Many infoboxes for places (schools, building, etc.) support geo coordinates in the infobox. These are usually displayed in the infobox and at the article title. There are articles with such infoboxes that have coordinates at the title only. I think this is rarely, if ever, a consious decision to display the coordinates only at the title. It probably happens mostly when an infobox was added to an article that already had coords at the title and they just weren't moved to the infobox, or "missing coords" are added to an article by an editor who did not realize/notice they could/should go into the infobox. I propose a bot move such coordinates into the infobox. JJMC89 has volunteered to do this. This would not affect any coords that are displayed "inline" anywhere else in an article. This should be straightforward, except in some cases with both an "inline" coord in the infobox and a "title" coord outside the infobox that aren't identical. Those would probably have to be handled manually. MB 02:19, 31 January 2020 (UTC)

A bot would also need to confirm that the omission of title was not intentional. I don't have any precise information at the moment but there are articles where multiple things are discussed, and some of the things have {{coord}}. MediaWiki makes certain assumptions about coordinates that have title. There should only be zero or one per page, and if one, it affects "what's near here" that can be used. Johnuniq (talk) 02:49, 31 January 2020 (UTC)

I recently stumbled upon this seemingly wonderful invention. I see great potentials in it. I see shortcomings in it, also - but not ones like the complaints that pollute its article's Talk Page. There, the complaints all seem to center around a kind of "English Language Bias". This is not helped by the fact that use-of-language seems to be Simple's only expressed selling point;

I believe it's a disservice to limit Simple's described audience to that. (It also opens up Simple to unwarranted criticism - particularly from those with a kind of English-only prejudice.)

I would argue that additional stated benefits should include concision and convenience. Its articles, nor writing guidelines don't need to change - they already appeal to these benefits. My first suggestion is that these benefits be put right on its front page, as well as the main Wikipedia article.

I'm highly technical and language-fluent, yet there are some articles in English's main Wikipedia that are so laborious and/or highly technical that it's a difficult read. This is not a complaint. This is, of course, what one wants from an encyclopedia! But sometimes I don't want to attain a PhD on a subject, but instead just a general perusal.

Doubtless Simple would benefit from more articles, ergo: more authors. For this, I don't think the challenge is a lack of people willing to write for it, but rather a lack of people aware of it. A decade and a half after its creation, and I only just stumbled upon it - and not from Wikipedia. I inadvertently found an article from a search engine. I should have seen a reference to it long ago on Wikipedia, itself.

This leads to my next suggestion, and this is an admittedly bold one; a "Simple" tab right next to the |Article|Talk| tabs. Right on top. Obviously, only for main-English articles for which there is a corresponding Simple article.

My fervent belief is this would create a massive wave a new awareness of Simple's existence, and a concomitant flood of new authors. Only then could Simple reach its potentials, therefore its true value.

     —  BoringJim (talk) 21:14, 25 December 2019 (UTC)

This has been discussed before (a quick search shows Wikipedia:Village_pump_(proposals)/Archive_66#Examples_TAB_beside_Article_tab_and_discussion_tab as an example). It could be possible to add at least a local script to include "Read on Simple English Wikipedia" or something if the interwiki to simple: exists. I doubt this would get integrated to mediawiki core. Problem at least for non-logged in users, we really wouldn't want to run a script an reformat the page every single time a page was read - and this wouldn't be effective for mobile readers. This could possibly be done as a gadget for logged in editors. — xaosflux ^Talk 23:04, 25 December 2019 (UTC)

What about a less complicated (dare I say, simpler) solution? We have {{commonscat}} and similar templates that can be added to let users know we have more content elsewhere. I don't know that we have that for simple, {{Simple}} is a redlink. Beeblebrox (talk) 21:43, 4 January 2020 (UTC)

You can also add the following code to your Special:MyPage/common.js. This code could also be added to MediaWiki:Common.js to apply to the entire wiki. The code will show a "Simple" tab as described by BoringJim if there is a language link to Simple English Wikipedia in the language list.

mw.loader.using('mediawiki.util').done(function(){
	var sl = $('#p-lang .interlanguage-link-targetlang="en-simple"').get(0);
	if (sl && mw.config.get('wgNamespaceNumber') === 0) {
		mw.util.addPortletLink('left-navigation', sl.href, 'Simple', 'ca-simple', 'View this page on the Simple English Wikipedia', 's');
	}
});

– BrandonXLF (talk) 03:44, 7 January 2020 (UTC)

That's interesting and I thnak you for sharing it. However, I think when it comes to the casual reader they don't really look much at those tabs, but they might notice a tag in the article itself. Beeblebrox (talk) 20:42, 13 January 2020 (UTC)

There's an existing script to do this: User:Operator873/SimpleWPTab.js. In regard to deploying something like that for everyone, I think it would be really helpful to help spread awareness that the project exists. Vermont (talk) 10:52, 16 January 2020 (UTC)

It'll be helpful because those learning english may not know simplewiki exists, and, in addition, editors who see it may want to edit the simplewiki. I support that being added to common.js. We also may want to consider mentioning it in school project pages and such. Computer Fizz (talk) 17:43, 16 January 2020 (UTC)

After looking it over some more, this should only be added to vector.js, not common.js, because it doesn't work with monobook, a seperate version can be made for that skin. Additionally, for the point about how casual readers may not look at the tab, how about the things that look like userboxes that mention things like the Wiktionary, Wikiquote, etc, when applicable, there can simply be one created to say "The Simple English Wikipedia has an article about title", or something similar. Does everyone agree with this? Computer Fizz (talk) 17:58, 16 January 2020 (UTC)

Computer Fizz, you also have to consider that the about script would make a request to the server for every person reading Wikipedia, seems like something that would be much better implemented in a Mediawiki extension of something. – BrandonXLF (talk) 04:40, 19 January 2020 (UTC)

@BrandonXLF: I think at first we should just use the top right box, then after that's been going for several months have another discussion about a Simple English tab. Perhaps there could be a bot to add and remove the boxes. The one problem though is that sometimes simple english can have different tiles than english (i.e. we say "movie" instead of "film") Computer Fizz (talk) 08:18, 19 January 2020 (UTC)

Computer Fizz, are you talking about the templates like {{Wiktionary}}? Those are added on a per-page basis and are generally placed near the bottom of the article from what I've seen. One could be created at the name {{Simple}} for the Simple English Wikipedia, that would be a good start since it doesn't involve any major changes.– BrandonXLF (talk) 08:28, 19 January 2020 (UTC)

@BrandonXLF: I think this would be better placed at the top, but i'm not entirely sure. I will see about drafting a simple template in my userspace tomorrow (37 after midnight for me right now), and if enough people like it I'll move it to template:simple. Computer Fizz (talk) 08:38, 19 January 2020 (UTC)

@BrandonXLF: Does this look good to you, and anyone else present in this discussion? Computer Fizz (talk) 21:37, 19 January 2020 (UTC)

Computer Fizz, I think it's good. I'd prefer if it said "on the Simple English Wikipedia" though. You'll also need one that uses {{sister-inline}}. – BrandonXLF (talk) 20:33, 20 January 2020 (UTC)

@BrandonXLF: Changed the text to "on the Simple English Wikipedia", but the inline template doesn't seem to support text for Wikipedias. Could you try making what you're thinking of in your own subpage? Computer Fizz (talk) 01:52, 21 January 2020 (UTC)

Computer Fizz, I made the two templates using Module:For nowiki at User:BrandonXLF/sandbox/3. Feel free to move them to your userspace. – BrandonXLF (talk) 02:16, 21 January 2020 (UTC)

