This page contains discussions that have been archived from Village pump (idea lab). Please do not edit the contents of this page. If you wish to revive any of these discussions, either start a new thread or use the talk page associated with that topic.
Push for the WMF to start a program involving select, reputable university departments. Editors can formulate a question on an academic or scientific topic; once that question gets vetted, it is shared with these departments, who are invited to provide short statements stating the academic consensus on the topic. These academic statements would inform our content discussions.
Rationale:
Why couldn't we accomplish the same thing on our own by just searching through peer-reviewed literature? Because (a) passed peer-review ≠ matches academic consensus, we're all Dunning-Krugers here, and (b) academics tend to not publish things that their peers would consider obvious, that may not be obvious to us.
It may help us de-escalate our high-conflict contentious topic areas. See for example the Holocaust in Poland debacle (and ArbCom case). It would also be a first step to countering civil POV pushing, which is currently completely unremedied, and it would make us less reliant on the vigilance of editors with the "right" POV.
It may also help address what Elemimeleargued in the previous discussion: Ideas That Went Nowhere after being published in peer-reviewed journals, whose current standing is hard for us to assess since they hasn't been addressed in more recent sources. We widely include such things, which poses NPOV concerns.
These statements would come from whole departments, not from individual academics, so we'd minimise COI and possible FRINGE concerns. We'd be the ones doing the editing, so this overcomes the WMF's unwillingness to get involved in content. And this doesn't depend on academics being willing to waste inordinate time learning how Wikipedia works and making their own contributions. It finally provides them with an approachable venue to offer content feedback, which is something academics mostly gave up on doing a decade ago due to their feedback often being ignored.
I'm sympathetic to the goals, especially w/r/t to the Ideas That Went Nowhere problem, but this doesn't sound workable.First, while individual academics might have the free time and interest to contribute to a project like Wikipedia, there are little to no incentives for a whole department to assign their already overworked resources to improving Wikipedia. I don't see how this is a more approachable venue to offer content feedback than e.g. article talk pages. Influential, sure; approachable, no.Second, the invitation-based selection process sounds fraught with problems w/r/t e.g. systemic bias. There's also the problem of how is WMF (or the community) supposed to determine who the right departments to invite are, if the whole underlying thesis is that the WMF/community does not have the expertise to evaluate the state of research.Third, I see no reason to believe that having a department ostensibly sign off on a comment would actually do much if anything to address WP:COI or WP:FRINGE. Just as individuals have their idiosyncrasies, so do communities such as departments. Framed in terms of the Hierarchy of Influences model, at the very best it removes the effect of the individual, but that is already largely achieved through our normal editing policies' requirements for reliable sources etc. At the worst (in the very realistic scenario where the statements by "departments" are de facto prepared by that one academic who has a bit of free time and wants to contribute to the project) this would instead entrench the idiosyncrasies of said individual, hiding them behind a veneer of "academic consensus".Fourth, w/r/t academics tend to not publish things that their peers would consider obvious: in my experience, this is because rather often these obvious things are more informed guesses or hunches and less positions well-grounded in data or research. If anything, the citation-based incentives of the modern academic world incentivize publishing a paper that says "that one thing everyone thinks is true, but doesn't have a reference for, indeed is true".Fifth, I don't see how this could be aligned with our policies on e.g. original research and synthesis. Either the comments simply regurgitate reliable sources - in which case this reduces to a more complicated multi-tiered variant of the standard editing process - or they contain OR/SYNTH, in which case we'd have to carve out some kind of an exception to those policies for a cabal of "approved original researchers", which, uh, I think would have a snowball's chance in hell of being accepted by the community. If the idea is that we'd interpret the comments as subject matter expert-authored self-published sources, then this just sounds like a lot of extra bureaucracy for what could be achieved by asking a department to write a blog post on their departmental website. Ljleppan (talk) 08:05, 27 February 2023 (UTC)
Removing disincentives is useful for attracting individuals (and certainly a worthy goal), but probably not enough to attract organizational partners (like departments). Overcoming the organizational inertia is no small task even if we find someone to "internally champion" participation, if all they have to go with is "it'd be nice" and the organization is likely already resource starved. Ljleppan (talk) 05:28, 28 February 2023 (UTC)
"then this just sounds like a lot of extra bureaucracy for what could be achieved by asking a department to write a blog post on their departmental website" hmm, despite my concerns about the policy in general this is something I've occassionally wanted and I might support a process for this. A problem I found in psychology was "pop-science" diagnoses proped by with a network of blog posts that would be better understood through other concepts. A good examples is Victim mentality which might be better understand through Trauma and things in Victimology. I guess WP:PARITY is the reason I'm happier with this - if there are no sources any source by an academic would be useful. On the other hand, I don't know how motivated most academics are by "there is insufficient literature for wikipedia to adequalty address a topic to its exacting standards". Talpedia (talk) 12:13, 28 February 2023 (UTC)
I think this could, in principle, already be done within the scope of the subject-matter expert exception of WP:SPS. WP:MEDRS would naturally rule out that stuff for biomedical claims, but that's probably for the best, and WP:EXCEPTIONAL should rule out any outrageous stuff in general.W/r/t I don't know how motivated most academics are by..., it's indeed tricky. Popularizing science or science comms tend to either be ignored, or are given very low weight in contrast to most other academic activities, when it comes to academic evaluations etc. so there's little external motivation. It's also not helped that we'd not be simply asking "hey prof so-an-so, what do you think about X?" by email, but instead "could you write a thing about your view on a potentially controversial topic, copy edit it to your standard of public writing, and post it for eternity on e.g. your department's blog". I'd imagine many (if not most) academics would be happy to give a brief answer to a private emailed query, but asking them to write something they'd be willing to sign their name on is a much bigger ask. Ljleppan (talk) 07:22, 1 March 2023 (UTC)
Things like this have been done before, usually with individual experts rather than whole departments. See Wikipedia:BMJ/Expert review for one example of an organized program.
The individuals who have agreed to do this in the past have generally suggested smaller changes and were happy to provide sources. Some want us to WP:SELFCITE for them (and sometimes we reach out to the person because their source was so useful, so that seems fair), but most are happy to recommend sources that they didn't write as well. There's no need at all to worry about academics trying to add "material—such as facts, allegations, and ideas—for which no reliable, published sources exist" (=the actual definition of OR according to our policy) because these people literally have their jobs because they know how to cite credible sources. Also, much of our content can be found in textbooks instead of recent papers, and sometimes what we need is someone to tell us to remove certain claims that the field generally considers disputed, unimportant or nonsense. WhatamIdoing (talk) 02:21, 28 February 2023 (UTC)
The problem comes with certain claims that the field generally considers disputed, unimportant or nonsense. Either this is something that can be easily identified from RS (i.e. a textbook or something like that says "X is generally not considered a credible theory", or it's not even discussed by others, at which point WP:FRINGE's Fringe theories may be excluded from articles about scientific topics when the scientific community has ignored the ideas. and A conjecture that has not received critical review from the scientific community [..] may be included in an article about a scientific subject only if other high-quality reliable sources discuss it as an alternative position. come into play), at which point there is little need to set up this massive bureaucracy, or it's not easily identifiable from RS, and we're at or approaching OR/SYNTH/a situation where there is no verifiable consensus.I'd also highlight that organizations - consisting of individuals as they are - have all kinds of idiosyncrasies, just as the individuals do. Every single research organization has their own collectve "hot takes" on exactly the kinds of issues where this proposal is intended to be useful. "Ask a group of people who work together" is not the panacea this proposal appears to assume it is. It might filter out the most burning hot takes, but I fear it will only entrench and hide behind a smoke screen the types of medium-hot takes that we'd -- presumably -- need the most help with. Ljleppan (talk) 05:59, 28 February 2023 (UTC)
Kind of doubtful. I think academic review and input is great.
My concerns are that you get this "false consensus" effect in academia. Basically, if you aren't working directly in a topic your understanding will be based on "the sort of things that people say" and what *must* be true rather than what you are certain of. I feel that a whole lot of wikipedia is and should be add odds with the consensus within the wider field. It should be the consensus of the few dozen academics who have actually spend time grappling with or synthesising the topic.
I can't help but feel that this approach seeks an authority that doesn't exist, and then elects one that won't necessarily care very much about what is actually true: "it's just science communication". I sort of view wikipedia as acting *against* this sort of "expertisation". We take the things that academics are happy to share with one another - where they can't just win by "being an expert" and then share this with the readers, thereby imposing academic scrutiny by proxy to issues. Normal science communication seems to more be along the lines of "what some guy said to a writer with a science-background simplified in lines with their values". Talpedia (talk) 11:26, 28 February 2023 (UTC)
Also fun fact. The Dunning-Kruger effect was a concept created by citogenesis with wikipedia and the original research applied to assumptions of competence within the population rather than ability to perform given tasks - though there has been later research, which may look into this. My impression is that the "generalised" version complete with diagrams including "mount stupid" isn't well supported by evidence and a convenient fiction that lines up with people's biases about expertise. Talpedia (talk) 11:37, 28 February 2023 (UTC)
Like many "fun facts", that's actually not true. The concept was unnamed but was known, Dunning and Kruger's research existed and was a topic of discussion before Wikipedia even existed. Dunning and Kruger published their initial findings in the late 1990s, Wikipedia itself didn't exist until 2001. It was not citogenesis that created the concept. However, the Wikipedia article about the subject popularized the name to the concept; that is all. The concept existed, it was the name "Dunning-Kruger effect" specific that was coined by a Wikipedia editor, not the effect itself. User Uucp coined the term "Dunning Kruger syndrome" at 19:21, July 22, 2005 (UTC) in creating the article. User Vaughan created the specific phrasing Dunning Kruger effect with This edit at 16:28, May 26, 2006 (UTC). To my knowledge, that's the first time the words "Dunning Kruger effect" are documented to have been used. However, this did not create the concept. As noted, it existed, it just wasn't named for Dunning and Kruger prior to 2005-2006. --Jayron3216:19, 3 March 2023 (UTC)
Considering these two discussions, I believe that we need to determine whether there is a consensus on enwiki for the UCoC; this will help inform how we consider the code of conduct locally, and it may help the WMF determine how to proceed with applying the code to larger self-governing projects like our own.
My preference is to do this via a widely advertised RfC, as I believe editors reasons for support or opposing the UCoC are as important as their support or opposition, but I understand some editors prefer a SecurePoll vote. I'm opening this discussion to determine whether we should proceed with such a poll, and if we do where it should be held and how it should be worded. BilledMammal (talk) 05:22, 4 March 2023 (UTC)
I am reticent as to whether such an action could be formed as a valid RfC, that complied with the underlying consensus policy.
It would also be an RfC that was hyper-vulnerable to interpretation difficulties. For example, do I believe there is a UCOC, yes. Do I believe it has a mandate? No. Would *I* carry out an en-wiki sanction based purely on the UCOC? Again no. But I also don't think that a) An RfC could be validly made that would comply with CONEXCEPT and b) any such RfC must also (if you're going to run), be done so with a basis of specifying what actions should be taken if/when it passes and declares UCOC-breaching but not en-wiki b reaching sanctions out of policy here - and what happens when en-wiki is bought to the U4C for systemic failure to uphold the UCOC.
Perhaps a better discussion and then RfC would be - what steps can en-wiki do to compel the BOT who, for reasons that still surpass comprehension, refused to have a community vote on it, to do so. Perhaps we should start with a formal community position that the Board's direct implementation of such was in error, as were all individual trustees who voted to ratify it without such a community agreement in line with the EGs, and propose that such a vote be included into the building committee stage. Nosebagbear (talk) 20:46, 5 March 2023 (UTC)
I don't think there is an issue with CONEXCEPT; CONEXCEPT only allows the WMF to ignore any consensus that we come to, and doesn't prevent us from coming to a consensus. I also wasn't initially considering a question on whether we should enforce the UCoC, but rather whether we should endorse it - but now that you have made the suggestion, I do wonder if that is the way to go.
It is very hard to plan a response to us being in breach of the UCoC without knowing what measures the WMF would take. However, I suspect that whatever measures they might take having a consensus behind the decision, rather than admins individually declining to enforce it, would make it harder for the WMF to take any action.
I also struggle to imagine circumstances where we are in breach of the UCoC; the closest would be the WMF interpreting it in a way we consider unacceptable (for example, WP:FRAM).
As for compelling the BOT, the only tool we have is the one proposed at the banners RfC, but I don't consider that tool to be appropriate here and I don't think you would support that tool. BilledMammal (talk) 10:04, 7 March 2023 (UTC)
Adjust scientific name across biology related articles
Wikipedia style, expressed at MOS:LIFE is that scientific names at the level of genus or below are italicized. I am not in the field, but the biological sources I see use italics much more often than they underline. SchreiberBike | ⌨ 16:46, 7 March 2023 (UTC)
(edit conflic) In the field of biology scientific names are usually italicized, not underlined (although underlining is an option if italics aren't available, e.g. when using most typewriters). If you find any scientific names that aren't italicized, please italicize them. There is no way for the Wikipedia software to know that a string of characters is a scientific name; editors must provide the wiki markup to produce italics. Plantdrew (talk) 16:54, 7 March 2023 (UTC)
The arguments to make section has really not been explored in the same fashion that the arguments to delete section has. Are there more arguments to make than listed? There are, but not having the WP hyperlink gives the impression that they don't exist or aren't important.
Adding a template and increasing the Wikipedia:Arguments to make in deletion discussions would be really helpful to newer or less experienced users. I know I don't know the full list of tools I have at my disposal, and perhaps making a template and adding to the arguments to make section would help.
I also do believe the nominator of an article for deletion has an outright advantage over the person defending their article in general and this might help that.
