manufacture 1,000 widgets" for example. I would use the word when I was giving payment to a request, as in "tender correct fare" or "tender cash" (as against debit card), rather than initiating the transaction when "proffer" might be more appropriate. All IMHO of course, I'm only an ageing observant amateur not an academician. Martin of Sheffield (talk) 09:58, 30 July 2017 (UTC)[reply]

Hi Nil Einne. I think you're probably right. It just pains me to see all those keystrokes landing there instead of in the mainspace. I do say "probably" because the current artworks are really lousy. If a great pic of the man were to be added, I do think it would at least reduce the amount of wasted resources. And a decent drawing wouldn't take long. We're not talking oil paint here, just a good sketch by someone who could really capture him. Anna Frodesiak (talk) 00:59, 31 July 2017 (UTC)[reply]

Hi Ntsimp and Martin of Sheffield. Tender, yes, okay, I think I got the wrong word. I've seen ads in newspapers headed "Tender" with the text reading '"...100 park benches required...place your lowest bid...", that sort of thing. The idea was to put a price on an image to be created, or maybe to see what people would bid, or maybe for people to link to their creation, and financial donators would offer money for the one they prefer. You know, 'tender'. Something like that. Anna Frodesiak (talk) 00:59, 31 July 2017 (UTC)[reply]

Request for tender, that sort of thing. Anna Frodesiak (talk) 01:00, 31 July 2017 (UTC)[reply]

Enlightening images and text - tied together

Moved from Wikipedia:Village pump (technical) § Enlightening images and text - tied together

– ―Mandruss ☎ 08:25, 29 July 2017 (UTC)[reply]

First I must say that also from a technical perspective, is Wikipedia a really great invention. But we have text and we have images (and also moving images). But if it's possible from a technical aspect, would I propose a third such option. In which one or a few images can be "tied" to a certain text, for explaining reasons. I imagine some kind of "block of image(s) and text", which could be included inside articles, at any location in the text. Just as an example (nothing else). A four stroke piston engine. One picture of each of the four strokes (and how the valves opens and closes), with a specific text tied to each of the images. (I know this nowadays can be done even better with a moving illustration, but this isn't about the example, but in general) This "block" of one or more images with one joint text or one text per image, should still be able to edit and can be moved. But if the common text either before or after "the block" becomes shorter or longer, will the "block" still remains as one unit. Another point here, is to be able to have several images on the same line (and from left to right, normally). The "block" becomes a "stand alone" object. Sometimes doesn't illustrative images quite fit horizontally with the explaining text. And common image texts are to be brief. The so called "block" can be edited by any contributor, but when other texts (located above or below a "block") changes size, should the "blocks" be intact. I hope this idea is understood. Usage can be for any type of processes (within history, math, physics etc) or for exemplification, that is best presented all across the screen horizontally. They also ought to be removed if their purposes are not obvious, and are not intended for, for instance picture galleries etc. An overflow of such "blocks" isn't what I hope for. There must be either "a line of explanation" or (in cases with a single image) an explanation which really needs to be tied to a specific image. Boeing720 (talk) 00:50, 29 July 2017 (UTC)[reply]

So templates ? —TheDJ (talk • contribs) 07:52, 29 July 2017 (UTC)[reply]

First I thought no. But the infoboxes have similarities. Main issue is to glue a certain text to one or a few images, for illustrated explanations purposes. I presume (what I like to call) "blocks" have a start point and an end point in the syntax. (And include at least one image) If the text above this "block" expands, will the "block" including its images follow, etc. (Unless a contributor finds it proper to move it/edit it/delete it) The "blocks" should be able to edit just as everything else. But as long as changes are not about the "block" itself, will it remain as it is. Today may explaining images stand to the right of a text which deals with other matters. Although the image once corresponded to the correct text.