@BrandonXLF: No at this point what we need to do is get the go ahead from a closing admin to move it into the template namespace and start adding it to pages. Computer Fizz (talk) 03:04, 21 January 2020 (UTC)

• @BoringJim: I appreciate your enthusiasm, initiative, and good will. There are many aspects of Wikipedia which people would enjoy if we made those tools and projects more accessible. Regarding Simple English Wikipedia, the situation is complicated, and in 20018 there was a proposal to close it. If things do not change in the near future I expect another proposal to close within a few years. Lots of people would have their own explanation for this, but briefly, the project has some problems which have brought it to the point where practically all Wikipedia outreach intentionally avoids directing new editors and readers to that project. Blue Rasberry (talk) 19:15, 16 January 2020 (UTC)

@Bluerasberry: True, simple english does have problems, but none of those problems will be solved by closing it. One such issue is that more people need to know about the wiki, which is being solved right now.Computer Fizz (talk) 21:22, 16 January 2020 (UTC)

It'd probably make sense to get people in this conversation who know how the Simple English Wikipedia works. User:Bsadowski1 and User:Djsasso are admins on both simplewiki and enwiki. User:BRPever is a sysop there and a global sysop, and User:Vermont is a sysop there and also at Meta. WhatamIdoing (talk) 19:00, 21 January 2020 (UTC)

Well I would love to see a tab here of some sort that contained simpler versions of articles (be that as part of this wiki or as a link to simple) as lots of this wiki's articles you need a degree to read and are of no use to younger readers or ESL people. But I don't know how likely it is to occur. It seems to me something along those lines comes up on this page every year or so with some support and some against. -DJSasso (talk) 19:03, 21 January 2020 (UTC)

I think having a tab to SEWP is a good idea and will help in solving some of those problems shown in 2018 (I think most of them). It is very difficult for us to do any offline events, Workshops, and outreach activities because unlike other medium-sized language wikis where most of the editors are from a specific region, editors from Simplewiki are scattered all over the world.--BRP ^ever 21:09, 21 January 2020 (UTC)

A few thoughts from me... I'm a SEWP sysop and member of the community for some time now. The aforementioned tab script and it's SEWP companion, (enWPTab.js, introduce pretty awesome functionality for me, personally. However, I do agree that hitting the API every single time a page is loaded by every user using this script would tax the system after a while. It would have to be reworked, ideally as an extension as BrandonXLF stated. Next, I'd like to take a moment to address the point brought by Bluerasberry regarding the future of SEWP. In the sited proposal's closing remarks, LangCom quite clearly stated:

Please note that in the future, consistent with the above, requests to close this project that are based on "no language code" or "not different enough from English" may be closed speedily. Requests to close this project that are based on "inactivity" or "vandalism" will probably also be closed speedily, unless things change pretty radically at Simple English Wikipedia.

As far as "merger" requests go, LangCom certainly likes the idea of Simple English content being more accessible from English Wikipedia. Whether that should be accomplished through merging the projects or merely better coordination between the two projects is a matter for the communities to decide, not LangCom. In light of that, we will not entertain a request to "close and merge" unless it is presented as the consensus of both communities, and unless at least the broad outline of a plan to accomplish that is in place.

As such, I feel SEWP is project that will remain active for quite sometime to come. With that in mind, I find the closer integration of the two projects, either from a "Simple English" tab or button or {{Simple}} tag to be ideal for both projects. Granted, I see the future need for the SEWP community to arrive at a consensus to redefine the appropriate MOS to more readily complement a user experience swapping from the technical and high-leveled language of enwiki to the more basic and readable version of SEWP. The articles contained on enwiki are very thorough, well written, and technically in-depth for the most part. However, this praise is also the very problem SEWP seeks to correct. SEWP works to have articles written in a manner that is easily read by the student, the adult English student, and those not interested in technical explanations.

I got a bit long-winded here, but in summary, I would support a trial integration of the two projects. I find the tab is the better implementation, but it would need to be reworked from the script. SEWP isn't going anywhere and the communities of both as well as readers would be better served by close integration, but not merging. Operator873^talkconnect 21:34, 21 January 2020 (UTC)

Personally I feel like the tab is a pretty big leap for a first step, but if people awnt that I'm okay with it. The one concern thoug his that sometimes english and simple can have different article titles, like movie vs. film, and we'd need some sort of way to tell it to point to a different title, perhaps with a magic word? Speaking of which, whether or not this tab will be an extension or javascript is another concern,an extension could be much more optimised but javascript would be easier to get added. Computer Fizz (talk) 01:23, 24 January 2020 (UTC)

Computer Fizz, we could query Wikidata for the page name on the Simple English Wikipedia since they should both be linked to the same Wikidata item.– BrandonXLF (talk) 22:12, 25 January 2020 (UTC)

@BrandonXLF: Not all pages are in Wikidata, but it does seem like a good idea to include for those who are, since it's both guaranteed to work if it is, and because it gives more of an incentive to keep wikidata up to date. Although this would most likely be an extension and not Javascript, are you any good at making extensions? Computer Fizz (talk) 22:15, 25 January 2020 (UTC)

Computer Fizz, they should all be on Wikidata, but if they aren't, we can fallback to seeing of the page of the same name exist on the SEWP. I would not say I'm good at making extensions, but I could try to make something. – BrandonXLF (talk) 22:19, 25 January 2020 (UTC)

@BrandonXLF: Sounds like a fantastic idea. Not sure how ENwiki goes about getting consensus for this stuff though, so I can't be of much help there but I can put in my strong support and post on phab. Computer Fizz (talk) 22:21, 25 January 2020 (UTC)

I have worked on the Simple English Wikipedia from 2018 until 2019, when I was blocked there two times. Anyway, I have proposed the following solution to the aforementioned problem. Firstly, there would be a tab called "Simple" between the "Article" and "Talk" tabs, no matter what skin is used. The "Simple" tab would use a different namespace called "Simple:" (e.g. if the article title is "SpaceX", clicking at the "Simple" tab would direct you to "Simple:SpaceX"). If you have concerns that the readers would not notice the "Simple" tab, do not worry. Many IP users encounter the talk page because they simply do not know what the "Talk" means in Wikipedia. --Soumyabrata (talk • subpages) 06:55, 23 January 2020 (UTC)

The Basque Wikipedia does something similar, with a kid-focused namespace. See w:eu:Frantzia (look for the button at the top with the colorful font) as an example. Most articles don't have an article in their kids' version. WhatamIdoing (talk) 17:30, 23 January 2020 (UTC)

• I missed this discussion apparently, but trying to integrate Simple more closely with English would be a Bad Idea. SEWP should burn to the ground; if there wasn't enough support to close it, then at least we don't have to actively encourage its development from here. I, for one, would strongly oppose any effort to make the existence of SEWP even remotely more visible. –Deacon Vorbis (carbon • videos) 18:01, 23 January 2020 (UTC)
  • I would put it a little more gently. I think simple.wiki suffers from an unclear, self-contradictory, or simply impossible rationale. I haven't kept up, but years ago at least, the theory seemed to be that simple was just another language, and that the content would not be dumbed down, it would just be written using a restricted vocabulary.
    But of course that just isn't possible, not unless you want to increase the length of articles to the point that they become effectively unreadable for other reasons. It sounds good to say that you don't really understand something unless you can explain it to your grandmother, but unless you have an unusual grandmother, it just isn't true.
    When I've looked at simple.wiki articles on technical topics, they are either stubs with essentially no content, or they aren't written in Simple English.
    There might be an argument for a simplified Wikipedia (simplified in content, not just language). But we shouldn't pretend it's just a language. It would be a separate encyclopedia and there's no way it could support a one-for-one mapping of articles; some topics just could not be covered. --Trovatore (talk) 18:20, 23 January 2020 (UTC)
    • Think of it. I'm a vandal/blocked/pov/fringe/racist editor. I can't edit the main article or don't have consensus or whatever, but hey, there's a tab to an article on another Wikiproject I can use. And the Admins/editors on en.wiki won't have power there/be bothered to look at it, etc. Doug Weller talk