KatoKungLee (talk) 14:37, 8 March 2023 (UTC)
Stay on topic, cite reliable sources, explain how policies apply, all seem like good things to do in any argument. --Jayron3215:03, 8 March 2023 (UTC)
The main page is already holding about as many sections as it can; more clutter does not improve usability. --Jayron3213:48, 3 March 2023 (UTC)
The motto of the day project focuses on topics related to editing. I don't think this would fit in on the main page, which mainly serves to showcase encyclopedic content and as a directory of useful pages. small jarstc16:36, 3 March 2023 (UTC)
I like the motto of the day and so put it it at the top of my user page. The project which created them is quite static now too and so the mottoes are drawn from the repository which was built up in its heyday.
The main page could use some additions though. To see the possibilities, try using the Wikipedia apps which have added several sections to their version of the main page including:
Top read
Random article
Suggested edits
Because you read
Nearby places
They also give the user the option to customize the selection of sections so that readers have a choice rather than a static format.
"Top read" would also be gameable by bad-actors. E.g., set up a bot to hit an article on your product from a zillion IP addresses just to vault your product into a highly visible placement on our site. And "Because you read" is just creepy. — SMcCandlish☏¢ 😼 20:05, 8 March 2023 (UTC)
Reflist by default
A new article could have a "References" heading and a Reflist template by default at the bottom. This would make it so that it doesn't have to be added manually. Born25121642 (talk) 23:53, 9 March 2023 (UTC)
Impractical. Not all mainspace pages have references, there are different ways to name this section, and depending on the citation style there may be two or three sections dedicated to references and citations. See MOS:REFERENCES and Help:Shortened footnotes for some documentation on different possibilies. Separately from it being impractical, I am against blank sections. —Alalch E.01:35, 11 March 2023 (UTC)
Category list scraping for WikiProjects
While updating project templates for a WikiProject, I'm finding there's quite a number of articles under the associated category (category:astronomy) that do not have an associated Wikiproject template (template:WikiProject Astronomy) on the talk page. It would be helpful if there was a script that could drill down under a category to a given depth and look for articles lacking a specific WikiProject template. For example:
list =: seek_no_WP_template(sWPTemplate, sCategory, nDepth);
@Praemonitus:, you want PetScan. Unfortunately, it's being having issues executing some searches for months now (searches that take forever to run and erroneously returns no results), and the situation is particularly bad when trying to search for templates on talk pages, which is exactly the functionality you want. I have a link on my user page to a search that for some bizarre reason seems to usually execute even when going directly to PetScan via the first link I provided has execution failures. I've adapted it to search for pages in Category:Astronomy in China, drilling down one level, that lack WikiProject Astronomy banners; so if the first link doesn't work, try this one (modifying for other categories as needed). (The Astronomy in China category was an arbitrary choice; I was just trying to find a category that had some articles lacking an Astronomy banner, so I could distinguish between 0 results being an execution failure, or a case where all articles in the category already had a banner). Plantdrew (talk) 19:47, 11 March 2023 (UTC)
Better communicating uncertainty in unconfirmed sports transactions
Hi folks, I'm looking for thoughts on how to better handle sports transactions, e.g. when a player is signed, traded, or transferred to another team, in the interim period between media scoops and an official confirmation. The most recent one to become an issue is D. J. Moore (wide receiver), which has been subjected to a lot of IP and new editors trying to "correct" our article.
Our current guidance is to tag the article with something like {{current sports transaction}} and to include specific in-text attributions to news sources are reporting about the transaction.
Unfortunately, that leads to a lot of confusion from readers who come to Wikipedia and see that we still have the "wrong" team listed. That's especially true for mobile readers, who probably skip over what becomes a very truncated template (web) or don't see it at all (Android app, and I presume iOS too).
What's a better way to communicate this uncertainty to our readers? I have a few ideas, but I've no clue if they're workable.
An infobox parameter that allows us to list a current team and an unconfirmed next team
Standard language that can be easily inserted article leads and bodies to communicate the reported trade ahead of official confirmation (or the trade falling apart)
A standard {{efn}} note that 1) describes the current reports and 2) notes that the transaction has not been officially confirmed
Have the sports bio page semi-protected, so IPs & Mobile editors can't add in transaction info, before it's confirmed. GoodDay (talk) 23:52, 10 March 2023 (UTC)
I believe the consensus at both the ice hockey and baseball WikiProjects is to only report officially confirmed transactions. isaacl (talk) 23:54, 10 March 2023 (UTC)
The general practice is to follow the WP:RSBREAKING guideline: It is better to wait a day or two after an event before adding details to the encyclopedia, than to help spread potentially false rumors. This gives journalists time to collect more information and verify claims, and for investigative authorities to make official announcements. If the deal is consummated, the major sports leagues invariably make a statement on the record at some point, typically in a few days. Stay the course, and mention nothing until then.—Bagumba (talk) 00:07, 11 March 2023 (UTC)
I feel like that part of the guideline is meant more for hard news and political journalism than it is sports reporting, which nearly often—though not always!—comes to fruition. Particularly with the nod to "investigative authorities". Ed[talk][majestic titan]04:23, 11 March 2023 (UTC)
Disagree with 1) "infobox parameter". This seems like an odd workaround. Our current standard procedure works fine with me, though this may lead to a lot of reverts or require page protection. On a side note, can someone protect D. J. Moore (wide receiver)? Natg 19 (talk) 00:18, 11 March 2023 (UTC)
None of the above messages address the fundamental issue of a reader seeing a reported trade, then coming to us and have those reports be completely unreflected or mentioned on the page. Even if we choose to not include transactions until they're completed, we're relying on {{current sports transaction}} to communicate that we're aware of the news reports... but many will have missed that banner, which is as I said truncated on mobile web and hidden on the apps. This is the sort of thing that lowers people's trust in us, and therefore a problem. So, other than the banner, what is better way of communicating to readers that a player is involved in a potential transaction? Ed[talk][majestic titan]04:20, 11 March 2023 (UTC)
We could just display this particular banner in all environments. Or prohibit editing from broken clients that do not display important banners. —Kusma (talk) 06:27, 11 March 2023 (UTC)
I believe every "currrent" event-type tag has in its documentation some note that Wikipedia:General disclaimer already says that article contents may not be accurate, and that the tag isn't all that necessary. As for the mobile web and apps, the problem applies to all tags, not just for this sports one being truncated or missing. That's a higher-level matter to discuss. —Bagumba (talk) 07:57, 11 March 2023 (UTC)
Until a transfer/trade is confirmed, it is WP:SPECULATION on whether it will happen, and so Wikipedia should not be reporting it as a fact. We absolutely shouldn't be adding expected teams to players' infobox stats or anything like that. Wait the day or two until it's confirmed, and then and only then should it be added to the article, once reliable sources confirm it's a done deal. Any speculative editing is unacceptable, as we shouldn't be presenting transfer speculaton as fact. Joseph2302 (talk) 07:39, 11 March 2023 (UTC)
I think it's wrong to frame this in terms of "speculation." The problem is that many sports transactions are reported as facts by reporters and news organizations before the transactions have been officially finalized. Drive-by editors see a headline at ESPN and rush to "correct" an article without understanding the subtleties. People who have been editing sports articles for a while know that in a day or two, there should be a formal press release from the team(s) officially confirming the transaction. But I think inexperienced editors genuinely don't understand what the problem is. Zagalejo (talk) 17:39, 11 March 2023 (UTC)
Someone like Adrian Wojnarowski is usually good about starting with "Sources say..." But secondhand reports, even from mainstream news services, don't always make it clear that the story is based on unnamed sources. (Local TV sports anchors, especially, will talk about these reports as if they were done deals.) The situations that Ed talks about are legitimately difficult for Wikipedia to deal with. It's easy to revert a baseless rumor, but the stories Ed has in mind have more journalistic heft to them. I was pulling out my hair over this stuff ten years ago, before I finally cut back on Wikipedia. Zagalejo (talk) 21:40, 11 March 2023 (UTC)
+1 to Zagalejo. Joseph2302, you do realize that WP:SPECULATION literally says "Individual scheduled or expected future events should be included only if the event is notable and almost certain to take place"? Transactions reported by large news outlets fall through only extremely rarely, so "almost certain" seems about right. Ed[talk][majestic titan]05:19, 12 March 2023 (UTC)
In association football, lots of transfers that are reprting on as "almost done deals" don't actually happen in the end, so reporting them like facts is just wrong. Joseph2302 (talk) 07:51, 12 March 2023 (UTC)
I'm not sure who you're responding to. To be fair, the reporting in these situations isn't actually "crappy"; journalists like Adrian Wojnarowski are very rarely wrong when they announce their scoops. We don't know who their sources are, but the reporters know very well what's going on among the teams, players, and agents. There are still formalities to take care of after the scoop has been announced, which is why Wikipedia should wait for official announcements, but when established reporters announce that a transaction has been agreed to, those reporters are rarely wrong. I obsessed over this stuff in NBA articles for years; probably 99% of the time, the transactions were soon confirmed as reported. I'm not trying to propose a specific course of action. Again, I agree that we should wait for official announcements. But I wanted to chime in, because I'm not sure people fully grasp why these situations get so frustrating on Wikipedia. Dismissing the transactions as "speculation" doesn't get to the heart of the matter, because the reporting is far more than speculation. Zagalejo (talk) 04:52, 12 March 2023 (UTC)
Though I do agree that we should not be changing teams and that our status quo generally "works", I can see the point that newer editors are confused or wondering why trades are not "reported". There have been some high profile deals that have fallen through (e.g. Carlos Correa this year), but on the whole, 95% of reported trades, FA signings, etc, pan out, and altogether Wikipedia is not a newspaper or meant to report breaking news, many new editors or drive by users are confused. It definitely took me a little while to understand our policies to know that breaking news shouldn't be reported without official team sources. Natg 19 (talk) 08:13, 12 March 2023 (UTC)
This idea confuses me, if there hasn't been an official announcement, then why would anyone think there was anything to report? Again, I don't care who the reporter is or what inside sources they claim to have or how often they have been right, until there is an official announcement I would consider this just rumors, not news. And I fail to see why anyone, no matter how new they are to Wikipedia, would think that rumors were appropriate for an encyclopedia.--User:Khajidha (talk) (contributions) 10:59, 12 March 2023 (UTC)
You have to look at the reporting that exists for these transactions. That will help us understand the perspective of a drive-by user. For example, most people reading this would assume the Moore transaction was a done deal, right? Sports reporters have deep connections inside the teams' they cover. The transactions they report are not just educated guesses; the decisions have been agreed to among the parties involved. We're essentially waiting for formalities. Yes, once in a while, the transactions do fall apart, but the reporters aren't pulling these stories out of thin air. Zagalejo (talk) 15:35, 12 March 2023 (UTC)
That source refers to an "agreement", which is a euphemism of sorts for "not completed yet". We can agree to purchase a house, but have the loan fall through or otherwise back out. We can agree to get married, and break off the engagement or say "no" at the altar. —Bagumba (talk) 07:19, 13 March 2023 (UTC)
Yes, I've seen articles use the word "agreement" in that matter, but most readers won't pick up on those distinctions. Again, I support the decision to wait until there is an official announcement. I just think some people don't realize how these situations look to people outside the Wiki bubble. Maybe Jayron32 is right that people just won't listen anyway... But I'd appreciate any outside-the-box ideas, because sports transactions are a miserable time sink for the editors who care. Zagalejo (talk) 16:45, 13 March 2023 (UTC)
I get the opposite impression of "trust", watching twitter. I've seen countless threads about somebody dying or an athlete/coach changing teams where people argue that it isn't official yet because Wikipedia doesn't say it yet. Schazjmd(talk)18:09, 12 March 2023 (UTC)
I actually agree that we can't write something as "done" when it's not, but I kicked off this section because our current process of completely omitting this info is clearly confusing people. What's stopping us from communicating that uncertainty? "ESPN reported on XXXX that YYYY has been traded to ZZZZ for AAAA. As of BBBB, neither team has confirmed the transaction." (edit: in practice, it would look something like this line) Ed[talk][majestic titan]18:27, 12 March 2023 (UTC)edited at 19:11, 12 March 2023 (UTC)
I'm fine with this, but there are some editors who don't like this. However, it still doesn't stop the issue of a lot of people who change infoboxes and the lede to reflect team changes. Natg 19 (talk) 20:02, 12 March 2023 (UTC)
Unfortunately, I am doubtful that adding "YYYY has been reportedly been traded to ZZZZ (or YYYY has reportedly signed with ZZZZ) will reduce drive-by editing. The only "solution" is page protection, and clear warnings to regular editors to not change infoboxes and lede sentences. Natg 19 (talk) 21:33, 13 March 2023 (UTC)
And what about the readers who don't edit?: That would be {{current sports transaction}}, subject to the limitations that have been already discussed above. Frankly, readers wanting breaking news should look to Twitter, not Wikipedia. That said, WP is pretty quick to update once an on-the-record statement is made.—Bagumba (talk) 07:08, 13 March 2023 (UTC)
I think a lot of people do use Twitter for breaking news, but then rush to WP to make a change on the player's page. Natg 19 (talk) 21:33, 13 March 2023 (UTC)
Shrug whether or not it is bad reporting, this is extremely common in sports. Even the NFL's official website does this: [1]Natg 19 (talk) 20:02, 12 March 2023 (UTC)
"Official" team and league websites have both news based on anoymous sources, as well as official press releases. The key is knowing how to tell the difference. —Bagumba (talk) 07:25, 13 March 2023 (UTC)
The most important thing to remember here folks is that rules are meaningless among people who haven't read them. The OP's main problem is that these kind of drive-by edits are done by people who aren't going to read any guidance on this matter, regardless of how it is given to them. The only way to fix this problem is to be diligent in cleaning up after the fact. There is no amount of rule-making, no notice that can be written, no giant flashing red sign that you can put in front of people, literally NOTHING that will have any effect on this phenomena. Efforts would be better spent just fixing these problems when they happen than on spending energy trying to stop them. --Jayron3212:14, 13 March 2023 (UTC)
Feature / Project idea: Use <time> tag wherever dates or times exist in Wikipedia?