While infoboxes are meant to be seen and noticed, is my idea sooner that such "blocks" shouldn't look differently from ordinary texts to our readers. The technical way to achieve this possibility, am I not really familiar with. Boeing720 (talk) 20:37, 29 July 2017 (UTC)[reply]

By the way, I noticed your four points about discussions. I liked them. And I "stole" them. I hope that was OK, elsewise will I remove them again, naturally. Boeing720 (talk) 20:51, 29 July 2017 (UTC)[reply]

Yes, it appears you're proposing templates, which are used for many things besides infoboxes. Or possibly just images with captions. Anomie ⚔ 12:05, 30 July 2017 (UTC)[reply]

Had a look. What I'm after is pretty much inline with that, but what I really would like to have (if it's possible from a technical perspective), is an option to "glue" text and pictures together, but in such a way, that our readers just see it as common text inside the text. Only when editing should it be obvious that there is a syntax which "glues" image(s) on to a certain part of the text. Have a look at for instance Complex numbers. Never mind the math, but this article is rather well illustrated. And perhaps the images fit the text on the left, but there is no guarantee for it to always remain like that. If someone adds some 15 lines below a headline with an image to the right, that image will no longer fit the corresponding text to the left. That's the main reason to why I've brought this issue up. But also the image texts are rather long. What I'm after, is way to locate such images into the ordinary text, and if the article grows, should this part of the text with its corresponding images not be affected. (Not until someone decides to edit inside this part). I hope I'm making myself clear enough. Boeing720 (talk) 22:25, 30 July 2017 (UTC)[reply]

You do seem to be having trouble making yourself clear here. The only way the images in Complex numbers will be separated from the adjacent prose is either (1) if someone inserts more prose between the image markup and the intended prose, or (2) someone puts so many images with so little prose that the images get pushed down the page. There's no real technical solution for #1 since it's an editorial problem, and #2 is just the way HTML works and there's little we can do about it without adding big blobs of whitespace in such a situation instead. Anomie ⚔ 22:45, 31 July 2017 (UTC)[reply]

Change FAC to A-class style review

Background: Over the past few years, I've been working on the Leo Frank article, receiving a peer review at the end of 2014, nominating it for GA in 2015 (the first time was quickfailed due to an edit war from a sock, and the second passed after an extensive review from SilkTork). I nominated it in FAC in the fall of 2016, but came up just short after another arduous review.

Recently I've considered reviewing the first FAC, making any appropriate changes, and renominating at FAC. But I realized after considering the process so far that I don't have confidence that a second FAC would actually improve the article. This is because of the time invested by one reviewer at GAN versus the comparatively ad-hoc nature of FAC.

To illustrate this, have a look at the article I worked on and mentioned above, Leo Frank. To a casual observer who's not familiar with him or his murder trial, it may seem long-winded for a Wikipedia article. Indeed, the GAN process significantly elongated the article, going from 78K characters in August 2015, peaking at over 125K in the review process, and back down to 108K in October after a considerable amount of trimming at the end by the reviewer. The article still grew by 30K, and this was a point of contention in the FAC. Brian Boulton recalled from the 2014 peer review that the article had grown considerably between that review and the FAC, and he and Sarastro encouraged me to further remove some of the content. I was reluctant to do so, and this became a point of contention between myself and Sarastro during the review. He felt that I was ignoring his comments, when I simply didn't want the work I had done in GAN to largely be reverted.

I realized that the reason for this difference in opinion for article length between reviewers who are all highly experienced and skilled, namely SilkTork versus Brian and Sarastro, is a matter of the review format. GAN allows one reviewer to dive deep into a subject, as SilkTork did with this article. He learned about Frank, his trial, and the source material extensively as to give a high-quality and thorough review. FAC reviewers, on the other hand, have a higher number of articles they are concerned with at any one time and generally just go through the bullet points, doing basic copyedits, source reviews, image licensing checks, etc.

I feel like the GAN process was a better system despite the lower standards. I also think there should be something that goes beyond GAN that includes a more stringent criteria for passing and solicits feedback from more than one person. A-class does this, using three reviewers with expertise in the area (military history and tropical cyclones being the two prominent examples currently) and a higher bar than GA while slightly short of FA.