That's a really good argument against this idea. Unless there is some way to mitigate that problem, I do not believe a tab of any sort can be implemented successfully. Vermont (talk) 16:01, 24 January 2020 (UTC)

From my understanding, the project isn't acting as though it's an entirely different language. Take a look at simple:WP:HOW. The intention is to create articles that describe things in a way much more easily understandable. And yes, many of the more technical topics do not have great articles about them, but it's a constant work in progress, as are all projects. Vermont (talk) 16:01, 24 January 2020 (UTC)

Entirely different is not the issue. What I'm saying is, I'm skeptical that "language" is the right model at all. The page you pointed me to says the language is simple, but the ideas don't have to be. What I'm saying is, that doesn't work. Complex ideas are frequently impossible to express clearly and efficiently in simple language. That's my objection to the whole simple.wiki concept. --Trovatore (talk) 19:25, 24 January 2020 (UTC)

Opposed Simply English has serious sourcing problems that would overwhelm our editors and lower the value of our information.--Moxy 🍁 23:44, 24 January 2020 (UTC)

• Comment — In desktop mode, I can see the "Simple English" tab between "Article" and "Talk" tab, which redirects you to Simple English article.

Edit: You will not see this, because the tab is disappeared. --Soumyabrata (talk • subpages) 07:04, 1 February 2020 (UTC)

Most non-free file tags have a "To the uploader: " instruction box that is pretty much identical for all of them. Because of this, a while ago I did write Template:Uploader information to hold the text of this instruction box for present and future non-free file tags, as using a metatemplate makes it easier to write new non-free file tags and to do mass changes on old ones. I did apply it to some templates but as someone observed on this talk page I should probably ask in a single place if this is a good move before continuing. Jo-Jo Eumerus (talk) 14:08, 29 January 2020 (UTC)

Seems like a sensible use of a meta template to me. Significantly reduces redundant code, standardize the message so it can more easily be improved and in general makes maintenance easier. ‑‑Trialpears (talk) 15:28, 29 January 2020 (UTC)

Sure. If as you say it's the same text that is repeated, no reason why it can't use a template. --Gonnym (talk) 15:39, 29 January 2020 (UTC)

Gonnym There is a bit of variation between each template as they often suggest certain rationale tags or the like; the metatemplate has a |additional notes= parameter so that one can add extra text if it's necessary. Jo-Jo Eumerus (talk) 15:44, 29 January 2020 (UTC)

No objection. Jo-Jo, I have combined all your edit requests into one. I hope this okay — Martin (MSGJ · talk) 00:00, 1 February 2020 (UTC)

I don't think that's a problem. Anyone who had the templates on their watchlist will have seen the edit through which the edit requests were originally added. Jo-Jo Eumerus (talk) 08:52, 1 February 2020 (UTC)

Is this the right place to say it? New3400 (talk) 23:31, 2 February 2020 (UTC)

@New3400: I suggest starting a discussion at Wikipedia_talk:WikiProject_National_Football_League RudolfRed (talk) 05:27, 3 February 2020 (UTC)

My name is Josh Nicholson and I am the co-founder and CEO of scite. scite is a new platform that allows anyone to see how a scientific paper has been cited, not just how many times and, specifically, if it has been supported or contradicted. We do this by analyzing millions of scientific articles, extracting the citation context, and then applying our deep learning model to identify citation statements as supporting, contradicting, or mentioning. We recently announced an integration to display our smart citations with BioTechniques (https://www.future-science-group.com/fsg-and-scite-partnership) and Europe PMC (http://blog.europepmc.org/2020/01/europe-pmc-integrates-smart-citations.html) and have signed indexing agreements with a handful of major publishers to analyze their full-text (BMJ, Karger, Future Science Group, Rockefeller University Press, and Wiley).

I believe that displaying scite's "smart citations" next to scholarly references in Wikipedia would make Wikipedia even more of an amazing resource for humanity and wanted to see if this is possible. I know I am not the first third-party company trying to integrate with Wikipedia but I think because citations are so central to Wikipedia and what we're doing at scite is really trying to evolve them there may be a way (https://www.wiley.com/network/latest-content/all-citations-aren-t-created-equal). You can see what such an integration could look like on references here. From a technical perspective, I think it should be quite simple (we offer integration via API or a small HTML snippet).

I'd love to hear your thoughts on this and would be happy to answer any questions.

Thanks so much, Josh

PS you can see an example of my own research here which should give you an idea of how scite works on an article you've never seen: https://scite.ai/reports/chromosome-mis-segregation-and-cytokinesis-failure-nM3vVr — Preceding unsigned comment added by Jdiogenes86 (talk • contribs) 14:21, 24 January 2020 (UTC)

@Jdiogenes86: Thanks a lot for your interest in improving Wikipedia. This indeed looks quite neat.

At this stage though, I don't think a consensus can be established to allow you to directly edit the WP articles to add scite data to them. What you can do is one or both of:

• Create a bot that scours through papers cited on WP, looks up the scite info for them, and identify the ones that have too high contradiction rates. This data can be posted to a subpage (such as User:Jdiogenes86/contradicted papers) so that editors can manually review them.
• Create a user script to allow editors to see these scite links next to the citations on articles.

For building a bot, there is the Action API and the newer REST API. The former is associated with good client-side libraries like mwclient and pywikibot.

You may also want to contact the folks at WT:WikiProject Academic Journals and WT:WikiProject Science. SD0001 (talk) 10:39, 27 January 2020 (UTC)

Let me start by saying that I think the concept of smart citations sounds very intriguing and is worth exploring. That said, implementing it would not be a small decision, but a big deal decision that requires the involvement in the signoff of the community. While I think @SD0001: identifies some good tools, as well as some places for further discussion, I suggest the cart is before the horse — we are not quite ready to fine tune the implementation, we need a lot more information before going ahead. I anticipate a lot of pushback about the notion of wholesale incorporation of material from an outside vendor without significant review.

One specific suggestion is that it might be useful to utilize a test wiki such as this (or perhaps there is a more suitable option). I'd like to see this in action on some real articles which could be copied over to the test wiki, but I want to see it in a testing environment before I see it in Wikipedia itself.--S Philbrick(Talk) 18:27, 27 January 2020 (UTC)

This is also discussed on the WikiCite mailing list (on Google Groups). Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 21:13, 27 January 2020 (UTC)

I like the general concept but it would be a very bad idea if Wikipedia was seen as endorsing any particular company or any particular proprietary product. The general concept of "smart citations" sounds good, but before adopting it, the protocols would have to be in the public domain. Ideally, any implementation would be in the public domain and not rely on data whose access would benefit any person or for-profit entity. That said, we do have precedent for putting things like {{IMDB}} at the end of movie articles, even though this links to the Internet Movie Database, which is run by a for-profit company. I'm just saying that as long as it sounds like someone will benefit financially from your idea, it will have a very hard time gaining traction. Also Jdiogenes86, please read WP:Conflict of interest and WP:Paid editing. davidwr/(talk)/(contribs) 21:31, 27 January 2020 (UTC)

@Sphilbrick: Thank you for the useful suggestions and feedback. I anticipated that this was not something that would just immediately or easily be rolled out to all pages linking to science articles. However, I figured it was worthy of discussion and I am happy to receive such constructive feedback from various members of the community already. In particular, I think your suggestion to build a tool for editors seems like a good practical first step. I am talking to two members from the Wikimedia team soon and I will bring up this idea and others on our call. I'll also reach out to WT:WikiProject Academic Journals and WT:WikiProject Science to get their feedback when I get some time. Thanks again!