My idea is this: every date and date/time in Wikipedia articles should be surrounded by the HTML <time></time> tags with the appropriate datetime= attribute. I think this would require some sort of batch edit process to update existing articles and changes to templates and/or editors to recognize date and time patterns to put the tags into the articles when edits are saved but I am a Wikipedia / Mediawiki neophyte so maybe it's not that simple? In any case the idea is to provide machine-readable date/time information in a standard format. That could then be used for future tools that do calendar-based things ("on this day", timelines, or things I haven't thought of) TimHare (talk) 05:43, 10 March 2023 (UTC)
I'm not sure how this would work out with things that are not a specific period in time (as MDN describes the <time> element. There are bound to be a ton of edge cases that would need to be figured out:
It extends from the earliest known use of stone tools by hominins, c. 3.3 million years ago, to the end of the Pleistocene, c. 11,650 calBP
Cleopatra VII Philopator (Koinē Greek: Κλεοπάτρα Φιλοπάτωρ, "Cleopatra the father-beloved"; 70/69 BC – 10 August 30 BC) was Queen of the Ptolemaic Kingdom of Egypt from 51 to 30 BC, and its last active ruler.
Ambrose Gwinnett Bierce (June 24, 1842 – after December 26, 1913)...
I'm not sure there is a "correct" date time string for all these, never mind one that could be identified automatically. I also note that as dates, the datetimes for <time> are supposed to be A precise date in the Gregorian calendar which will be an... interesting challenge in terms of mapping all the dates, especially when there's any fuzziness around them. Ljleppan (talk) 06:35, 10 March 2023 (UTC)
Good points. It could not be used for estimated durations. We could, however apply the <time> tag to strings like "at 11:59 PM Eastern Standard Time on August 10, 1969" or "December 7, 1941" and maybe develop a heuristic to avoid the other cases. Sticking to the Gregorian calendar might be as simple as avoiding all dates prior to its inception - I say "might" because all things calendar-related can be more complex when we look closer. TimHare (talk) 15:33, 10 March 2023 (UTC)
If I propose this as a project, I will change it to "use <time> tag wherever a specific Gregorian date or date/time exists" TimHare (talk) 06:05, 14 March 2023 (UTC)
Wikimedia maps Update
Hello,
I think Wikimedia maps (commonly used in Template:Maplink) could benefit from the inclusion of names for some large/notable parks, rivers, etc... Without context, a lot of these areas blur into one color when in fact, there are multiple different parts to them (especially the green color for parks and protected areas). It might look a little like how Google Maps makes their map but with far less named areas and no pinpoints. Thank you for your time and have a great day! DiscoA340 (talk) 18:08, 15 March 2023 (UTC)
In the news criteria
In the news has an issue in that the selection process for articles is almost entirely on the basis of editors personal preferences and biases - current editors, in the case of non-recurring stories, and past editors in the case of recurring stories.
This has resulted in a systematic bias issue that prioritizes topics of relevance and concern to Wikipedia editors, rather than to the world in general.
This can be prominently seen at ITNR. Defined very broadly, 73 of the listed events cover international topics. Of the rest:
23 European (11 British, 1 German, 1 Spanish, 8 generally; 1 per 32 million people)
16 North American (13 American, 2 Canadian, 1 generally; 1 per 36 million people)
6 Asian (3 Indian, 2 Japanese, 1 generally; 1 per 760 million people)
5 Oceanic (4 Australia, 1 New Zealand; 1 per 9 million people)
2 South American (2 generally; 1 per 211 million people)
1 African (1 generally; 1 per 1216 million people)
It also violates the core policies of WP:OR and WP:NPOV. By deciding which events are significant enough to publish ITN through our personal criteria we are engaging in original research and placing undue weight on those events that are included based solely on those personal preferences.
To fix this we need a more objective criteria for inclusion at ITN, and my initial proposal is for one of the following:
The event has received significant original (not syndicated or from a wire service) coverage in a wide variety of reliable international news sources.
The event has received significant original (not syndicated or from a wire service) coverage in a majority of 21 selected reliable international news sources.
The second proposal would involve selecting those sources; I expect these would include Agence France-Presse, Associated Press, BBC, Deutsche Welle, The New York Times, Reuters, and The Times of India, while the others would be the subject of considerable discussion. It has the advantage of being more objective and better suited to reducing systematic bias by forcing editors to consider coverage in a wider range of sources, particularly if the list includes non-English language sources, but it may also be harder to get an initial consensus for due to difficulty in creating the list of sources.
I think is a very minor concern. Most of what doesn't get through ITN or gets through ITN is mostly determined by the quality of the article update. Which is as it should be, the purpose of the main page is to tell people about Wikipedia articles that are of a high quality. Everything else is irrelevant. ITN is not "Tell the world what Wikipedia thinks is important", that's a terrible way to think about it. Instead, it is "What is something recent that Wikipedia readers may appreciate reading a high quality Wikipedia article about" with an emphasis on the high quality. I'm far less concerned with playing cultural gatekeeper and making sure that the main page is not sullied by things from certain geographic areas, and I'm pretty much only concerned that Wikipedia articles are improved, and that we let readers know what those articles are. --Jayron3215:55, 14 February 2023 (UTC)
ITN is not "Tell the world what Wikipedia thinks is important" – The problem is that this is exactly what it's become. Most people !voting on blurbs don't even mention article quality. It's just the personal opinions of editors as to whether they feel it's an important news story or not. If you want ITN to be centered around article quality, then it's going to need considerable reform. Thebiguglyalien (talk) 20:21, 14 February 2023 (UTC)
+1, One article that I worked up to GA borderline passed ITN despite it winning the video game equivalent of an Oscar, with the event attracting tens of millions of viewers. If it was based upon article quality, it would have been accepted much more readily, but the main opposition was centered around video games being "not important enough for ITN". The Night Watch(talk)15:37, 15 February 2023 (UTC)
“It's just the personal opinions of editors as to whether they feel it's an important news story or not.”
That’s typically how consensus works, and from skimming this discussion it seems to be a lot of the usual suspects on ITN who disagree with the mere idea of consensus because it usually goes against their opinions. The Kip (talk) 16:11, 15 February 2023 (UTC)
That's not how it is supposed to work at Wikipedia. At Wikipedia, we ignore commentary that is not based in a sound footing of Wikipedia policy and guidance. If people are commenting with rationales that don't align with established principles at Wikipedia, we're supposed to ignore those comments when assessing consensus. From WP:CONSENSUS, and I quote, "Consensus is ascertained by the quality of the arguments given on the various sides of an issue, as viewed through the lens of Wikipedia policy." (bold mine). ITN has written standards; if people ignore those written standards in their commentary, then their opinion should not be taken into account. So know, we should not let consensus discussions be overwhelmed by people whose opinions run counter to long-established Wikipedia principles. --Jayron3213:36, 16 February 2023 (UTC)
The problem of late is that we have stories that bring a lot of "drive by" !votes (editors that are maybe at ITN the first or second time in their history) and vote, such as in death blurbs "very important person" without further explanation. When that happens en masse, the ITN admins in charge of posting seem to not consider the "views through a WP policy lens" and are more or less !vote counting. (The posting of Betty White's death and Carrie Fisher's were examples of that). That also applies to general stories too. Not so much that the recent one on LeBron breaking the scoring record being flooded with such but there were definitely drive-bys that were basically "Support - a key record was broken" without any further explanation. And that was briefly posted before pulled as the posting admin seems to take those !votes into account when they should have been discounted.
That's getting away from this point on importance based on RSes, but it is also tied to it, since I've seen frequent calls on stories that get !votes like "All over front pages" or "leading story on (key sties)", which may indicate importance to news media, but not to an encyclopedia per NOTNEWS (eg the whole mess with the US House's Speaker election). I do believe there are some objective standards that incorporate frequency and type of coverage by RSes to evaluate relative importance to the world body of sources, but it should definitely not be the sole driver (in addition to quality) for consideration. Masem (t) 13:58, 16 February 2023 (UTC)
"Not a news ticker" means it isn't our job to tell people what is important, it is our job to highlight quality content only. It doesn't really matter much whether or not we think something should or shouldn't be "important", we should be showing people quality article content, period. Any decision making that attempts to assess worthiness is making ITN a news ticker. Instead, it should be a tool to direct people to high quality articles about current events. --Jayron3214:23, 16 February 2023 (UTC)
Disagree, as the more blindly include topics based on quality followed by news prevalence without any further consider, the more we mirror what news media believes important, and we become a news ticker to them. It should be obvious that not every story that gets huge promotion on the news makes for a reasonable topic (first and foremost we rely on enduring coverage and not bursts, which us what the typical story in the news is), so we clearly need a filter to eliminate burst-news coverage. Then we need to add the issue if systematic bias that even with a good selection of worldwide press, we will still find ourselves favoring Western topic, and hence the need to increase the visibility of events in less covered places that our significant (eg the Turkey/Syria quake has been getting less coverage than all the 2024 campaign hijinks or even the Ohio rr srory). That's why blindly following news media makes us more like a news ticker. Masem (t) 14:49, 16 February 2023 (UTC)
You can disagree, but your rant about bias is irrelevant here. You fix bias by making articles about underrepresented topics better. The idea that we should eliminate bias by making Wikipedia articles on Western topics worse seems like a bad idea. --Jayron3215:10, 16 February 2023 (UTC)
Far be it from me, a clueless non-admin, to make a judgment about how admin work usually goes at ITN, but I've found that Masem's observations are more or less correct, that the significance of a story is based on who shows up to vote and in what quantity. So any story that provokes an emotional reaction from editors will likely sway the balance because we weigh all of these "significant/not significant" votes equally. This comes down to us not having a clear objective criterion for significance on WP:ITNCRIT, instead going based off of a subjective consensus. I expect that when The Boat Race is nominated for 2023, despite the article being a FA year-after-year (the quality that Jayron32 is looking for), there will be a visceral outpouring of opposition, more than we'd see for any other sporting event, based on the fervor that was accumulated during the discussion calling for its removal from WP:ITN/R.
In regards to bias, the difficulty with nominating underrepresented topics is that we are still stuck in a precedent-based mindset of "well, we never posted this before, we shouldn't post it now" or "it's not disastrous enough" or "I've never heard of it". That's the sort of acts that perpetuate systemic bias. Indeed, it's present across all of ITN. These sorts of arguments would never fly in an AfD, so it's a real shame that they can be effectively weaponized in this setting. Moreover, newer contributors who are nominating something for ITN for the first time tend to be driven off by petulant line-towing semi-regulars who scream at them "Of course this isn't notable, SNOW close, why the fuck did you nominate this?" which is most discouraging. It's a big part of why people complain about ITN's atmosphere.
I don't agree with Masem on a lot of things. I think during the start of the 2022 Russo-Ukrainian War, he took an incredibly skeptical line in challenging just about everything the reliable sources stated about events such as massacres or missile strikes. But I think he's right on this one. ⛵ WaltClipper-(talk)13:40, 17 February 2023 (UTC)
I disagree with you that our job is to only highlight quality content, because even if that might be our mission, that's not what the criteria states and that's not consistent with our established consensus which calls for identifying a story's significance. I agree with you on making articles on underrepresented topics better, but if our goal is to get it onto ITN and onto the Main Page, I don't believe that improving it to even an FA will be enough to get it posted to ITN. ⛵ WaltClipper-(talk)13:53, 17 February 2023 (UTC)
Let me also hasten to add that I want our job to be highlighting only quality content, just like you do, but I don't think we'll get everybody to drink from that cup. --⛵ WaltClipper-(talk)13:57, 17 February 2023 (UTC)
The larger problem is that I think we (WP overall) have lost sight of why NOTNEWS exist. Newspapers serve one function - to report information as fast and broadly as possible - while an encyclopedia serves a different one - to summarize a topic for posterity. If we put those functions into a Venn Diagram, there would definitely be an overlap in that there are topics that started as news that become clearly important for enduring knowledge. And with covering current events, we generally had been pretty good about being predictive that an event is going to have the long tail that makes for a good encyclopedic topic. But we getting a lot of new current event articles that may be well-backed from newspapers as part of their function, but fail to prove out as long-term events of significance. In other words, we have editors trying to write like a newspaper and thus we get a lot of noise at ITN, particularly from overrepresented areas. I think we do need to tweak how editors approach NOTNEWS and NEVENT, understanding that just a mere burst of coverage is not necessarily quality sourcing for an encyclopedic article. Whereas news stories from underrepresented areas that fit into encyclopedic content, there may be the long-tail of coverage but the number of works covering it will be low, and that's why I think any type of "counting" of story coverage is a problem that feeds, not fights, systematic bias.