My proposal is to make the FAC process more like A-class, while retaining the FA criteria and gold star. We could keep A-class for the WikiProjects that choose to use it, but have a pool of FA reviewers and have any three (or whichever number we decide by consensus here) review a particular article. The reviewers could decide which articles they want to review based on their interests and expertise, while not needing to be a part of any WikiProject. That way, you don't have the ad-hoc issues in FAC of reviewers coming and going to a particular article, and the A-class problem of not enough skilled reviewers in all but two projects to make it happen, in addition to the other issues outlined above. This will make FACs more thorough and completed in a shorter time frame. I'm open to any feedback below and will condense this into an official Pump proposal if there is enough interest. Tonystewart14 (talk) 19:08, 29 July 2017 (UTC)[reply]

The problem with A-class reviews – and the reason that MilHist is pretty much the only project which still does them – is that in most areas, there simply aren't enough specialists active on wikipedia who are willing to put in the time to review articles in-depth in the way that ACR and FAC require. (And, notwithstanding your experience with the GA process, most GA reviews are not by people who are experts in, or put the time in to learn about, the article they are reviewing.) I also am unconvinced that the suggestion would actually solve the problem you had – an editor who thinks an article is too short and one who thinks it is too long just fundamentally disagree.

I have added a notification of this discussion at WT:FA. Caeciliusinhorto (talk) 08:45, 1 August 2017 (UTC)[reply]

I don't think this proposal would solve the problem you want it to solve, which is to get a dedicated reviewer who thoroughly engages with the article. This does happen at FAC, and in my experience it happens a bit more often at FAC than at GAN. A good reviewer is worth a lot in any process, and WP:PR, for example, can yield better feedback than either GAN or FAC if you're lucky. The changes you're proposing don't make it more likely that an article would receive the in-depth review that you (and we all) would like, do they? I also think it's an advantage at FAC that you can have many more than three people show up; sometimes that's necessary for a thorough review. Re the Leo Frank article: I think good editors can legitimately disagree about the level of detail needed in an article, but a discussion between those editors is likely to be of value to the article. If that discussion is not resolved, I don't think it would be appropriate to award an article A class or FA class; consensus needs to be reached first. Mike Christie (talk - contribs - library) 10:13, 1 August 2017 (UTC)[reply]

Or we don't think it's important. WPMED has the experts, but we don't care enough about the ratings to bother with A-class. Or FAC or GA or PR, for that matter: sometimes people nominate articles, and we usually pitch in on the reviews, but going through an official process does not seem to be a priority for the group. WhatamIdoing (talk) 20:31, 2 August 2017 (UTC)[reply]

Sister project for references

Since the citing system (references) has become quite complex over the years and can vary a lot throughout the Wikipedias and other sister projects, I started to think about special project that would deal just with references.

The main idea is to collect data of publications (similar to library identifiers) that are used as sources in articles. Every "unit" would be given it's own number (U9748 for example or something else, similar to Wikidata's "Q") that would be cited in the code of the article (Some statement.<ref>U9748, pages 123–127.</ref>). Every Wikipedia should then create it's own module with definition of style of citing.

The project could probably be offering automated categorisation of publications by authors, years, languages ... Every unit would have the list with global usage, similar to Commons. One of the advantages of this feature is also a simple tracking of fake sources. User identifying one can easily see wherever the unreliable publication is cited. Besides, codes of articles would become much lighter, since all citing templates are minimized. --Janezdrilc (talk) 19:48, 30 July 2017 (UTC)[reply]

@Janezdrilc: You will be interested in meta:WikiCite. --Izno (talk) 19:56, 30 July 2017 (UTC)[reply]

Thank you for link. I didn't know it before. But, still I wonder if it's better to have WikiCite within Wikidata and not separate - as independent project with Wikidata engine. --Janezdrilc (talk) 17:04, 31 July 2017 (UTC)[reply]