@Davidwr: I fully agree with you that it could be dangerous for Wikipedia to endorse products and companies. And, obviously, I am biased and stand to potentially gain financially from such an integration between scite and Wikipedia, even if only indirectly. With that said, I am not sure there is a way to fully implement such an idea in the public domain now or ever. Here is why I think that:

1) In order to provide smart citations, scite (or anyone for that matter) needs access to the full-text of scientific articles. As you may be aware, most research articles are under restrictive copyright licenses held by the publishers. We have signed licesning agreements with these publishers to be able to text mine their papers in order to create the scite website. As part of these agreements, we cannot re-distribute the full-text articles or the citation snippets to any third party. We can and do share citation tallies (the counts of the citations organized by classification) and are happy to open these up even further. However, these numbers without the snippets or links to the snippets themselves, would not be useful in the proposal above. I think it is necessary if we are to classify a citation statement as supporting, contradicting, or mentioning that readers can easily verify this information. It would be doable, of course, but it would be a silly implementation.

2) Citations, let alone smart citations like we are implementing, are still mostly closed and not in the public domain. There has been a massive effort to open citations (https://i4oc.org/), which we applaud, but even this implementation, which is fundamentally more basic than what we are proposing is having difficulty.

Thus, I think the question we should be asking is not can this be made open or not (I don't think it can given the restrictions from publishers) but do we think the benefits of this resource (allowing millions of people to see how a study has been cited, specifically if it has been supported or contradicted) outweigh the costs (driving traffic to a for-profit company). Maybe there is some agreement between wikimedia and scite that could limit any overreach scite would have? This we would certainly be open to discussing.

Also, I appreciate the links to paid posts and COI policies of Wikipedia, I was aware of such policies at a high level I hope that it is clear to everyone reading these messages that I have no intent of trying to misrepresent myself or scite. — Preceding unsigned comment added by Jdiogenes86 (talk • contribs) 19:42, 28 January 2020 (UTC)

@Jdiogenes86: The failure of the perennial proposal Use reCAPTCHA may give you some background for preferring "free" - as in non-proprietary - solutions to problems. If a non-profit entity was doing work similar to yours as a public good, the "who will get rich" issue would be much less of an issue, or likely it wouldn't be an issue at all, assuming the non-profit had good governance, good accountability, had no highly-paid employees, and was otherwise seen as a responsible non-profit. Hopefully, as academic research moves more towards a "free access journal" model, even the non-free-access journals will open up their abstracts, bibliographies, and other "meta data" more than they already do. davidwr/(talk)/(contribs) 20:39, 28 January 2020 (UTC)

@Davidwr: Thanks for the context and history on reCAPTCHA. It makes sense that a non-proprietary solution was implemented there. If such a non-proprietary solution existed for scite, it would indeed make sense to go with that. However, there is no alternative (non-profit or for-profit) that offers what scite offers today. Moreover, if scite were a non-profit it would not have the resources to do what it does, which is quite costly computationally—our funding comes primarily from NIH and NSF SBIR grants which require companies to be for-profit. Why not implement what exists today *freely* to the potential benefit of many millions of people and if a non-profit alternative is created it could then be used in place of scite? Surely, being able to easily see if a scientific report has been contradicted or supported or more generally how something has been cited has more value than seeing who was the lead actor in a recent movie via IMDB, which has been made an exception somehow (their CEO must be better at making his/her case;)). In this scenario, the "worst" outcome is that one private company (scite) benefits from it and that doesn't seem so bad given the upside and the possibility to put some checks in place against potential behavior. I guess the other thing that I think is worthy of mentioning is that wikipedia does link to scientific aritcles that are closed and run by profit publishers, this seems like it would fit under the same umbrella, no?

Thanks again for the discussion and for the help with formatting on these replies. Jdiogenes86 (talk) 22:49, 28 January 2020 (UTC)

Pinging User:JLaTondre, User:Headbomb, and User:OlEnglish, who have worked on Wikipedia:WikiProject Academic Journals/Journals cited by Wikipedia/Questionable1. WhatamIdoing (talk) 21:32, 30 January 2020 (UTC)

Right now, I'm not seeing much use for... whatever this is supposed to be. Taking a random paper like "https://scite.ai/reports/10.1103/PhysRevLett.23.935", we see it's been mentioned 83 times or something. I don't really see what a reader would gain from knowing that in relation to the claim the citation is meant to support in Quark ("Quarks were introduced as parts of an ordering scheme for hadrons, and there was little evidence for their physical existence until deep inelastic scattering experiments at the Stanford Linear Accelerator Center in 1968"). Even if, somehow, the paper had "42 support" in Scite.ai, that still wouldn't mean it supports what we are using it for.

It might be a resource that's useful to editors, so if someone wants to write a WP:User script making use of Scite.ai, sure. But for readers? I don't see it. Headbomb {t · c · p · b} 21:41, 30 January 2020 (UTC)

@Headbomb: Thanks for chiming in. However, I am not sure I follow your argument. It is not useful for readers to see if the reference that underpins a sentence or argument is reliable or not but it is useful for an editor to see it? This might be true if scientific articles did not receive citations over time but they do and these citations can and do change the interpretation/validity of the article itself. This debate can be informative for readers.

Looking at one example in WP:Marriage and health you can see the third sentence says,

"There are gender differences in these effects which may be partially due to men's and women's relative status.¹"

This reference, mentioned 7 times on that page, however has been contradicted, which can be seen on scite very easily: https://scite.ai/reports/10.1007/s11199-011-9968-6. I think this would be useful information for readers and editors to see. Of course this is just one study on one page. I'm now looking at this more systematically and hope to be able to share a report soon. In brief, I've compiled a list of all scientific articles linked on Wikipedia and am now crossreferencing them against scite. I think such an analysis will be interesting and informative about the scientific foundations of Wikipedia. Jdiogenes86 (talk) 15:10, 2 February 2020 (UTC)

Can I ask..what is the protocol for calling a vote similar to other proposals? Jdiogenes86 (talk)

A more interesting proposal would be something like having a bot report on talk pages, where a bot compiles a list of possibly contradicted/outdated citations. Headbomb {t · c · p · b} 19:23, 2 February 2020 (UTC)

Here's what I see as the best chance for you to get something like this implemented on Wikipedia.

• Form a non-profit associated with your company, with an agreement that it can use your technology for specific non-profit purposes only, e.g. doing a project on Wikipedia for open access journals only
• Do this demonstration project and openly license the results and interface used on Wikipedia. In detail, non-profit doing the computation, output only for open access journals, interface (or link) to the non-profits site is freely licensed.
• Best if the non-profit has some continued funding. We wouldn't want to do this for just 2 years and then have to remove all the links!
• The non-profit could have an editor (maybe even a Wikipedian in Residence) inform other editors on article talk pages that the links are ready and where to find them, but unaffiliated editors make the actual decision to insert them.
• You get your proof-of-concept at large scale, all the technology developed along the way, except for the part owned by the non-profit.

I think that would have a 50/50 chance of flying after the details are worked out.