I'm willing to hear about any system to help improve objectivity and reduce the drive-by voting, and there may be something in source counting, but I don't see an obvious solution that still creates a systematic bias problem. Masem (t) 15:46, 17 February 2023 (UTC)
The change proposed runs counter to the stated goal, but that doesn't matter because the premise here is ridiculous. ITN/C aggressively combats bias and engages in affirmative action by applying lower standards to quality and significance for under-represented regions. GreatCaesarsGhost16:44, 14 February 2023 (UTC)
I'd support trying either of these proposals, or pretty much anything that will make ITNC more objective and less subjective. It's always felt bizarre to me that widespread media coverage does not count as "significance" at ITNC, and instead, "significance" is just the sum of subjective opinions of participating editors. As between these two proposals, I'm split; I can see both the upsides and downsides of specifying a number. Other reform ideas: eliminate blurbs altogether and just post the links to articles (so everything would look like ongoing or RD, and we could use the extra space for more pictures, so we could have, e.g. three picture slots and no blurbs); replace ITN with a "most-edited articles" or "most-viewed articles" list; eliminate ITN altogether and rearrange the main page with the remaining elements (FA/FL/FP, DYK, OTD). Levivich (talk) 20:42, 14 February 2023 (UTC)
Wikipedia is not a newspaper, and ITN is not a news ticker. We're there on the front page to feature quality articles about recent events that happen to be in the news, not to feature news stories that have quality articles. That's why significance and quality are the key determinants in ITNC discussions. Following the news does not give us the broad range of topics that we want ITN on the Main Page to be. I do think that the "significance" factor has been watered down and/or weakened which started with the whole issue of mass shootings in the US, and there's lots of bitter feelings on that that which has made objective evaluation of significant far more difficult to come to on other topics. Masem (t) 03:45, 15 February 2023 (UTC)
I have for a long time thought that this section has departed from its encyclopedic purpose, which is to highlight Wikipedia articles about subjects that are in the news, rather than specially created articles about the news events themselves, which should usually be deleted per WP:NOTNEWS rather than highlighted on our front page. People are given the impression that this is a news site rather than an encyclopedia. Phil Bridger (talk) 20:57, 14 February 2023 (UTC)
ITN was established on the basis of how fast the community worked to produce a quality article on 9/11, and type of effort been repeated multiple times since. Now part of the problem is that NOTNEWS and NEVENT ofter go disregarded because nearly every event is claimed to be notable because of a burst of coverage (notable requires more enduring coverage). We really need to be more enforcing on NOTNEWS which should with some of the topic noise at OTN (currently exemplified by the UFO shhotdowns) Masem (t) 21:48, 14 February 2023 (UTC)
I know I'm not with general consensus here, but we should really only be covering topics that are covered by secondary sources. Most articles about news events are primary sources. Of course newspapers sometime publish secondary sources, such as reviews of a situation, but the general opinion here seems to be that we should accept lots of primary sources, with some geographical distrubution, as the basis for an article. Phil Bridger (talk) 19:30, 16 February 2023 (UTC)
Good idea. I want ITN to reflect what's actually in the news, not what some of us wish got attention. Maybe 21 sources is too high an estimate for a practical mainstream core, but the seven offered above are certainly a good start. InedibleHulk (talk) 21:27, 14 February 2023 (UTC)
Will not oppose. Well, we could try one of these, but I don't know what's going to happen. Many of the folks on ITN think it'll just heavily prioritize celebrity news or gossip. But who knows, it might work. I don't think we should do away with our current ITN/R items wholesale based on this new proposal. Just continue to nominate those for addition or removal on a case-by-case basis. ⛵ WaltClipper-(talk)21:55, 14 February 2023 (UTC)
Support something along these lines. I would add a stipulation that topics be in the "world" or "national" news sections of each publication to prevent NOTNEWS pop culture/sports/etc. creeping in. JoelleJay (talk) 23:07, 14 February 2023 (UTC)
I think I agree with that, but the wording is a little ambiguous; I assume you mean any event that is classified as either international, related to a country, or related to a continent? BilledMammal (talk) 03:49, 15 February 2023 (UTC)
Yeah, news articles that are classified under those sections of the newspaper. Like The Times of India homepage has a ticker menu at the top with "India" and "World", and articles map their directory path back to the parent category (e.g. here where it says NEWS / WORLD NEWS / CHINA NEWS / [article title]). JoelleJay (talk) 20:24, 15 February 2023 (UTC)
If we're going to be using some set number of sources, editors would have to come to a consensus on which sections of each paper qualify for assessing significance. JoelleJay (talk) 01:24, 22 February 2023 (UTC)
I don't really expect this model to work. In addition to the problem of choosing which wire services/newspapers count (nothing from Pakistan, home to 100 million English speakers? nothing from Nigeria, the third-largest population of native English speakers on Earth?), I think editors will be unhappy with the results. We'll want the culture section when the Nobel Prize in Literature is announced but not when the next Harry Potter-equivalent is released. there is a significant fraction of editors who want Wikipedia to feel "serious", and relying on external sources that don't have a bias towards "serious" will not achieve their goals. WhatamIdoing (talk) 02:56, 22 February 2023 (UTC)
I would suggest that it is better to collect evidence without changing the process first. It should be possible to, given an ITNC nomination, how many times it appears in the selected list of sources. Or alternatively, figure out a means to determine the top 5 stories of each work each day and count repetition across sources. With, say, a good two weeks or a month of data, it would be far easier to understand the impacts on ITNC without actually changing it. My gut remains that this type of a approach will overwhelm ITNC with Western and English topic per WP:BIAS, but it would be best to prove that wrong before making any change. --Masem (t) 03:38, 15 February 2023 (UTC)
I don't think such a test will work as I don't know what sources the community will chose and the result of this change will depend heavily on that. My overall position is that something has to be done; we can try this, and if after trying it for a few months we discover it doesn't work we can try something else. BilledMammal (talk) 03:49, 15 February 2023 (UTC)
Then the first thing is to determine the "jury" of news sources, first and foremost, before even applying that. This has far too many working parts to implement without evidence and other testing beforehand, and could fundamentally break ITN if its not thought out well. Masem (t) 13:17, 15 February 2023 (UTC)
However, there are times when key news happens first in foreign language sources (particularly SE Asia and South America) that using the known RSes from those regions are fine as well. That's another flaw in this system is that not all news breaks first in English. Masem (t) 13:27, 15 February 2023 (UTC)
This is a terrible idea. Firstly, because ITN is not, and should not be, purely a news ticker. Secondly because far from avoiding systemic bias, having a special list of 'blessed' sources that determine what is worthy of ITN will deeply entrench that bias. I also don't think we should be reliant on English-language sources, which will also drastically skew our coverage. If a Hindi, or Chinese, or Brazilian Portuguese source is the main source for a significant story, we should reflect that. We had great trouble in this respect with trying to build consensus on the Nagorno-Karabakh blockade story - very few of the sources were even in the Latin alphabet, but that did not mean we should not attempt good, unbiased coverage of it. GenevieveDEon (talk) 13:14, 15 February 2023 (UTC)
I think the last thing we should do is have any sort of minimum source requirement. That really isn't the issue we are currently facing. Most are concerned about not enough items being posted. I think a minimum source threshold will only exacerbate preexisting balance issues at ITN, as quite frankly most media coverage nowadays is about what generates more clicks. For example, many nations in Africa suffer from power instability and are subject to military coups, but usually this is just a passing news story for many Western publications. However, a transfer of power (especially one done by force) is clearly more noteworthy and impactful than some of the events we have and will post, such as the Ohio train derailment, which may not get posted but is on the cusp as of right now. DarkSide830 (talk) 18:45, 15 February 2023 (UTC)
On the contrary, it's probably the most important place to remind editors of that policy. "In the news" exists to showcase Wikipedia articles about topics that are in the news, not to provide a news service. Phil Bridger (talk) 18:02, 19 February 2023 (UTC)
I think the difficulty is that "In the news" doesn't make it obvious what the missing word is ("[Wikipedia articles] in the news"), and so most readers see the section as just the news section of Wikipedia, rather than "Articles in the news." :3 F4U (they/it) 19:01, 23 March 2023 (UTC)
One of the reforms we should do is to get rid of the "showcase" purpose, as pretty much no article about current events is worth showcasing; there's not enough time to bring them up to GA or FA quality. We flatter ourselves with "showcase". The purpose of ITN is (and should be) to help readers find articles about topics that are in the news. Levivich (talk) 15:38, 20 February 2023 (UTC)
Exactly. ITNQUALITY isn't good enough to be worth "showcasing", thus ITN doesn't fulfill the purpose of showcasing quality content, thus we should remove this purported purpose (not to be confused with a purported porpoise) from the list of purposes (not to be confused with the list of porpoises). Levivich (talk) 16:08, 20 February 2023 (UTC)
Unless you are proposing the same pruning of article quality that DYK has, this is unworkable. We recognize many articles about news topics that do have encyclopedic purpose can get up to a perceived quality within a day or so, but the process to get through GA or FA is far longer than that, so we'll accept something that we can tell is likely to be of high quality in the short term. Masem (t) 16:10, 20 February 2023 (UTC)
The Main Page of WP is to showcase quality articles in their relevant sections. We can't remove that. We are not expecting GA/FA quality (though clearly will accept them), just as DYK doesn't expect those. The purpose of ITN is to help readers find encyclopedic articles about topics in the news, which means that not all topics in the news will necessarily be featured. Wikinews is better suited for the latter function where there is no encyclopedic requirement. Masem (t) 15:53, 20 February 2023 (UTC)
Sometimes I think if I said "we should turn the light on" you would respond with a paragraph explaining that it is dark here. Levivich (talk) 16:10, 20 February 2023 (UTC)
You keep introducing a meaning of ITN that is not based on the actual meaning of ITN, instead wanting to turn it into a news ticker. Masem (t) 16:13, 20 February 2023 (UTC)
Hahaha yes that's called "change". I keep suggesting change, you keep explaining how things are as if I don't know. But seriously: please stop, it's annoying af. Levivich (talk) 16:16, 20 February 2023 (UTC)
Things are as they are for a reason. The changes you want to make would turn Wikipedia into something it is not meant to be. Blueboar (talk) 16:22, 20 February 2023 (UTC)
Oh man, you're not seriously pulling out the "things are as they are for a reason" line? Trust me: that line is a weak argument wherever it's deployed; it's akin to rhetorical surrender, because "a reason" is not necessarily a good reason. Sometimes things are as they are for a bad reason, like in this case: editor vanity, wanting to "showcase" average work. Levivich (talk) 16:28, 20 February 2023 (UTC)
I do think we need some change to ITN, because looking through the candidates often makes me often wonder what does "significance" mean? Some sport news are considered significant, some are not. Some mass shootings are significant, some are not. Some deaths are notable, some are not. We need some more clearly defined criteria, otherwise it just becomes heavily subjective. My question about the first proposal is what does "wide variety" mean? Does that mean 5 sources, 10, 15? Natg 19 (talk) 18:04, 21 February 2023 (UTC)
You're pretty much on the mark. Although people have their own individual standards for what constitutes a significant story, in the end, overall significance is based on a headcount. That's the dirty word that people around here don't like to say, but I feel it's true in the case of ITN. A !vote that says "it's notable" has just as much weight as one that says "it's not notable". ⛵ WaltClipper-(talk)17:10, 22 February 2023 (UTC)
ITNC is the only page on Wikipedia I can think of where closers seem to never weigh votes; it's pure headcount. Even a vote that says "only significant in one country" gets weighed, despite being against the instructions on the very page. IMO, pretty much all the problems with ITN would be fixed if closers applied our WP:PAGs and weighed votes when closing discussions. If they explicitly stated the kinds of votes that they weren't considering (the kind that are contra PAGs, contra WP:ITN), editors would eventually stop making those kind of votes, and the whole enterprise would improve. But, alas, easier said than done. I'd do it myself if it didn't require running for RFA. Levivich (talk) 17:16, 22 February 2023 (UTC)
Oh I disagree there; there are many examples of successful RFA candidates who ran specifically to work in just one area (e.g. SPI, DYK, CCI), and I think "main page admin" or "ITN admin" is a perfectly valid reason for someone to run for RFA. I'd never encourage anyone to run for RFA because I think the process is awful, but Walt if you're inclined to do so, I'd say go for it. Unlike me, I think you'd actually pass and ITN could use more admins. (ERRORS, too.) Levivich (talk) 17:43, 22 February 2023 (UTC)
I'd only run if three people nominate me in good faith, one of them preferably being another admin. I've long bemoaned the gauntlet, and I also have a ton of skeletons in my closet that would be dredged up from my early editing days. ⛵ WaltClipper-(talk)17:48, 22 February 2023 (UTC)
As a further note, the only "guidelines" ITN has are the article has to be of sufficient quality (WP:ITNQUALITY), have "updated content" and the significance section, which itself states ultimately, there are no rules or guidance beyond two: (1)The event can be described as "current", that is the event is appearing currently in news sources, and/or the event itself occurred within the time frame of ITN. (2)There is consensus to post the event. There is a lot of explainer text following these two "rules", but unfortunately, this process is way too subjective to determine what is significant, so we need something to clarify "significance". Natg 19 (talk) 18:39, 22 February 2023 (UTC)
We have worked to address the issue of deaths, in that as long as the person has an article, it qualifies for the RD line once quality has been assessed. That might still leave debates over whether the person should get a blurb about their death, but that's not core to this concern. ITNR is also there to make sure a wide variety of recurring world events get covered (again, barring quality issues). One thing that gets us, and why using any type of source based counting causes problems, is that per NOTNEWS, we shouldn't be covering topics that have a burst if coverage but no long tail as its own story. We frequently have the heaviest discussions on such "burst" news coverage, and we need to adherents NOTNEWS better by focusing on topics with long coverage of events. --Masem (t) 18:47, 22 February 2023 (UTC)
I disagree with ensuring our news story choices are proportional to the population of the continents, and also whether that is evidence of systemic bias within Wikipedia; our stories are based upon the major news sources in English, if there is bias it's the sources. But regardless of what we do someone will criticise us (whilst not being prepare to volunteer), so let the editors choose based on thier whim, becuase we often have a far wider range of stories than elsewhere. Wakelamp d[@-@]b (talk) 14:50, 20 March 2023 (UTC)
Alternative - quality only
The issue is that our current assessment of significance is subjective. The alternative to using a less subjective method of determining significance is to remove the significance requirement entirely; change "In the news" to "Good articles on recent events" and have the requirements be that the article meets the good article criteria (with some leeway for stability given the event was recent) and that the event covered is more recent than the oldest currently listed.