@Janezdrilc: Which, regardless, is not really something regarding en.WP. Feel free to introduce yourself there and see if anyone there agrees or disagrees. It's probable that they've talked about that already--which, for the most part, we won't have talked about here. --Izno (talk) 20:25, 31 July 2017 (UTC)[reply]

We should be able to thank anonymous users

I really like the Thank user for edit feature of wikipedia, and I think we should also be allowed to thank IP editors for good edits. I think this would help spread wikilove and positivity! --HighFlyingFish (talk) 19:52, 31 July 2017 (UTC)[reply]

IPs are almost always dynamic. Since the person using the IP at the time you thanked it would almost certainly not be the person who made the edit for which you're doing the thanking, the only effect this would achieve would be to confuse the new user as to why they were being thanked for something they'd never done. We discourage people leaving messages on IP talkpages for the same reason. ‑ Iridescent 20:08, 31 July 2017 (UTC)[reply]

I wasn't aware of that about IP talk pages. So there is no good way to communicate with an IP except to drop a message in some other talk space (article talk, noticeboard, etc.) and hope they happen to see it. As the collaboration process relies heavily on communication, add one to the long list of reasons not to allow unregistered editing. ―Mandruss ☎ 05:45, 1 August 2017 (UTC)[reply]

Not really. If you know that the IP isn't dynamic and that only one person uses it—a situation which is fairly rare, except in some rural areas and a few fiber-optic providers—then you can leave a message on the talkpage as you would any other user. In most cases—particularly in areas with heavy internet saturation like the UK, New York City and California, which unfortunately are where Wikipedia's editors disproportionately are, the number of IPv4 addresses has long since run out, and your ISP will just allocate you a temporary one on the fly each time you log on—thus, the person using a given IP address at present won't necessarily be the same person who was using it ten minutes ago. (You can test it for yourself—go to www.whatismyip.com to find out your current IP, leave a message on that IP's talkpage, switch your router off and on again, visit Wikipedia logged-out, and see if you get a "You have new messages" notification in relation to the message you left. Or, just go to www.whatismyip.com, switch your router off and on, and go to www.whatismyip.com again and chances are it will have changed. And that's just for wired home/workplace connections—if you're using a cellphone or a mobile broadband connection, your IP address will quite literally change every time you go under a low bridge.) Even in those situations like academic institutions and big corporations with dedicated lines where the IP address don't change, you still have no way of knowing who you're addressing when you leave a message for an IP since the message will in that case be seen by the next person who happens to use that computer to read a Wikipedia page, whoever that person happens to be. ‑ Iridescent 16:42, 2 August 2017 (UTC)[reply]

Oh I'm all too familiar with IP addresses that change frequently. I recently spent way too much time - over a span of days - trying to communicate with one such user about their ongoing activity. By the time I saw an edit fitting their pattern, in their IP range, and dropped a message on their talk page, it was not their talk page anymore. Again, again, and again. I finally gave up in frustration, unsure whether they ever received one of my many messages on their many talk pages. And that is my point. The only thing that was news to me was We discourage people leaving messages on IP talkpages for the same reason. - that the community consciously acknowledges that (most) IPs cannot be full partners in the collaborative process. As I said, that's only one of the many serious problems with unregistered editing, but there are still plenty of experienced editors who defend WMF's unyielding position on unregistered editing. In my opinion that is one of the few most pressing problems with the project today, and it's getting no attention aside from occasional scattered, off-topic comments like mine above. ―Mandruss ☎ 20:52, 2 August 2017 (UTC)[reply]

While I don't personally agree, I can at least understand the WMF line—most Wikipedia editors began as someone who just wanted to correct a typo or a minor error but end up staying, and those people who aren't intending to stay probably won't bother setting up an account. I think it creates more trouble than it's worth, but it's certainly a legitimate position. ‑ Iridescent 21:04, 2 August 2017 (UTC)[reply]