But take away any part of it, I don't think it would fly. Smallbones_(smalltalk) 02:04, 4 February 2020 (UTC)

If this is somehow used on Wikipedia, limiting this only to open access journals is silly. Debunked/unsupported claims in closed access journals are just as important to flag as those from open access journals. Headbomb {t · c · p · b} 21:44, 5 February 2020 (UTC)

@Jdiogenes86: The idea is certainly interesting, although with some issues others have pointed out. Looking at the link you posted, I have a few questions:

• You mention that the citation is contradicted twice. This really makes it sound like it is contradicted in two different papers, but really it is two instances in the same paper. Can a paper simply reference another several times to influence how it is portrayed on your software? Have you looked into whether this (citing something multiple times) could happen more frequently in supporting or contradicting papers?
• Does scite take into account if a paper contradicting another paper is itself heavily contradicted?
• Looking further into the article you linked, six of the results are from one paper: "Marital Quality and Health: A Meta-Analytic Review". I would say that none of those contradict it. Here are the two marked as contradicting it:
  • "While our failure to detect overall gender differences in the relationship between marital quality and health endpoints may appear counter to previous narrative reviews (Kiecolt-Glaser & Newton, 2001; Wanic & Kulik, 2011 b), those prior reviews made suggestive, but not strong claims about the empirical status of gender differences, marital quality, and health outcomes. Moreover, both the interpersonal-orientation and subordinate-reactivity hypothesis focus on "the pathway from negative marital conflict behaviors to physiological functioning""
  • "We found no significant gender differences in the relationship between marital quality and cardiovascular reactivity or cortisol responses, counter to prior reviews (Kiecolt-Glaser & Newton, 2001; Wanic & Kulik, 2011b). Methodological differences between our meta-analysis and prior narrative reviews account for much of the difference."

I'm working on a report now that analyzes all scientific articles on Wikipedia, which is nearly done. Once it is, I will share it here and the data will be made open under a CC BY NC license. I think it will greatly inform this discussion about possible next steps if any. Thanks to everyone for the feedback and questions. Jdiogenes86 (talk)

Please see relevant new discussion at Wikipedia:Bot_requests#Bot_for_lint_errors, essentially I want to get someone to write a bot, then run a bot through some ancient (at first) signatures reformatted by user on varios talk pages (errors on all page spaces), this will fix one main "important" category of errors and some fixes will flow into others. This will allow identifying current users with errors in thier bespoke signatures and advise them to correct span or font issues in the same, reduce overheads going forwards etc.121.99.108.78 (talk) 10:01, 28 January 2020 (UTC)

When I said that this should go to VPR (Village Pump/Proposals), I meant that there should be an actual proposal made at VPR, with a clear explanation of what exactly is proposed and what problems this would address (with an estimated number of affected pages for each 'error'), with a full RFC on the matter. Having more people at WP:BOTREQ won't establish consensus that the proposed task request is a good idea or not. Headbomb {t · c · p · b} 15:55, 28 January 2020 (UTC)

Headbomb, please review the proposal below and comment, adjust or give guidance as required.

Amend signatures that cause lint error

My proposal is to seek approval by consensus to allow adjustments and reformatting of editor signatures which have either formatting errors or depreciated parameters that cause various Lint errors. These errors appear on various talk pages across all page spaces and cause errors mainly in the Tidy bug affecting font tags wrapping links Cat. (currently showing 5,220,072 errors). These errors date from 2006 onwards, fixing these errors would also, in some cases flow through to fix other errors not in the Tidy Cat., help identify current signatures that continue to cause these errors, reduce the overhead in running and maintaining affected pages while making it easier to navigate through the error reports.121.99.108.78 (talk) 07:34, 29 January 2020 (UTC)

1. I want to emphasize that you appear to be talking about a number in excess of FIVE MILLION TWO HUNDRED TWENTY TWO THOUSAND AND SEVENTY TWO PAGES, apparently including an unknown and conceivably larger number of additional unidentified pages.
2. This is not required, but I suggest/request you log in. It's awkward dealing with IPs, and particularly unhelpful when you're making such a massive bot request.
3. Headbomb said above you should include "a clear explanation of what exactly is proposed". I don't feel your proposal is very clear about what changes you want to make, and I find it entirely opaque regarding the unknown number of unidentified "other" groups of changes you want to make.
4. Headbomb said above you should identify "what problems this would address". You have given no indication whatsoever how or why these changes would be beneficial to us. I do not consider lint-numbers inherently significant.
5. Given the above I can't really say anything about your proposal with certainty, other than it's big and unclear. However want to note that moments ago I opened a proposal for a new Village Pump page. There are a number of big issues I hope to raise there, and I believe it is possible that one of those issues might directly exterminate any rationale or desire for your bot run. So unless there's urgent need and immediate benefit, I'm in no rush for such a massive bot at this time.
   Oppose, at least until a more clear explanation and immediate benefit is provided. Alsee (talk) 12:48, 29 January 2020 (UTC)

As someone who has spent a considerable amount manually fixing linter errors myself, I must concur with every point that Alsee raised. Although we're already beginning to see readability issues in talk page archives from deprecated HTML, it's not a serious enough issue to warrant the resources necessary for changing countless signatures, and even if it was, a bot is hardly the way to do it, given that no two lint issues are alike, and require different fixes depending on the lint issue. OhKayeSierra (talk) 15:51, 29 January 2020 (UTC)

• As someone that's written a bot for this purpose, I wanted to chime in here. There are really three separate issues here:

1. Signatures with issues caused by the switch from Tidy to RemexHTML that cause formatting to bleed onto other parts of the page
2. Signatures with issues that cause them to display differently under RemexHTML than they did under Tidy (usually fonts and colors not being applied to text) but that don't affect other parts of the page
3. Signatures that generate Linter errors but appear identical under Tidy and RemexHTML

In the first case, fixing the signatures is important, and has an immediate benefit. However, identifying the error and fixing it in a manner that is consistent with the editor's intent cannot be done automatically. User:Ahechtbot was programmed to run against a manually-created listed of problematic signatures that appeared on hundreds or thousands of pages, but generating and testing these lists is too time consuming for signatures that only affects tens of pages, so there is a long tail problem. The only way to automate it would be to replace the signature with a default unformatted signature, but even that isn't straightforward (just look at the trouble that archiving bots have with certain sigs).

In the second two cases, I don't see any urgency to fix the pages that justifies making edits to literally millions of pages. Linter errors are just that -- errors that show up in a list on an obscure page somewhere but don't do any harm. If you're worried about polluting categories, there's a drop-down that allows you to filter out talk pages. In the long term, a better solution would be to modify the parser so that it goes back to interpreting <font>Foo</font> as Foo|<font>Foo</font>, which would take care of a majority of the errors. There are other cases, such as <font>Foo|Foo</font> which more likely to be problematic and are less straightforward, but could also be corrected by the parser if there is a desire to do so. --Ahecht (TALK
PAGE) 16:48, 29 January 2020 (UTC)

Thanks for the expert info Ahecht. That's pretty much what I was thinking. I believe the real problem here may be well-intentioned-harm that the Foundation has done to Wikitext. The Foundation is working to redefine Wikitext, some individual changes may have been fine, but some are bad. I plan to show that harm on the new Pump page, and explain the long term Foundation strategy behind those changes. (Spoiler alert: Making Wikitext user-hostile is ok because everyone should be using VE instead.) Alsee (talk) 19:34, 29 January 2020 (UTC)

Thanks all, I am going to try to run some numbers in the background, but to clarify it is ~5mil or so errors, not pages, often one talk page will contain multiple versions of the same signature and error.121.99.108.78 (talk) 06:57, 30 January 2020 (UTC)

It's easy enough to get the numbers from the Toolforge DB replicas, although I don't see how User:Alsee121.99.108.78 got 5,222,072 unless someone has somehow fixed half of them in the past 5 days. I count 2,034,469 instances across 712,180 pages. That would still be a lot of bot edits, just to change the colors of some links in user signatures. If we're reasonably confident that user signatures are the only instances remaining, it might make more sense to just ask that Linter no longer report this "error". Anomie⚔ 13:23, 3 February 2020 (UTC)