This does open the possibility of abuse by paid editors to increase the profile of their product so articles likely to be of interest to paid editors would also need to be excluded. BilledMammal (talk) 09:05, 19 February 2023 (UTC)
A less stringent requirement than good article class would be B class, but there have been past objections to using the class system for ITN on the grounds that it is too subjective below good article class. BilledMammal (talk) 09:09, 19 February 2023 (UTC)
I'll go over the reasons why this won't work. First, it'll make the systemic bias issue worse than it is now, since topics from underrepresented regions will be harmed by the lack of extensive sources and the lack of people working on those articles to bring them up to the lofty requirements of the B or GA criteria. Second, having the lack of a significance standard will hyper-prioritize minuscule developments in those areas of Wikipedia which are well-developed, such as American politics, sports, celebrity news, business news, gaming, etc., and while that might not sound so bad in theory, you will have a very hard time getting ITN users to buy into that. Finally, it runs contrary to the goal of "[emphasizing] Wikipedia as a dynamic resource", since the increased quality standards would result in a stagnant article base. ⛵ WaltClipper-(talk)15:58, 19 February 2023 (UTC)
I'll second Walt Clip's comments here. And, in the end, do we really have such a colossal issue here that we need to make such a massive change that a lot of editors aren't going to buy into. DarkSide830 (talk) 16:19, 19 February 2023 (UTC)
I don't think we'll have enough GAs to fill the pipeline. However, drop that, and it could just be "new articles" or "recently-updated articles" that meet the minimum ITN requirements (like DYK). There's also the possibility of somehow combining DYK and ITN. Levivich (talk) 17:18, 23 February 2023 (UTC)
As in, all articles, regardless of significance, as long as they meet the quality requirements? I would support that, so long as we include appropriate protections about it being abused by UPE's. BilledMammal (talk) 01:49, 24 February 2023 (UTC)
BilledMammal I think quality-only is the way to go, but I don't think it needs to be restricted to good articles. Any event article that meets basic criteria on quality should be included. WP:Notability (events) should be the "significance" standard. If it doesn't have its own article, then it's probably not worth putting on the main page. And if there's a recent event that you think should be included, just create a high quality article about it. Thebiguglyalien (talk) 18:36, 17 March 2023 (UTC)
Trending Topics
I believe supplanting with a "trending topics" of some kind is long over-due. We had posted this idea and got some amount of interest, but that thread as with many other ideas died because we did not know where to take it. In some sense the iOS app already does this.
Background: We often get comments tying stories and nominations to their potential popularity particularly as measured by page views. However, we all broadly agree that we should not be conflating WP:ITN significance with WP:PAGEVIEWS. Also, we agree that WP:ITN is not a news ticker.
Suggestion: I think this might be time to introduce a trending topics section either as a part of the WP:ITN box or outside of that. It does reflect quite poor if our mainpage after all these years is still fairly static in its content refresh capability and is not dynamic i.e. tailored either based on audience interest (trending topics), geographic interest (trending near you), or personalized reccos (tailored for you). Trending topics reflects the lowest level of personalization but is still dynamic, whereas tailored for you is the highest level of personalization, while trending near you is in between. This can either be text-based links or better still, images. Requires some amount of creative thinking and might not be in the remit of this group which is largely in a maintenance and operations mode.
Complexity: This is not an easy problem to solve since it requires a technical solution, which might or might not exist within the Wikipedia realm. Furthermore, there will have to be new sets of processes including of reviews and such that might need to be baked in.
Next Steps: Would love to get this group's input on the interest for such an idea. More importantly who would be the right group to take this idea forward, if at all.
I like this idea (trending topics), and I especially like it if it can be more than "just" WP:PAGEVIEWS (call it "general trending"), such as the examples you gave, "trending near you" (geographic) or "tailored for you" (personalized). Here are my concerns about it:
I'm no expert in these things, but I don't think there is a technical impediment, I believe a bot could use Pageviews API to grab page views and then write them to a Wikipedia page, which could be updated periodically (not sure what frequency is possible/desirable). I could be very wrong about this, as I've never tried to do it before.
I have no idea if the Pageviews API lets you break it down geographically. I don't think MediaWiki has the capability to deliver personalized suggestions (for logged-in or logged-out users), but I might be wrong about that, too. In theory this is doable, though, as many other websites do it.
I'm concerned about "false positives", which could be exploited. See [2]. This happened in January with topviews reporting Index (statistics) (don't know why) and Cleopatra (see [3]). However, we could account for this issue by allowing human editors (admins) to override the bot algorithm and exclude certain pages when appropriate.
Human editors could also override the bot to pull listings based on poor quality, if we wanted to do that.
It's possible to do this and not have it automated at all, but it would slow down the rate of update.
Still, I think it's an idea worth exploring, mostly because it takes the "significance" requirement of ITN out of the subjective hands of editor opinion and puts it into the objective hands of readership statistics. Levivich (talk) 17:16, 23 February 2023 (UTC)
Pageviews API is heavily rate limited; I don't think that is a viable solution. However, the WMF does do (daily?) dumps of pageviews, so pulling that file could work as well. Unfortunately, the WMF does not provide view localization, although I am certain that they do have this information and might provide it if requested?
MediaWiki does support content localization; we already see this with banners. It will take some work to make it work for page content but I believe this is something we can do and don't require the WMF for.
I would generally support a solution like this; we could also use weighted pageviews, to give extra weight to articles whose content has recently been updated.
Question. Who can take this idea forward to see if it has some merit toward implementation? Should I be posting this in some specific group? Thoughts? Ktin (talk) 16:40, 1 March 2023 (UTC)
Thanks @Levivich. I wish I knew of a way to share a mockup here. I can draw on powerpoint or a piece of paper and share that as an image. Would that help? Alternately, just replicating what the iOS app has by way of top read articles would be a win. Thoughts? Ktin (talk) 21:57, 10 March 2023 (UTC)
@Ktin: I think that looks really good. But my guess is that what'll get people really engaging with the idea is to see a picture of what the main page would look like with "trending topics" (or "top read" or whatever you call it) on it. My guess is an image (as opposed to a coded, functional mock-up) would be fine for discussion purposes. Unfortunately I lack the graphic design skills to create one myself. Levivich (talk) 00:00, 11 March 2023 (UTC)
Tool for bypassing redirects in see also sections
I recently noticed the link List of student newspapers in the United States of America in an article's see also section. On clicking it, it redirected to List of student newspapers in the United States, so I fixed it, but I suspect that there are many examples of this phenomenon. There might be a few instances in which the redirect is used intentionally/preferred, but in most I'd think we'd want to correct it. Would someone be interested in creating a semi-automated tool that'd allow you to quickly go through redirects in see also sections and bypass them if needed? Cheers, {{u|Sdkb}}talk19:42, 15 March 2023 (UTC)
@WhatamIdoing, this task would seem to me to be compliant with the advice there. It talks about e.g. not changing e.g. [[Franklin Roosevelt]] to [[Franklin D. Roosevelt|Franklin Roosevelt]], which would be a cosmetic task. But changing links in see also sections is not a cosmetic task because they're fully written out. Redirects there seem most analogous to redirects in navigation boxes or on disambiguation pages, which WP:NOTBROKEN specifically lists as instances where it's okay to bypass. Cheers, {{u|Sdkb}}talk20:46, 15 March 2023 (UTC)
Yes, NOTBROKEN is important and we must not replace redirects indiscriminately. (One LTA I deal with gets blocked for that regularly) However, See also does seem like a good place to clarify the target article's actual title, assuming that the redirect being bypassed is a near-synonym rather than a subtopic with possibilities. Certes (talk) 21:51, 15 March 2023 (UTC)
Why is the ==See also== section "a good place to clarify the target article's actual title"? How does that comply with NOTBROKEN, which says There is usually nothing wrong with linking to redirects?
Maybe the ==See also== section is a good place to link to Lorry instead of Truck, especially if the article that the link is being written in uses British English.
Or maybe the redirect in question has potential for expansion, and bypassing the redirect would eventually end up sending people to the wrong article. The powers that be have rearranged a couple of rare cancers, and the result for us will be that one article needs a new name, and one existing redirect will be repointed to the new name, instead of to its present target (which has been declared to be unrelated). If you bypass the redirects, people will end up in the wrong place.
Those sorts of examples are why I'm thinking of this as a semi-automated task with a human in the loop rather than a fully-automated bot task. If I saw a link to Lorry, I'd check the article's ENGVAR before changing it. I'd also use my judgement in the broader sense — if it turns out that 9/10 instances are straightforward fixes, great, but if every other instance requires more careful examination or is clearly a valid use of a redirect, I'd drop the task. Best, {{u|Sdkb}}talk03:05, 16 March 2023 (UTC)
I suspect every other instance would require more careful examination. They're also going to be hard to find. In theory, Quarry etc. could list the millions of links to redirects, but 95% of those links will be outwith See also and should be left alone. Some redirects should be bypassed and some should not, with a large grey area in between where good editors will politely disagree, but that's no longer relevant if we can't easily find them. Certes (talk) 10:45, 16 March 2023 (UTC)
Honestly just a tool that highlights links that don't match their title in general could be useful in a variety of circumstances outside of See also. :3 F4U (they/it) 14:11, 24 March 2023 (UTC)
Tool for highlighting links that don't match their title
Ooh, Freedom4U, I really like your idea here. Splitting that out into its own subsection so we can give it full consideration. Would anyone be interested in building this as a script? {{u|Sdkb}}talk14:30, 24 March 2023 (UTC)
As in, possible WP:EASTEREGG finder? There's something like that already, for non-matching interlanguage-links, which generates a report somewhere; I'd have to go look. Mathglot (talk) 22:06, 24 March 2023 (UTC)
I'm talking about redirects and piped links-- maybe they should be highlighted in different colors.... I can think of countless examples where something like that would have helped me out when cleaning up an article. :3 F4U (they/it) 00:48, 25 March 2023 (UTC)
@Freedom4U, it won't help with piped links, but you can use the .css for your skin to change the color of redirect links; see User:Schazjmd/vector.css for an example. (The last line in that file turns external URLs in the body hot-pink, which makes it super easy to clean them up.) Schazjmd(talk)14:35, 25 March 2023 (UTC)
Most of the 60 million links to redirects shouldn't be changed, as they are not broken. Piped links are a different issue. Clicking the "Random article" link ten times and checking the wikitext revealed 71 explicit piped links, plus an unknown number via various templates. If those articles are typical, we have about 40 million links that don't match their title. The good news is that the vast majority of them will already be either better than or at least as good as an unpiped link. Certes (talk) 18:04, 25 March 2023 (UTC)
There are a lot that can be fixed wrt redirects from my experience. Just recently I found a link to grilled beef that redirected to carne asada. There's lots of other instances I've seen this too-- I think one of the biggest things it'll help point out is when links aren't capitalized correctly. For example, I recently found an article where Front homosexuel d'action révolutionnaire was miscapitalized as Front Homosexuel d'Action Révolutionnaire. Nothing deal-breaking, just slight improvements. :3 F4U (they/it) 18:15, 25 March 2023 (UTC)
Not a miscapitalization as in outright wrong, but rather its a capitalization in title case, and Wikipedia uses sentence case. :3 F4U (they/it) 18:26, 25 March 2023 (UTC)
Keep a list of Problems/Issues
Topice Was Problem : Problems/Constraints are not specified before ideas We should agree on the importance of problems by creating a problem list which we can then refer to on ideas/proposals.
Many editors suggest changes/ideas/proposal or are frustrated with various problems. Part of the problem maybe that we jump to ideas and proposals, and then get stuck on discussing whether it is a problem or pointing out constraints.
If we had a list of priorised problems, then we could focus on evaluating ideas/proposals with the knowledge that there is consensus that the problem needs fixing.
If there are constraints that are causing us to stop fixing the problem (resources, WMF policy, WP policy, etc), then these can be looked at in the same way.
For the non IT peoples, In most organisations systems , there is a a set process that all changes goes through (except for emergency system changes or CEO whims) .We skip most of it.
Problems typically go through a process, something like this
Discover - gathered/elict processes
Analyse - the problems using lots of different methods Five_whys to find the causes ...
What is the extent? What is the evidence of the problems? What are the benefits?
What are the constraints? Can we solve part of the problem?
Prioritise. Is the problem worth fixing? Is there agreement? Is the problem within our control, or can we make it in our control?
This is a good idea. For example, the pending-changes protection for FAs seems to have worked, but no one has RfC'd it, something that imo is a solution that has just been forgotten. At the same time, I am also worried that this would create extra bureaucracy and creep, such as criteria for putting a problem on the list, preventing canvassing of solutions to those problems, and extra bots to manage the page(s). Sungodtemple (talk • contribs) 00:35, 24 March 2023 (UTC)
Maybe it would help to do an example using the FA pending changes protection
Problem : Should FA articles have pending changes BUT its best to have an issue stated as an issue, so "Reducing vandalism on FA articles"
Criteria for putting a problem on the list - Its probably better to have a bug/issue reported than not. But if the issue already exists then merge /redirect it
Importance Editors: Number of readers that click on FA account, Number of reverts on FA account, editor health (# of editors harassed after reverts), evidence provided Strength of evidence - Query # of reverts
Priority :Number of editors voting for it, number of proposals/solutions created, Importance
Canvassing of solutions - is already covered with current canvassing rules I think. But I think off wiki discussion and brainstorming as long as it is reported would be fine and needed
Bugzilla as an issue repository You mentioned maintenance issues - to avoid them I looked for existing open source issue systems that were linked to media wiki. There are Bugzilla extensions that write to mediawiki software such as (all descriptions below are from the link).
BugSquish Mediawiki Extension This extension checks your Bugzilla installation and adds a strikethrough to interwiki links for closed bugs.
Mediawiki Bugzilla Extension This extension allows you to list bugs from Bugzilla in a Wiki page using Bugzilla's REST API.