I too understand the WMF line. At some point a position is so indefensible that it can't be called "legitimate". The cost/benefit fail here could not be much clearer. IP editing was clearly needed to get the encyclopedia to a certain point of development, but we're years past that point. We could do more to clarify that registration requires absolutely no personal identifying information, not even a throwaway Yahoo email address. And we could stop calling unregistered editors "anonymous" while they are exposing their IP addresses to the world. ―Mandruss ☎ 21:18, 2 August 2017 (UTC)[reply]

You can thank such editors, and publicly too. Instead of the thank feature or the user talk space try the talkspace of the relevant page, hopefully that's usually going to be an article talkpage. A section saying thanks to whoever fixed my typos/added the flora section/ or just thanks for this (diff). If they are interested in the article they edited they may well see your note. Ϣere SpielChequers 21:02, 2 August 2017 (UTC)[reply]

Or there is no good way to communicate with an IP except to drop a message in some other talk space (article talk, noticeboard, etc.) and hope they happen to see it, in other words? ‑ Iridescent 21:04, 2 August 2017 (UTC)[reply]

3-segment wikilinks

For use in making wikilinks, particularly in languages with case-endings and similar, try allowing 3-segment wikilinks:-
1. [[X]] displays and links to X , as now.
  - [[X]]Y links to X and displays XY , as now.
2. [[X|Y]] links to X and displays Y , as now.
  - [[X|Y]]Z links to X and displays YZ , as now.
3. And idea: let [[X|Y|Z]] link to XY and display XZ . For example, "he [[procrastinat|ion|ed]]" would be more compact than "he [[procrastination|procrastinated]]".
  - And if so, [[X|Y|]] would link to XY and display X .
- Anthony Appleyard (talk) 09:24, 1 August 2017 (UTC)[reply]

That could be interesting. --167.58.26.73 (talk) 19:55, 3 August 2017 (UTC)[reply]

Text searching

Currently the searching templates {{look from|xxxxx}} and {{intitle|xxxxx}} seem to look only for letters and numbers and ignore punctuation etc. This can be a nuisance, because, currently there is a policy to remove the comma when a page name ends in ", Sr." or ", Jr.", and to help to do this work, there is no easy way to search for these two character sequences, but someone must look at every page name by eye, and requests to remove these "senior and junior commas" come in endlessly in dribs and drabs instead of someone being able to quickly call search for ", Sr." and ", Jr." and get the job complete and done and over. Please provide an option for Template:look from and Template:intitle to search for all characters, not only for letters and numbers. Anthony Appleyard (talk) 05:34, 2 August 2017 (UTC)[reply]

The latter is phab:T156510. I'm not sure how you mean to use the former for your use case, but it doesn't seem reasonable off the cuff. --Izno (talk) 12:31, 2 August 2017 (UTC)[reply]

@Izno: As regards Template:intitle , what is the likely progress with https://phabricator.wikimedia.org/T156510 ? Ability to search for characters other than letters and numbers would be useful. Anthony Appleyard (talk) 05:17, 3 August 2017 (UTC)[reply]

I do not know. That is something to ask on the phabricator task. :D --Izno (talk) 13:37, 3 August 2017 (UTC)[reply]

There is a beta search tool on wmflabs that uses regexp to search article titles. Unfortunately, you're on your own for constructing efficient regexp searches. Also, the tool can be very slow and can return a LOT of results if you put in a too generic a regexp. For example, this search for ", Sr." took quite a while to run, but it looks like what you describe. (and I'll refrain from commenting on whether it is a worthwhile use of time to "fix" such trivial formatting differences). older ≠ wiser 14:11, 3 August 2017 (UTC)[reply]

Having regular expressions in page search seems a bit excessive. With a one-character sign ('.') and an any string sign ('*'), that should be enough. --167.58.26.73 (talk) 19:58, 3 August 2017 (UTC)[reply]