Ah, probably the count came from Special:LintErrors rather than the linked Special:LintErrors/tidy-font-bug. If that's using the method of estimation I guess it probably is, then "Note: The counts for categories are not exact, but are based on estimates" is what happened there. Yes, those estimates can easily be off by that much or more. Anomie⚔ 13:30, 3 February 2020 (UTC)

OhKayeSierra and User:Ahecht, I wonder whether it'd be possible to make a list of high-traffic talk pages with lint errors, and set about fixing just those. Ancient talk page archives don't feel like a high-priority task, but https://en.wikipedia.org?pojem=Wikipedia:Village_pump_(technical)&action=info says there are four lint errors on that page today, and those might be worth cleaning up. Whatamidoing (WMF) (talk) 00:03, 31 January 2020 (UTC)

I'm not sure I see a reason to fix pages like that. It would just add needless work, correcting invisible errors as they are introduced. On the page you linked to, none are actually bad signatures with the "Tidy bug affecting font tags wrapping links" linter error. One is a typo leaving out an apostrophe when bolding something, one is someone removing text from a substed template and accidentally removing a closing <span> tag, one is someone using <font> tags (which are deprecated, but still supported) in their signature, and one is someone deliberately giving an example of text formatting that doesn't work. A better solution to reducing errors on high-traffic pages would be to (a) scour the prefs database for editors with signatures that generate linter errors and notify them somehow and (b) have something like an editfilter warning to warn editors that their edits are introducing markup errors. --Ahecht (TALK
PAGE) 01:20, 31 January 2020 (UTC)

:Oppose until I can take an IP seriously. >>BEANS X2t 13:34, 3 February 2020 (UTC)

• BEANS X2 If you have nothing nice to say then please keep your opinions to yourself, Not every IP is a dumbass. –Davey2010^Talk 20:30, 5 February 2020 (UTC)

Sorry, wasn't having a great day. >>BEANS X2t 13:38, 6 February 2020 (UTC)

• Oppose - I'd imagine there's a whole ton of problems when it comes to Lint and sigs ..... For instance my 2013 sig heavily used </font>.... I don't see how fixing these can really be achievable and if they could then there wouldn't really be any benefits to such a huge move, Sure for instance font is deprecated but it still shows and I assume it will continue to carry on showing .... –Davey2010^Talk 20:27, 5 February 2020 (UTC)

I want to submit the following proposal. Why not make the profiles of the editors look a little more professional by making available (optional) details about yourself like a photo, educational background, the city and country and so on. That will look more professional and of course, the information that you additionally put, will be optional. I think that will improve Wikipedia even more.— Preceding unsigned comment added by Albi Ndoni (talk • contribs) 16:01, 4 February 2020 (UTC)

Albi Ndoni You can already put most of the information you describe on your user page if you wish. Your user page is for introducing yourself to the Wikipedia community in the context of your Wikipedia editing or use. It isn't a place to tell anything and everything about yourself, though; please see WP:USERPAGE for what is considered to be acceptable user page content. 331dot (talk) 16:12, 4 February 2020 (UTC)

I assume you're talking about editors' user pages? Some people do exactly that. Some people don't bother. Some people intentionally don't do it so as to preserve their anonymity. In general, we leave decisions about how editors manage their own user pages pretty much up to them. Personally, I think the emphasis should be on the quality of the content that's being produced, not who's producing it. -- RoySmith (talk) 16:16, 4 February 2020 (UTC)

I agree with the last point. I don't keep a user page myself, and only reveal personal details (apart from my name) when it is relevant to the context, for example it may be important in a discussion for people to know what my reading level of a particular language is. If you want to know more about me then I'm sure you could find enough in my editing history to distinguish me from the handful of other people who share my name, but I prefer not to make it too easy to do so. If you want to make it easier for editors who wish to do so to supply details then there are almost certainly templates that format things professionally, or you could write one yourself, but this is not part of Wikipedia's core mission. Phil Bridger (talk) 19:26, 4 February 2020 (UTC)

Correct me if I'm wrong, but don't we generally discourage the use of PII where unnecessary? A userpage (not a "profile") is generally for sharing information relevant to editing Wikipedia, primarily interests and language proficiency. Anything more in depth is discouraged, with common sense dictating that you wouldn't want to give trolls more ammunition against you than they already have - especially if you're an admin or work in an area that's filled with partisans. —A little blue Bori v^_^v ^{Onward to 2020} 20:25, 4 February 2020 (UTC)

@Albi Ndoni: I suspect what you’re looking for is the template {{Infobox Wikipedia user}}, which allows you to display that information in a format similar to the Infoboxes found in many articles. Of course, you don’t need to fill out every field. You can find examples of how to use the template here. —pythoncoder (talk | contribs) 21:32, 4 February 2020 (UTC)

Because of Identity theft, a degree of caution is needed.--Dthomsen8 (talk) 02:33, 8 February 2020 (UTC)

World Book Encylopedia Encyclopaedia Britannica

(indentation)As we can all see in the above examples, there are no indentations at the beginning of paragraphs on English Wikipedia, but there are in World Book and EB. American and British English are both important parts of English, and after reading some books from those regions and getting an education in the USA, I've become convinced that indentations at the beginning of paragraphs are definitely a part of the English language on some level, even if they can be shown not to be used in some circumstances or to be looked down upon in some manuals. I suspect this issue has come up before many times, but I would like to bring it up again if only so that you all can know that there are and will always editors dissatisfied with the current state of affairs regarding paragraph indentation. I urge the community to respect the culturally distinct formatting used by the English language speakers (especially American English, etc) and not force us to use what I would see as an alien or continental non-indented paragraph convention. It would be a sign of respect for the history of the English language pre-Wikipedia to just go ahead and use the conventions of the English language as they truly are and not as we might wish them to be or plan them to be. At bare minimum, we ought to be allowed in some circumstances and on some pages to write in our native language unimpeded, especially pages marked for American English. Thank you for your time and consideration. NOTE: Because I personally feel this topic is likely to impugn the authority of the powers that be on this website on some level, I do not plan to follow up to any responses here because I am in dread fear that I may be blocked if I use the wrong wording here. Even to bring up this topic may accidentally cause myself to get a ban. But I feel quite passionate about this, and I think that as an editor of several years and many edits, my continued silence would make me complicit in what I would see as a mistake. I make the post in a spirit of friendliness with a hope to improve the encyclopedia by returning to what I see as bona fide English instead of "hip & cool" no idents on paragraphs. Because of the likely/possible controversy of my proposal, I reiterate that will not respond to pings on the issue. Thanks again. Geographyinitiative (talk) 04:33, 14 January 2020 (UTC)

You seem to think we will ruthlessly crush a dissent as powerful as reads up... asking ... for indents... Friend, I don't think you have anything to worry about. Then again, since you don't care about the reaction, I guess this is just a fun time dance party now? (Also, you may notice that while print works that you linked are indented, internet pages are routinely not. It's because writing for print and for the screen are Different Activities, with Different Rules. --Golbez (talk) 04:39, 14 January 2020 (UTC)

I wonder if paragraph indents are something users can create with their .css or .js settings? --jpgordon^{𝄢𝄆 𝄐𝄇} 04:42, 14 January 2020 (UTC)

Ayup. Something along the lines of div#mw-content-text p { text-indent: 2em; } in your CSS will do it. —Cryptic 04:45, 14 January 2020 (UTC)