Hi! Curious if y'all think there are any aspects of editing where new editors may need more guidance/advice than what our current PAGs and explanatory essays provide, such as RFCs. — Ixtal( T / C ) ⁂ Non nobis solum. 01:26, 3 March 2023 (UTC)
I'll suggest the opposite: there's too much guidance, and the way it's organized makes it really difficult for new editors to know where to start. Thebiguglyalien (talk) 05:04, 3 March 2023 (UTC)
I'll second Thebiguglyalien's comment: too much "stuff", poorly organized. We can expect editors to go through Help:Introduction. But practically no one reads Wikipedia:Contributing to Wikipedia or any of those pages (I certainly didn't). They're both too basic and too verbose, making them pointless, while far less prominent pages, like MILHIST's Academy or userspace essays, are more useful by a mile. Nobody reads a manual before driving their new car; our help pages shouldn't try to address everything, they should focus on frequently asked questions, like how to copy edit (with specific guidance and examples), and how to find high-quality sources (too bad we can't just link out to LibGen/Sci-Hub, but people can still find tons of books/papers for free on the web or archive.org, yet most don't know how to). DFlhb (talk) 11:54, 3 March 2023 (UTC)
@DFlhb Maybe we need a new space called history/archive that we move all the old/duplicate stuff to. (I didn't know that the milhist academy exisited. Wow. ) Wakelamp d[@-@]b (talk) 15:14, 27 March 2023 (UTC)
The Wikimedia Foundation's Growth team has been providing very brief bits of information. The Editing team's new WP:Edit check will do more of the same, but right inside the visual editor, based on what the editor has typed.
One thing I haven't seen during the last decade is someone sitting down with Special:RecentChanges and figuring out what kinds of edits brand-new registered editors are making. Are people creating accounts to add paragraphs or to fix typos? The last time this was done, brand-new accounts could create articles in the mainspace (the Draft: space didn't exist then), and creating a new article was the first edit made by a quarter of accounts. Whatamidoing (WMF) (talk) 16:48, 14 March 2023 (UTC)
I'll third Thebiguglyalien's comment and add to and expand on it. "Too much" "not organized" "unvetted" "with no navigation (which for a newbie means "hidden")". One suggestion: add a "Getting started at editing" link lower down on the main page and have it land at a highly vetted / lots of eyes main navigation page for newbies. And direct links from it only to highly vetted pages. North8000 (talk) 18:11, 14 March 2023 (UTC)
Allow administrators to enforce structured discussions in CT/GS
This is an idea related to Wikipedia:Administrators' noticeboard#A plea: Propose a better solution, and before I actually propose it I'd like to workshop it a bit here first to see if it actually has legs and if there is anything obvious I'm missing. I think a small improvement we can make to the community processes around CT/GS, both at AN/ANI and in content discussions is give uninvolved administrators the tools to enforce a formalized discussion structure when they see a need. This could be a benefit in long, drawn out ANI threads that may have been better suited to AE, but have grown too large to move to that venue, or in RFCs and content discussions which are clearly generating heat to prevent them from igniting. Tools to step in and prevent bludgeoning, unconstructive back and forths, and sprawling, unreadable clusterfucks will make it easier to attract uninvolved closers, keep heated discussions focused and on-topic, and keep editors from crossing lines that may lead to sanctions. Hat tip to Tamzin for writing up most of the following legalese, although I've made some changes.
Any uninvolved administrator may take reasonable actions to structure a discussion in any area designated as a contentious topic or under community general sanctions, included but not limited to requiring section based (rather than threaded) discussion, imposing maximum word limits, and closing a discussion that has become unconstructive pending uninvolved closure. Administrators may enforce such restrictions using admin tools if necessary. Decisions can be appealed to the administrators' noticeboard.
The administrator could pick and choose how to structure the discussion, and not every remedy would be needed for every discussion. Sometimes just limiting those taking part to 500 or 1000 words would be enough, and there are other times when strictly enforcing sectioned discussion with word limits would help the situation. For an RFC or topic ban discussion, enforcing a structure where only !votes of up to 200 words are allowed under Option A, Option B, and Option C, and discussion may be threaded, but editors are restricted to 500 words may be enough to keep a dumpster fire from flaring up. Hopefully this would also serve to encourage editors to think carefully about their response, and have their supporting evidence ready at hand when responding. There's also a large benefit in seeing that a discussion is over-cooked, and stopping it at that point, even if you don't have the time to close it yourself at that point. Generally, the longer a discussion is open after it's well done the more acrimonious it becomes, which only serves to harm the community.
This is not a large change, and it won't solve all of the problems with CT/GS discussions or ANI, but small changes are easier to implement, and any improvement is an improvement. Any input on this would be appreciated, and if it seems well received I'll make an actual proposal at WP:VPP. ScottishFinnishRadish (talk) 12:34, 6 March 2023 (UTC)
Yes, please. Questions that would need to be answered: User:NoWords4U implements a 200 word / 10 diff limit on a given conversation. User:IHave11Diffs wants to post 11 diffs. Would any admin or just NoWords4U be able to grant said request (i.e., would granting extensions count as modifying a sanction)? Also to be considered: how would !votes on resolutions work? At AE, there is a section for patrolling admin comments; would there be something like this for everyone? For only uninvolved editors? Any of the above, depending on what an admin decides?HouseBlastertalk15:26, 6 March 2023 (UTC)
As far as granting extensions, I don't see a reason why any uninvolved admin wouldn't be able to do so. As far as closing, it would depend on the specific structure implemented by the uninvolved admin, but I don't see most of the discussions needing specific admin closure. The aim isn't to make it exactly like AE, it's to enforce enough structure to keep the discussions constructive and on point. An admin enforcing a structure could say that those responding to the discussion must declare in their section header if they are previously involved. The hope is that it would be a broad toolset so an admin could enforce the structure that best fits the situation. ScottishFinnishRadish (talk) 16:22, 6 March 2023 (UTC)
I dislike the idea of implementing word or diff limits; sometimes editors need more space for their statement, and the issue resolves itself because the longer a statement is the more likely it is to be ignored. BilledMammal (talk) 15:45, 6 March 2023 (UTC)
Is the idea to propose these as general sanctions so that there would be a presumption against overturning once implemented or to be resolved through the normal ANI/AN method of actions that sometimes get undone (and redone)? Best, Barkeep49 (talk) 15:59, 6 March 2023 (UTC)
I'm a bit up in the air about that, but I think it would be more effective as a general sanction, so that if there is an issue with the structure it has to be discussed at AN rather than changing the discussion parameters another time, possibly unnecessarily. ScottishFinnishRadish (talk) 16:25, 6 March 2023 (UTC)
Even if, as the community cannot directly modify the Contentious Topic procedures. So someone could either ask the Committee to do it or it could be done as a community General Sanction. If you're going to ask the Committee to do it after an RfC, I would recommend (and this is me speaking only for myself) making sure that's noted when people participate in the RfC. Historically the committee has been reluctant to take something the community has passed over without some reason for doing so (such as part of a larger case). This reluctance is why the committee has tried to provide the community a way of aligning its GS with the Contentious Topics procedures so that the community could use AE for its GS. But ultimately that remains an option and the community can decide it doesn't want to align with the GS it passes (and so far hasn't chosen to re-align any existing GS). Best, Barkeep49 (talk) 20:09, 6 March 2023 (UTC)
There are enough other big conversations in which editors want to engage that I suspect the current status quo (where a clerk already modified the General sanctions page to say "... when the community designates a topic as a contentious topic") will be considered good enough for the near term. isaacl (talk) 22:42, 6 March 2023 (UTC)
I think SFR is proposing it be done as a general sanction. That is, we "make a rule" that says something to the effect of "Admins may do [the things outlined in this proposal] to discussions related to topics that are listed here" I believe this would be within the purview of the community, though I definitely agree it would be much easier if ArbCom passed a motion affirming the result of the RfC if it is successful.HouseBlastertalk23:20, 6 March 2023 (UTC)
Yes, I know and I agree. I feel the arbitration committee would see benefits in keeping the standard set of page restrictions in one place, rather than having them split across the Contentious Topics page and the General sanctions page, and thus would be willing to pass a motion, but if it disagrees, then it can be split up. isaacl (talk) 22:35, 6 March 2023 (UTC)
I think @ScottishFinnishRadish hit the nail on the head: small changes are easier to implement, and any improvement is an improvement. I am strongly in favor of introducing a more efficient structure, even if it's through trial and error, because the issue is urgent. I've said this in the original thread and I feel strongly about WP:WALLOFTEXT editors at ANI who often don't even edit in GENSEX. We're technically not a bureaucracy but somehow it's precisely the shortcomings and endless complications of certain bureaucratic processes that makes this entire process more challenging for many editors who work on some of the most emotionally and intellectually taxing areas of en-wiki (and many of whom already feel very disenfranchised in the real world). Ppt91talk19:36, 6 March 2023 (UTC)
The administrator could pick and choose how to structure the discussion
not a fan of this, also not necessarily a fan in general - I feel this is a bit of a last resort thing. Controlling a discussion aggressively is often equivalent to controlling its outcome. Would far prefer a canned set of solutions chosen to limit this effect, and less invasive constraints like "we won't talk about this for a while". Also do we know what works? And are we sure the community actually needs coercion rather than a good model? Talpedia (talk) 22:09, 6 March 2023 (UTC)
Personally, I don't feel choosing a discussion format is equivalent to aggressively controlling a discussion. I appreciate some formats may be less appropriate for certain types of disputes, thus I think it's good that the proposal is providing flexibility for an administrator to decide upon a suitable format. I'm not sure that a canned set of solutions is less invasive than setting a format for discussion, and in particular, a single person deciding that the community won't talk about something for a while feels like a very controlling approach. isaacl (talk) 22:58, 6 March 2023 (UTC)
Not equivalent... more potentially enables in the absense of limits. Not talking at least doesn't reach a *conclusion*. I might read up / summarize some literature on influence in meetings for clarity. Talpedia (talk) 10:37, 7 March 2023 (UTC)
IMO probably only about 2% of Wikipedians and 5% of admins could successfully execute that role than you describe. IMO unless you could identify the 5% I think this would do more harm than good. Sincerely, North8000 (talk) 22:51, 6 March 2023 (UTC)
This feels to offer quite a lot of control to any individual admin and could be a somewhat de facto change to ANI format for CT topics if used with a significant level of frequency. What would be the mechanism should admin A decide to bring in said rules for a discussion and others (editors or admins) disagree with them? Other qs - the proposal suggests that this will be used for complex/scrawling discussions - would they only apply to comments made post-addition? That could potentially be imbalanced. However, something like rate limitation I could see being of more used than word/diff-count or, more drastically, sectioning. I'd also have zero issue with an obligation to disclose involved status - I think that last one would be a good general AN/ANI rule. Nosebagbear (talk) 00:21, 7 March 2023 (UTC)
section based (rather than threaded) discussion Ugh, no. Have you ever tried to go back and read a "discussion" that was held that way? It's nearly impossible to get any idea of what the actual discussion was. It only works for Arb cases because mostly everyone is talking to the Arbs, and the Arbs have their own threaded discussion on private mailing lists, and because few ever read those "discussions" after the fact (the final decision is the important part). imposing maximum word limits Seems like another measure to shut down discussion rather than to encourage better discussion. Which side has more editors, to have a higher word limit? Will we have people going back to delete their old comments to be able to make new ones, further hiding the actual discussion from anyone who hasn't been following it in real time? Anomie⚔13:37, 7 March 2023 (UTC)
Just on the word count limit thing, and I express no opinion about any other aspect of this proposal -- word counts are meant to keep the matter readable and to ensure that one or two passionate editors don't drown out everyone else with long conversations. I get that. But they're a significant barrier for uninvolved people to participate. My admittedly limited experience with Arbcom is that I don't know how many bloody words I'm going to need and I always want to have some in reserve in case I need to answer other editors anyway. I find word count and diff limits a significant stressor. A much better rule would say "be as succinct as possible" and enforce that with clerking by uninvolved sysops.—S MarshallT/C19:14, 7 March 2023 (UTC)
I'm sorry, I don't recall that my access to the tools gives me the power to officially determine who is being succinct and who isn't. (And now, I am being snarky.) Donald Albury01:26, 8 March 2023 (UTC)
Sysops routinely officially determine who's being disruptive and who isn't. Sysops routinely officially decide when someone's engaged in undisclosed paid editing, and block them despite their denials. You can find the consensus in an unstructured and undisciplined discussion. You read text and decide what's fair and what's out of line, and thereby keep the encyclopaedia running smoothly. If you say to me that you can't read a textual conversation and tell who's being succinct and who isn't, then okay, I believe you. But *I* can and I don't think it's a difficult thing to do.—S MarshallT/C08:52, 8 March 2023 (UTC)
I don't know, I feel like making it a judgement call would allow too much bias to creep in. This whole discussion was triggered by dysfunction in the GENSEX area, which is an area rife with all kinds of different biases. In the last few ANI cases from that topic area I saw several people go after the accused editor for "bludgeoning their own ANI case". Whether you think that's the truth or not, I strongly believe that's exactly the kind of situation we need to avoid, because it is both an accusation that is really hard to fight back against once you've received it and entirely subjective. A word limit with limited extensions would avoid that pitfall a little more efficiently. Preferably, those extensions would have to be granted based on something other admins can verify. --Licks-rocks (talk) 09:28, 8 March 2023 (UTC)
Well, we're writing an encyclopaedia. So our job is to read the sources and summarize them. We explain complex and difficult matters in simple, clear words. That's an academic discipline, called precis, that I and many other editors my age were taught as part of our English studies in grammar school. And with all due respect, it's easy to know what's brief, simple and clear and what isn't. Any fool can tell when people are repeating themselves, and the other aspects of "brief, simple and clear" are objectively measurable with Flesch-Kincaid tests. There are websites where you can paste disputed text and get the Flesch-Kincaid score as a number.My favourite example of horrible prolixity and slack precis skills in text is our Wikipedia article on Attention deficit hyperactivity disorder. Its Flesch-Kincaid score suggests an audience of professionals educated to at least college degree level who are fluent at reading technical language. It's like they designed it to be incomprehensible to ADHD sufferers.Anyway, being brief, simple and clear is a core skill for anyone who edits text and it shouldn't be hard for any encyclopaedist to construct a brief, simple and clear explanation of their position.—S MarshallT/C08:56, 9 March 2023 (UTC)
I think the main skill in arguments is having been in enough discussions to know when arguing more doesn't help your case... unfortunately that can depend on who you are talking. Talpedia (talk) 10:11, 9 March 2023 (UTC)
I've only skimmed this proposal, but I'd like point out one advantage of word limits vs clerking; they can't be construed as favoring one side or the other. In a heated debate, it's good to have strict rules, uniformly applied. Otherwise, you risk somebody saying, "You let the other guy say more, why won't you let me say more?", and then you're down that rabbit hole. What arbcom does seems to work: you start with a fixed limit and if you run into that, you can ask a clerk for more. That still brings human clerking into play, but at least the first barrier is a fixed number. -- RoySmith(talk)19:34, 7 March 2023 (UTC)
I think it's unavoidable that people will (inaccurately) construe CT-enforcing sysops as favouring one side or the other.—S MarshallT/C22:29, 7 March 2023 (UTC)
I think a system like this would need clear rules around when and how it is being/able to be enforced. for example, different tiers of enforcement depending on topic area or subject matter. (e.g. "random vandalist IP #2989 active again" probably doesn't need the same level of structured discussion as a highly controversial topic in the GENSEX area does.), ranging from none at all to the AE like format Billedmammal proposedhere. I think under those conditions, this would form a massive improvement to the way things currently work at ANI. I don't think it's a way of shutting people up, I think of it more as enforcing WP:COAL by force. That doesn't sound too bad to me. As for word limits: I think they're fine as long as people get an extension they can use. ----Licks-rocks (talk) 19:52, 7 March 2023 (UTC)
I can generally support this, as it doesn't propose diff limits. That would be a major failure point. Much of the back-story of this proposal is that the earlier proposal was to force all GENSEX ANI reports to AE, but AE is poor for establishing long-term patterns of abuse because of AE's diff limits and AE admins' general hostility to evidence that is not brand new (AE is only good for evidencing and stopping recent "flare ups"). — SMcCandlish☏¢ 😼 22:38, 7 March 2023 (UTC)
I think one discussion format that may be helpful in some cases is a round-robin discussion phase, where each contributor can make one comment until a moderator decides the next round of comments should start. The moderator would also decide when the round-robin phase ends. This would help prevent rapid escalation of contentious statements, and give everyone a chance to weigh in before the next round of responses. isaacl (talk) 18:18, 8 March 2023 (UTC)
The complaint that AN/I is awful, unworkable, untenable, dysfunctional and a terrible stinking mess is one that I've been hearing pretty much the entire 18+ years I've been editing here, and yet AN/I goes right on sanctioning those who need to be sanctioned, not sanctioning those who don't, warning people when they need to be warned, and -- in general -- getting the job done pretty well. True, it's messy, it's sometimes hit-or-miss, things slip through, and there are aspects of it which are as annoying as all get out, but the same can be said of any semi-democratic process which involves large numbers of people. It really doesn't need a lot of "fixing", and it certainly doesn't need to be "fixed" in a way that makes it work like AE or any other process. AE works because -- although anyone can comment -- it's pretty much functionally limited to a smaller population of edits, the admin corps, and in actual fact, to a much smaller group of admins who bother to participate there. I would oppose any change to AN/I which would involve forcing structured discussion of any kind. That doesn't mean that I think that there's nothing that can be done to improve AN/I, just that an attempt to force it to be something it isn't seems to be to be a bad idea. Beyond My Ken (talk) 23:56, 11 March 2023 (UTC)
The intent isn't to change ANI entirely. For the most part ANI does a find job of handling actual incidents. The intent is to, when the need arises in any discussion, implement some structure to keep a discussion productive. ScottishFinnishRadish (talk) 15:35, 13 March 2023 (UTC)
Thanks SFR for splitting this out to discuss. I've gone back and forth on this, but I think ultimately it doesn't really successfully address the issues with ANI, or rather introduces a bunch of unwanted additional issues. Randomly empowering random admins to reformat or structure discussions the way they see fit is going to feel even more arbitrary or random than discretionary sanctions/general sanctions feel now, especially to newer users. I think for it to work, there'd need to be a much clearer and documented process, or something that kicks in automatically. Der Wohltemperierte Fuchstalk15:29, 13 March 2023 (UTC)
I thought about triggers, so that it would kick in in certain circumstances, but that's exceedingly difficult to try and nail down. It's a lot easier to recognize that something has become, or is in the process of becoming, a quagmire, or dumpster fire and trying to head off the likely escalations. All in all, it's a tough situation to deal with because the arbitrary nature is a problem, but so is trying to write up a bunch of if statements to meet the threshold for deciding that unlimited threaded discussion isn't the way to handle a discussion that is quickly going off the rails. ScottishFinnishRadish (talk) 15:34, 13 March 2023 (UTC)
No. This is basically to enable Admin-rule on discussions where they have a significant conflict of interest or such a strong point of view they will effectively be given licence to suppress opposition they dont like. Pretty it up how you like, but this conversation (and the discussion at AN) has come about because the "wrong person" ended up sanctioned and some people dont like it. Which is a terrible basis for a change in procedure, a change explicitly to prevent the same thing happening again. Only in death does duty end (talk) 12:15, 15 March 2023 (UTC)
Oppose. For the exact reason given by Only in death. The ANI that inspired the idea proposed here is an ANI where an editor considered to be "fighting the good fight" in favor of transgender causes had violated Wikipedia policies and community guidelines more than once, had received warnings, but continued to push the envelope until it could no longer be tolerated. If you read some of the comments in that ANI, editors who supported a topic ban based on the diffs provided, including editors who had a history of dealing with said editor, were accused in general of being "out to get" that editor by some who were comrades of said editor, and the ANI was a weapon being used to prevent this person from contributing to transgender-related articles — because of course, in their eyes this editor was a warrior who could do no wrong. Gensex articles are going to draw contentious editors and contentious editing because they attract too many activist editors with personal agendas. Pyxis Solitary(yak yak). Ol' homo.10:46, 20 March 2023 (UTC)
I support this and furthermore think it's probably a better solution than the previous two suggestions. I think that being able to "cool down" a discussion like this will do a lot to cool down bad drama-based ANI bans while not having much effect on good ones. Loki (talk) 00:53, 20 March 2023 (UTC)
I feel Wikipedia's gotten far too conservative, and there's far too much inertia for anyone trying to make any improvements. Why not try this, or try the suggestion below, for a single CTOP and for a limited duration (a few months), and then reflect? When did we get so risk-averse? We need to restore a culture of experimentation: run trials, analyze, iterate. DFlhb (talk) 16:49, 27 March 2023 (UTC)
This, 110%. Our "if it ain't broke..." culture is detrimental to the development of Wikipedia. We do not protect FAs because they are "ain't broke" to prevent them from being "fixed". Sure, FAs may be small-g good articles, but that doesn't mean we shouldn't try to make them even better. Same thing with our PGs/norms/practices: experiment, and see what happens. Things that are commonplace and (I believe) widely seen as Good Things had "there is no problem" opposes. Then the change happened, and Wikipedia was made better. To name a few, in no particular order: de-mopping inactive admins (e.g. opposes 2, 6, 8, 16), introducing rollback (e.g. opposes 79, 88, 115, 120), letting 'crats remove the mop (e.g. opposes 1, 2, 3, 5, 6, 9, 10, most of the others), and ACTRIAL (e.g. the 101 people who supported the linked proposal by Ironholds).Heck, from WP:5P5, "Wikipedia has policies and guidelines, but they are not carved in stone; their content and interpretation can evolve over time". Can we please let them evolve?HouseBlastertalk19:04, 27 March 2023 (UTC)
5Day Limit at ANI
User:Tamzin has asked us for ideas (back at WP:AN) on how to deal with reports at WP:ANI. I have a proposal that does not involve ANI reform in the usual sense. I suggest that any thread at ANI that involves a contentious topic that is still being discussed after 5 days should be closed and transferred to Arbitration Enforcement. The ArbCom is intended to adjudicate disputes that the community cannot resolve, because the dispute divides the community. A contentious topic dispute that is still being discussed (argued about) after nearly a week is a recurrence of disputes that divide the community and have been delegated by ArbCom to Arbitration Enforcement.
@Robert McClenon: Given that a siteban proposal takes 72 hours (except in SNOW cases), and it usually takes a bit of time to get to the siteban proposal, do you think that this would effectively hinder the ability of the community to siteban bad actors as a community sanction? — Red-tailed hawk(nest)17:36, 17 March 2023 (UTC)
User:Red-tailed hawk - That's an interesting valid concern, and illustrates why brainstorming here is useful. Perhaps the proposal should be for any thread at ANI that is still being discussed and is not considering a siteban. After all, my stated intention was not to prevent the community from dealing with trolls, flamers, and similar bad actors, and trolls and flamers are often banned. The objective is to identify ANI threads that are going nowhere and send them to AE, as opposed to those that the community can handle. Robert McClenon (talk) 19:06, 17 March 2023 (UTC)
5-day limit? No. Sometimes it takes a few days for an editor to find out there's a discussion at ANI about a problematic editor. Limiting ANI discussions to 5 days will become a way to keep editors who aren't attached at the hip to ANI from participating in discussions they would have otherwise become involved in. Pyxis Solitary(yak yak). Ol' homo.14:11, 18 March 2023 (UTC)
Making ANI into arbom-light could be a good idea. Some of the same methods used at arbcom could be translated into an ANI version, that is less restrictive. arbcom has time limits, but 5 days is actually more restrictive than arbcom. -- GreenC14:19, 18 March 2023 (UTC)
While I'm not entirely opposed to the idea of a time limit, I think it'd need to be a lot longer than five days. ANI is sort of a double-edge sword: it overreacts to drama stuff without a clear policy violation, but it also is noticeably better at dealing with cases of long term bad behavior without a clear policy violation (like civil POV pushing)... unless the person in question is just randomly popular, in which case AE is a lot better.
I may not have been clear that I was only proposing an ANI time limit on discussions that involve a contentious topic. I was not proposing to close down discussions after 5 days if there is no other forum. Robert McClenon (talk) 16:17, 21 March 2023 (UTC)
I would support such a proposal, although with a longer limit. I'd say 7 days seems a bit less restrictive, and interested editors can still comment at AE. — Ixtal( T / C ) ⁂ Non nobis solum. 16:52, 21 March 2023 (UTC)
Opposed - I don’t have a problem with an admin closing an ANI thread they think is just going in circles or is otherwise unproductive, but I don’t think we need to put a firm time limit on the discussions. Sometimes it just takes longer for an issue to be resolved. Blueboar (talk) 17:33, 21 March 2023 (UTC)
Support. Five days is enough. This does not put a time limit on a discussion. Editors are able to comment at AE. And support modification that there should be no referral to AE if a cban has been proposed during this period; but if the cban discussion resulted in no consensus (not consensus not to ban): yes, refer automatically to AE. —Alalch E.11:29, 27 March 2023 (UTC)
Google's continued unrealiability/censorship is something that should be talked about
I'd bet almost everyone uses Google here to try and find various sources. For those who haven't noticed, Google has become increasingly unreliable and there seems to be large amounts of information being hidden, whether it's intentional or unintentional
For instance, let's say I perform a search on "oranges". A non-political, non hot-topic issue. The search says it has found 16 quadrillion results, but when you start scrolling down the page, it limits the search to 247 results and gives this message (In order to show you the most relevant results, we have omitted some entries very similar to the 247 already displayed.If you like, you can repeat the search with the omitted results included). While you can repeat the search, you are not going to get anywhere near the 16 quadrillion results you are told exist. And as you go further down into the original 247 result search, you start getting news reports about Orange County. This isn't a one-off either as every search is like this and as it gets into political issues, the results clearly become more swayed in various directions.
This is a really big problem when it comes to research and resource gathering on this site. While you can deal with the results about oranges being hidden, searches on less known or less popular things are severely impacted. The information is still there, somewhere, but you really have to have to know what you are looking for in order to find it. The issue is not just with google either, as other search engines like Bing are doing this as well.KatoKungLee (talk) 14:55, 19 March 2023 (UTC)
Google limits the results displayed to 1000 and then eliminates duplicates, so you will never see more than 1000 results displayed, and for most subjects a few hundred. As far as I am aware it has always done this. Other search engines behave similarly. Phil Bridger (talk) 15:08, 19 March 2023 (UTC)
You can cast a wider net by searching for phrases. If "oranges" doesn't find enough interesting links, try "orange fruit", "orange tree", "orange color", etc. Searching Google Scholar, Google Books, or Google News will give you results that don't turn up in a plain Google search, and which are often more useful for sourcing WP articles than what you find with a plain Google search. Donald Albury15:39, 19 March 2023 (UTC)
Not to mention that there are other search engines. And most libraries have their catalogs on-line. Plus a wide variety of commercial databases and archiving services available through WP:LIBRARY. -- RoySmith(talk)15:51, 19 March 2023 (UTC)
Google used to have options to limit searches, but over time they have become increasingly less functional. Someone more cynical than me might guess that degradation to be due to their business model. Perhaps it's time for an "open search project" that was more concerned with accurate hit lists than with selling eyeballs? I miss Deja News. --Shmuel (Seymour J.) Metz Username:Chatul (talk) 18:03, 19 March 2023 (UTC)
This isn't a reliability or censorship issue. Unless there are good quality sources that Google absolutely will not return results for (like, if they suppressed links from the NYTimes), this seems like a standard approach in search engines to keep the most relevant results for you. Masem (t) 18:13, 19 March 2023 (UTC)
Is throttling search results to 1000 a problem, though? I doubt many people would be looking past the 1000th search result even if it were available, and I suspect the proportion of relevant reliable sources coming up at that point would be pretty low. Editors would be better off developing their search engine skills to effectively find sources than worrying about whether Google will cut them off after 1000 results! Caeciliusinhorto-public (talk) 15:24, 20 March 2023 (UTC)
yes and no. It still knows your location if you search like that and will tailor your results to your location and language. —TheDJ (talk • contribs) 22:22, 20 March 2023 (UTC)
Using search engines is a skill that can be learned. What you need is not to go through more than 1,000 results, you need better search terms instead.