• I know the MOS-ites are a powerful crew not to be crossed, but I don't think they've escalated to acquiring community bans for raising the topic of paragraph indenting, please feel free to engage the conversation, it makes it much easier to discuss it. Jpgordon raises a good point that what you really need to show is a significant group of English writers who indent online paragraphs. I've not spotted that but for all I know I just don't note it. Nosebagbear (talk) 12:22, 14 January 2020 (UTC)
  • We don't even need to show that other websites do it. We just need to show that editors want it. User:Geographyinitiative, the code that User:Cryptic posted should let you have indented paragraphs. It might be good to try it out, and see how well it does (or doesn't) work for you. Just put it in Special:MyPage/common.css (and don't worry about the warnings – malicious code could screw things up, but this isn't malicious, and if it doesn't do what you want, then you just go back to the same page and edit it to remove that code). WhatamIdoing (talk) 18:53, 21 January 2020 (UTC)
• I just pray that nobody blocks or bans the user for using that CSS code. – Levivich 01:05, 4 February 2020 (UTC)

I have spent my while life indenting paragraphs. The issue here is that we don't do it just because everyone else does. I am suppose I would be open to "discussing" a change but will not likely change the css code to accommodate this. I don't think I would "correct" such an article (haven't previously thought about it) with consistent indented paragraphs should should I run into it. Without clear guidance or a mandate (consensus) I would certainly object to editorial sanctions if someone did use indention's (a code allowing it), unless maybe there were issues like: It was objected to (changed or reverted) because of our current practices and it was reverted without discussion.

Also, I would be more responsive to a discussion where I don't see "I reiterate that will not respond to pings on the issue". Change may be a good thing but someone "might" have had reasoning to want to present questions to the initiator. I am not a fan of "discuss it here but leave me out of it" that seems to be presented. Otr500 (talk) 13:13, 8 February 2020 (UTC)

I don't think it is fair that the community has no say who gets to be a CheckUser and who gets to be an oversighter. I propose that the requirements be similar to an RFA (having an account and being extended confirmed), however, an additional requirement would be that the candidate must have signed the access to nonpublic information. If a candidate is nominated and has not signed it, the candidate would be asked to sign it or if they choose not to sign it, they cannot be nominated. Interstellarity (talk) 23:55, 8 February 2020 (UTC)

Community comments were invited for the last round of appointees, at least: Wikipedia:Arbitration Committee/CheckUser and Oversight/2019 CUOS appointments –xeno^talk 00:41, 9 February 2020 (UTC)

(edit conflict)As an opening thought, while an ARBCOM member's usage of OS/CU (which is mainly informative) could operate without an admin's toolkit, many of the tasks that a CU/OS role is meant to fulfill would require an admin rights - CU rights in particular are normally used when considering socking issues, which is inherently tied up with blocking. If that then required a separate person to action you'd be duplicating effort. Beyond that, Wikimedia Legal have said that access to even general deleted content requires a full RfA, I suspect a more general community discussion wouldn't be enough to tick that box. Nosebagbear (talk)

Still, that's all about the candidate pool, certainly the community could do the consideration of CU/oS rights, however, I'm not sure how much help it would serve. The community gets encouraged to provide feedback on CUOS candidates when ARBCOM considers them (at least in my time), and I've not seen individuals accepted who've had significant community resistance. Nosebagbear (talk) 00:47, 9 February 2020 (UTC)

Suggest reading though this somewhat recent discussion before deciding if you want to have a new one Interstellarity. (Wikipedia:Village_pump_(idea_lab)/Archive_29#Idea_Poll_-_independent_CUOS_Elections). — xaosflux ^Talk 00:55, 9 February 2020 (UTC)

@Xaosflux: I didn't know this has been discussed before. I read through the discussion. Not sure what to do from here. Should we continue with the discussion or should I delete this post? Interstellarity (talk) 01:04, 9 February 2020 (UTC)

@Interstellarity: up to you, if you want to continue by all means continue; if you want to withdraw your proposal for now just leave a note below and it will eventually be archived. Best regards, — xaosflux ^Talk 01:07, 9 February 2020 (UTC)

@Xaosflux: I will let the proposal run and see where the consensus goes from here. Interstellarity (talk) 01:08, 9 February 2020 (UTC)

FYI

 – Pointer to relevant discussion elsewhere.

The AfD suggesting merge/delete and redirect has closed, and another discussion about merging all three LotR volumes' articles into The Lord of the Rings has now opened at Talk:The Lord of the Rings#Proposed merge of The Fellowship of the Ring etc into The Lord of the Rings, as of February 4.

Listing this here because I think this renewed proposal has implications (one way or the other) for our treatment of other multi-part works, including non-fiction ones (cf. Das Kapital, and Das Kapital, Volume I through Das Kapital, Volume III), and non-book ones (e.g. separate TV episode articles, and many other such things). This is a bit of a meta-discussion, on what concerns/plans/wishes/complaints can be considered to simply override WP:GNG for alleged broader WP:ENC rationales.

Also, I find this procedurally a bit dubious. The AfD closed with a consensus against deletion (in a proposal to merge and delete then redirect), albeit not an overwhelmingly strong one. A back-to-back proposal to merge and delete then redirect via off-AfD talk page discussion seems a bit out-of-process, even if the AfD closer expressed some uncertainty on the matter (and that last bit means the proponent is surely acting in good faith to clarify/resolve, not trying to forum-shop). I think this verges on WP:CENT-worthy, however, because zero relevant talk pages were notified, even of the affected articles and of WikiProject Middle-earth (though all four articles received a merge tag; some process was followed). I have rectified the missing notifications (also to WikiProjects Fiction, Fantasy, Novels, and Literature, and WP:PM). Nevertheless, the discussion has been going on since Feb. 4 without many people knowing about it other than a few who followed it directly there from AfD. Perhaps the notices today are sufficient to draw in a lot more eyes and brains to consider the matter. I'm not listing it at CENT yet, but would not object if someone else wanted to. I think the horse is too far out of the barn to close it as out-of-process, so it just needs to run with sufficient input for a proper consensus.  — SMcCandlish ☏ ¢ 😼  10:24, 9 February 2020 (UTC)

I agree, for what it's worth. I only saw the discussion because you posted it here. The Drover's Wife (talk) 10:32, 9 February 2020 (UTC)

We have the 100K Edit Star, shown here. Should we award a 200K Edit Star, perhaps 300K, etc.?--Dthomsen8 (talk) 02:25, 8 February 2020 (UTC)

Although edit counts go into the millions, stopping at 400K or 500K would include most of the editors over 100k. See Wikipedia:List of Wikipedians by number of edits for details.--Dthomsen8 (talk) 03:25, 8 February 2020 (UTC)

This would appear to be a singular waste of time. Anyone who's been around for that length of time (there are only 174 of them, many of whom haven't edited for years and quite a few of whom were actually unauthorized bots) is pretty much by definition going to have enough experience on Wikipedia to know that once you get past around the cutoff of around 1000 edits that separates "inexperienced" from "experienced", edit count is one of the most meaningless metrics on Wikipedia. (I could have racked up 10,000 search-and-replace typo fixes with a script in the month or so it took to make this single edit; it doesn't mean it has 1⁄10000 of the value of fixing an instance of "doe snot".) If anyone really feels the urge to pretend a high edit count somehow makes them important, they're already perfectly capable of putting "I have 200,000 edits!!!!" in 50-point dayglo type on their userpage. ‑ Iridescent (303,408 edits at the time of writing and not pretending that makes me some kind of super-user) 09:15, 8 February 2020 (UTC)

So true. · · · Peter Southwood ^(talk): 16:52, 8 February 2020 (UTC)

Hmph. BD2412 T 22:38, 8 February 2020 (UTC)

I oppose anything, such as this proposal, that values quantity over quality of edits. Phil Bridger (talk) 21:47, 8 February 2020 (UTC)

Sheesh this award is just for fun. It should be noted that bots were removed from the list many years ago. Also, most of the names on that list are still editing. I know because they are hard to catch up with. MarnetteD|Talk 23:21, 8 February 2020 (UTC)

Additionally this award simply acknowledges quantity - it in no way shape or form "values" that number over quality. BTW there are numerous Wikipedia:Barnstars to be awarded for quality editing. MarnetteD|Talk 23:29, 8 February 2020 (UTC)

Perhaps I expected too much form my proposal. An easier approach than to "award" the stars with a bot would be to create the stars, and let the users award them to themselves or to friends who qualify.--Dthomsen8 (talk) 23:38, 8 February 2020 (UTC)

Oppose per Iridescent and Bridger. It may be just for fun, but it remains that a vast number of editors don't and wouldn't see it that way. I long ago stopped displaying the Wikipedia:Service awards because they were becoming increasingly misleading. To proclaim that I am now a Master Editor (or Illustrious Looshpah) (soon to be promoted to Auspicious Looshpah) would be a gross and embarrassing overstatement of my contribution to this project. Those awards are likely here to stay as a result of inertia, but we don't need to be doing things to increase the magnitude of the problem. Edit count means less than nothing. ―Mandruss ☎ 23:59, 8 February 2020 (UTC)

I tend to agree with the critics above. Cute badges like this are primarily for noobs and semi-noobs. There's a certain subset of incoming people (especially video game players) for whom "badges" or "achievements" are personally meaningful, and there's even a small subset of people for whom this sort of thing is appealing always and forever, but its a pretty small slice of the editorial pool. And some of us who don't psychologically need these things will display them anyway, as noob encouragement. But anyone who's hit the 100K edits mark (by whatever means) is probably already over it; they don't need new badges for 200K and 500K to encourage them to keep participating and feeling good about it. They're already here for the long haul, and we sure hope! for WP:HERE / WP:ENC reasons. And noobs are not going to be encouraged any further by 200K awards than by 100K awards, which already seem like they're almost impossible to achieve when you're that green.  — SMcCandlish ☏ ¢ 😼  01:00, 9 February 2020 (UTC)

Heh. An edit count of 100,000 still seems almost impossible to achieve and I've been here for thirteen years. Give me ten more and I might make it. Useight (talk) 15:55, 10 February 2020 (UTC)

That comment prompted me to have a look when I got to 100,000 edits, and I was disappointed to find that I'm not even half way there despite my similar length of tenure. I think I need to pull my socks up by doing in ten edits what I currently do in one. Phil Bridger (talk) 17:59, 10 February 2020 (UTC)

Try getting into the counter-vandalism game, it's great for inflating your edit count! Automatic tools, just click one button to revert an edit, and every rollback you do counts double since you then post on the user's talk page! You'll get to 100k in no time at all. ^creffpublic _{a creffett franchise} (talk to the boss) 18:04, 10 February 2020 (UTC)

@Dthomsen8: We already have Barnstars and Wikipedia:Service_awards. If you want to create a new award, I suggest you create a new Barnstar that can be awarded like any other. RudolfRed (talk) 01:59, 9 February 2020 (UTC)

Apologies if this is not the right forum for my question/proposal. I frequently notice how clunky and difficult it is to generate proper references on Wikipedia. I am at a loss to understand how a tool that works quickly and properly to expand references is not part of the standard interface. Yes, there is the citations gadget, but it does not work half the time and takes forever the other half of the time. And there is WP:Refill. And there is the manual cite web/book/journal tool under cite on the main toolbar. But none of these are fast and reliable built-in solutions. I don't understand why things are so clunky on a site where one of the most important aspects of what we do is providing proper referencing. So if there is proposal to be made here, it is perhaps that Wikimedia, or whoever has the money to hire a team, should get on this?ThatMontrealIP (talk) 20:12, 8 February 2020 (UTC)

The ultimate problem is that there is no reliable method to gather the citation information from any arbitrary web site, as the information is not presented uniformly. For better or worse, the more seamless a tool is, the more people will assume it is reliable. Reusing citation data is perhaps a more achievable goal, using Wikidata as a central repository. Unfortunately that won't help editors add new citations. It's been a while, but last time I added some assertions to Wikidata and then added citations for them, I went down a rabbit hole of creating items for every part of the citation. Unless that has gotten easier, it's pretty cumbersome. isaacl (talk) 21:18, 8 February 2020 (UTC)

Yes, it is no doubt tedious. But if it is doable by a human in a couple of minutes, it is certainly doable by software (perhaps with a dash of AI) much more quickly. I don't see why significant funds have been not expended on this. It is just a matter of looking up the URL, intelligently parsing the site, and formatting that as a reference. At minimum most cites will include the html title tag, which fills in the title part. it is not a trivial software job, and not an impossible one either. But it is certainly doable and I do not understand why it is not a priority to make it part of the principal interface. Or maybe they are already working on it?ThatMontrealIP (talk) 21:30, 8 February 2020 (UTC)

I'm not sure if this is exactly the same issue, but it has long been a bugbear of mine that it is far easier, with the automatic tools available, for editors to nominate an article for deletion than it is to add references from sources such as books and academic articles indexed by Google Books and Google Scholar. Anything that encourages editors to actually look for sources has my support. Phil Bridger (talk) 21:46, 8 February 2020 (UTC)

There is a tool that automatically generates citation data from a website, it's called Zotero. It's a little cumbersome to set up: you have to install an application on your computer and a browser extension. It also doesn't work for every website, but the times that it does work, it can save quite a bit of time. More info can be found at Wikipedia:Citing sources with Zotero, and feel free to ask me any questions you might have because I think that page is a little outdated now. --Secundus Zephyrus (talk) 21:49, 8 February 2020 (UTC)

Secundus Zephyrus, thanks for that. Integrating something like Zotero into the interface is exactly what i am talking about. Zotero is open source. At the moment we are relying on user-generated tools like ReFill to fill refs. Why isn't the WMF working on this? Zotere is Open source so it is an example of something that could be forked to work. There should be a bulletproof way to add refs reliably, and lots of money should be thrown at it until there is. At the moment we just have hacks, workarounds and optional gadgets, many of which were written by external parties. Phil Bridger's point that it is easier to AFD an article than it is to add references is a point well taken.ThatMontrealIP (talk) 22:17, 8 February 2020 (UTC)

Here is what the WMF research section says they are working on, sources-wise.ThatMontrealIP (talk) 22:21, 8 February 2020 (UTC)

This isn't the place for it, but I do somewhat wonder whether this is the type of thing that leaning on some of our partners, ranging from Google to the Open Software Foundation and trying to nudge towards some vaguely more co-ordinated form of presenting the key info would be worthwhile. I'd have absolutely no complaints about the WMF working towards a zotero-esk approach, but I'm well aware of how tough it could be to do. Sometimes I find a ref and extracting the information manually can be tough enough - AI is not automatically better at recognition tasks. Nosebagbear (talk) 00:37, 9 February 2020 (UTC)

@ThatMontrealIP: Zotero is already integrated into the interface: the VisualEditor and 2017 wikitext editor have a truly built-in tool for adding (and expanding) citations, powered by Citoid, which in turn uses Zotero. SD0001 (talk) 12:34, 10 February 2020 (UTC)

Handling PDF's from a URL could certainly be improved. When I look at PDFs' properties there is usually a date given. The author though in there is pretty useless. In the visual editor I would like to see it handle patents and theses. Does it handle news paper sites to generate a cite news? Graeme Bartlett (talk) 22:11, 10 February 2020 (UTC)