For most Wikipedia research, just build a query that goes: site:nytimes.com OR site:... and include all the sites listed at WP:RSP that are relevant to your area (whether culture, technology, video games...), or rely on WikiProject-specific reliable source lists (WP:TV and WP:VG have those, I believe). You wouldn't believe how many results you can surface with this method, that would never come up with "casual" search queries. Remember Google Search is built for the use cases of the masses, not for us. It'll rarely surface old articles, for example. Use your search engine's "date" tool (for example, all results before 2005) for that. DFlhb (talk) 22:00, 20 March 2023 (UTC)
If you want to search the NY Times, go to nytimes.com and use their search tool rather than relying on a third-party search engine. -- RoySmith(talk)22:17, 20 March 2023 (UTC)
Agree with others. If you need to look at > 100 search results, you need more specific search terms, or there are problems with your topic not being notable. Regardless this is not a Wikipedia issue that users here need to deal with. Natg 19 (talk) 23:14, 20 March 2023 (UTC)
This SEO type site states that less than 0.63 % of searchers ever look past page 2. Are there better search engines that might be suitable for edits ? With more operators? Or where you can save searches? (The film project has a custom search engine, which I think is based on this [| google tool]. I used to like clusty which did clustered search but it seems to have gone, and I couldn't find any search engine result comparison tools Wakelamp d[@-@]b (talk) Wakelamp d[@-@]b (talk) 07:56, 28 March 2023 (UTC)
OpenAI and ChatGDP
Have you tried to chat with openAI bot ChatGDP? It is amazing, breathtaking to say the least. But what are the implications for WP? As I realize, ChatGDP has a "training data pool", with info retrieved from WP, among other sites. Whilst I think OpenAI is a great tool for enhancing education and research, I think there are potential dangers. I would like to suggest, we, the WP, suggest to openAI to only include reviewed articles ie Good articles or better. Have you guys discussed OpenAI issues anywhere else in WP? I 'd like to read others perspectives on the issue. Cinadon3611:13, 28 February 2023 (UTC)
I think it is actually better for LLMs to include “junk” in their training input, especially if low-quality articles are clearly distinguished from the Good articles, as this allows the AI to learn what good looks like and what bad looks like. Barnards.tar.gz (talk) 16:04, 1 March 2023 (UTC)
Large language models are good at reproducing coherent language, and bad at writing properly referenced encyclopedia articles. They produce something that at first glance looks like a really good Wikipedia article, but don't (for example) actually read, cite, synthesize, and create coherent articles from source texts. They can give something that has all of the trappings of a really good Wikipedia article on the surface, but is actually nonsense once you scratch it away. --Jayron3220:51, 1 March 2023 (UTC)
I feel you are a little harsh here @Jayron32. You are placing the bar too up high. We are at the beginnings here. I have asked chatGDP a couple of questions in my field of expertise and the replies I got were moderate to good. Which is astonishing, especially if you keep in mind that this is a new technology. Yes, I have also noticed the problem with references. Also, I feel there is an element of vagueness and cliches? Maybe, I am not sure. But, it is a project that is still developing. Cinadon3608:02, 3 March 2023 (UTC)
I too can search for common text strings and copy and paste the information in a way that shows no understanding on my end, but I know enough basic English to reproduce reasonable text. I can even search and replace synonyms and alter sentence structure so that it isn't readily obvious where I got the information from. It's a more efficient version of that. --Jayron3211:46, 3 March 2023 (UTC)
@Cinadon36 It's impressive in many ways but, at least as of now, it generates fake sources and doubles down on them, including fake ISBN numbers and Google Books links if you keep pushing, even though the information it provides appears factual. Which is to say, it can answer even complex question with correct data (eg. what were some of the main economic reforms of Reagan's presidency?), but if you ask for where the information came from (eg. cite actual sources) the result will be made up. It can even use actual names of scholars and existing publishers to create a veneer of authenticity. So really dangerous for review process at this point. I have found it to be most useful for creating outlines and organizing existing thoughts into a clear format. Ppt91talk22:52, 6 March 2023 (UTC)
Yes, @Ppt91, I remember my frustration when I asked bot for references and handed me a couple of fake sources. But that was not my point. My suggestion was that we, WP, tell OpenAi to include only good articles of WP or above- just as an advice. As far as I can tell, current version of bot is not adequate to create context for article.Cinadon3608:17, 7 March 2023 (UTC)
@Cinadon36 Thank you for clarifying and I see your point now. I think that this is something worth discussing as OpenAi develops the project. Consider me involved as this conversation is moving forward. :) Ppt91talk16:40, 7 March 2023 (UTC)
I think we need to make template to disclose when ChatGPT or similar tools are used to write an article. Similar to automated machine transition, people need to be warn that they need to check the work before they post it and use that template to allow others to double check.
on "I would like to suggest, we, the WP, suggest to openAI to only include reviewed articles ie Good articles or better. " specifically, It's an interesting suggestion, but I think there's a risk this encourages gaming the system. It's really not something that's within our control anyway, they can use any/all of the articles as they see fit. JeffUK15:03, 28 March 2023 (UTC)
Adding LLM edit tag
See WP:AFTAGS and Special:Tags for what I'm talking about. "LLM" refers to large language models, aka "AI chatbots". Here are three ideas:
Add a tag for LLM-assisted edits. This makes LLM usage disclosure far more consistent, systematic, and easily searchable than requiring such disclosure in the edit summary.
Add a checkbox to the "Publish box" so users can voluntarily apply that tag (in addition to the current checkboxes, "This is a minor edit", and "Watch this page", add something like "This is an LLM-assisted edit"). This can only be proposed on Phabricator, but it needs discussion first.
For users that fail to voluntarily tag, the WMF could add an edit filter based on an LLM-detection algorithm, and tags these edits with a second tag: "potentially LLM-assisted". Such detectors aren't great (low sensitivity, but good specificity), but they'll help with RC patrol.
Are we solving an actual problem, a very incidental problem or a theoretical problem ? Put in another way... Why do we not have the same checkbox for "This is edit is a copyright violation", "This is just me storytelling" etc etc... —TheDJ (talk • contribs) 15:54, 23 March 2023 (UTC)
Do you believe LLM use should be disclosed in the edit summary, or do you also think that's pointless? The tag is intended as a better alternative to edit summaries, but if it's pointless in edit summaries, then indeed it's pointless as a tag too. DFlhb (talk) 16:09, 23 March 2023 (UTC)
From an edit filters perspective, detecting language models' outputs using regular expressions would be an impossible job. It would have to be done by a bot (Like how ClueBot NG is run).
From an editor's perspective, it might just be not worth it to hunt for these kinds of edits. I myself have used ChatGPT to rewrite a section of an article and I don't think it would be a big difference. So I think that LLM use doesn't need to be disclosed, and that doesn't mean we can't catch problematic edits. 0xDeadbeef→∞ (talk to me) 17:29, 23 March 2023 (UTC)
True on regex vs bot. And after thinking about it, I agree that the practical benefits of disclosure are minimal, even in edit summaries.
At the WT:LLM draft proposal, we came to a weak consensus (low participation, nothing formal) in favor of disclosure, but come to think of it, we've likely overreacted and thrown everything but the kitchen sink into that draft; ChatGPT's been out of months, and the LLMpocalypse hasn't happened. And after checking the long WP:VPP discussion on LLMs, I'm not even sure where we got that "mandatory disclosure" idea from, because I'm not seeing any community consensus for it. DFlhb (talk) 18:34, 23 March 2023 (UTC)
I say that disclosure is an unachievable goal. It's like asking people to not raise their kids with the iPad/Youtube, a lofty goal but pointless. —TheDJ (talk • contribs) 14:48, 24 March 2023 (UTC)
A few years ago, there was an ad where a cute little girl said, I'm going to make this photo better, then clicked a button and said, there's, it's better. I think as more sophisticated technology gets deployed more widely into writing assistant tools (and it's already in many photo processing tools), users won't know if specific technology X was used. Plus I don't think it's a scalable approach to have a checkbox for technology X, then Y, then Z, and so forth as new methods are developed. We need to instill a culture in authors of reviewing changes carefully, no matter what tools they used in the process of creating them, and not relying on tools to have automatically created text suitable for Wikipedia. isaacl (talk) 18:16, 23 March 2023 (UTC)
I don’t think we care that much whether an LLM was used, as much as whether edits comply with policies and guidelines. If an LLM manages to assist an editor in producing good edits, great. Barnards.tar.gz (talk) 16:50, 24 March 2023 (UTC)
It looks like there isn't going to be a consensus to require disclosing LLM-assisted edits either via edit summary or tag. —Alalch E.19:38, 26 March 2023 (UTC)
I am opposed to Idea #2. This has a good chance of backfiring, as a checkbox saying "Was this written by ChatGPT/similar" is only going to make people think using ChatGPT to edit Wikipedia is encouraged. That's the opposite of what we want, which is somewhere between "Use LLMs responsibly and manually check everything you write" and "Dont use them at all/Avoid using them". Until we get overwhelming consensus that LLM use for editing should be encouraged, a checkbox like this will be a bad idea. Soni (talk) 19:13, 27 March 2023 (UTC)
I agree, highly likely to backfire. It would make people aware of the opportunity to use LLMs on Wikipedia even if the idea hadn't crossed their mind. This would, guaranteed, increase usage of LLMs on Wikipedia. —Alalch E.22:13, 28 March 2023 (UTC)
Creating minimum inclusion criteria for lists involving subjective categorization
This has been touched on at the Village Pump over the last couple months, but I want to present a specific example for comment this time. 2001 has been updated with the intention of making it more encyclopedic (before and after). Given how much of a massive change it is, I want to get as much input as possible before similar edits take place on other year articles. I'm particularly looking at WP:LAYOUT, WP:PROPORTION, and WP:SUMMARY, but any criticism is helpful. Thebiguglyalien (talk) 03:36, 23 March 2023 (UTC)
Looks solid on all three counts, I'm impressed. I expected WP:PROSELINE to be inevitable, but that article avoids it skillfully.
Shame that I missed the prose RfC; this article is a great showcase for the benefits of prose in Year articles. Cleaner, more readable, better organized. DFlhb (talk) 14:54, 23 March 2023 (UTC)
I do think the list element is visible in a few spots. The "accidents" section is basically "this happened on this day and killed this many people" repeated several times (and overall the "disasters" section is the one I'm least happy with). It's one of the things I'm hoping to get worked out. Thebiguglyalien (talk) 23:00, 23 March 2023 (UTC)
I almost changed skilfully to mostly, but couldn't come up with any specific improvements to make. It'll be a fun challenge anyhow. DFlhb (talk) 23:27, 23 March 2023 (UTC)
Looks great! The only criticism I have is the imagemap (is that what it's called?) at the top is a little too US-centric. I don't have specific suggestions, and it is a year where events in the US had massive repercussions around the world. So maybe it makes sense for 2001? But for other years, I hope it's a little more global. 🙢 - Sativa Inflorescence - 🙢18:05, 23 March 2023 (UTC)
Those imagemaps have been the subject of huge disagreements. No one can agree what should go in them or whether they should even be on year articles at all. Thebiguglyalien (talk) 22:54, 23 March 2023 (UTC)
@Thebiguglyalien, this is awesome. I really like what you've done. I think it makes sense to have a ==Section== for each of the main "2001 in..." articles (e.g., ==Sports== for 2001 in sports).
On possible improvements, I offer two ideas:
I think it would be a good idea to merge the ==Nobel Prizes== section at the end up into the relevant areas (e.g., culture for the literature prize, etc.) instead of keeping it as a separate section.
The births/deaths section, which is empty, could have some simple statistics about world population change, like number of births, number of deaths, and the estimated population at the beginning and end of the year. Years with unusual mortality (e.g., 1918 influenza pandemic, 2020 COVID-19 pandemic, the world wars) could have a sentence about that.
I really like this, and I feel like it gives me a better picture of the year overall, rather than a laundry list of individual items. WhatamIdoing (talk) 19:12, 24 March 2023 (UTC)
WhatamIdoing, births/deaths and Nobel Prizes (along with the events timeline) are the remnants of what the article looked like before, so now they kind of exist in a limbo. There's a "health and society" section near the top that kind of touches on population and such, but that's one of the sections that I'm not sure what exactly it should look like or what should go in it (especially since there's no 2001 in health or anything like that). Thebiguglyalien (talk) 02:19, 25 March 2023 (UTC)
This is a very nice format. If there isn't prose you could probably move the births/deaths links to the See also, but I agree with WhatamIdoing that putting it in demographics might be good for some years. Perhaps don't use a 2010 flag for the African Union being mentioned in 2001! CMD (talk) 02:39, 25 March 2023 (UTC)
Should the "Events" timeline remain on year articles after they're destubbed?
Related topic, so I'm adding this as a subsection instead of making a new post. Currently, most year articles are dominated by a timeline that lists a bunch of events from that year. But with 2001, the first half is an article and the second half is a timeline. The way I see it, there are three ways to handle this:
Keep the article and the timeline on the same page. A bit clunky and some redundancy, but all the info is in one place.
Allow year articles to develop and then WP:SIZESPLIT the timeline from developed year articles (so an article would be written at 2002 and then the timeline would be moved to Timeline of 2002 or Events of 2002).
Going forward, WP:MOVE the timeline to a separate page and then create a new article under the old title. This would have the same end result as the previous approach, with the advantage of preserving the timeline's history.
Keep a timeline on a main year article, but it shouldn't be the only part. Not every important event in a year fits neatly into a prose paragraph, and some people are a bit "data-crunchy" in their reading style. Keeping a timeline in addition to prose on the same article hits two birds with one stone. InvadingInvader (userpage, talk) 18:38, 26 March 2023 (UTC)
Keep both prose and timeline on same article - It will minimize the confusion of the general public, and will kill two birds with one stone. - L'Mainerque - (Talk - Sign) -18:54, 26 March 2023 (UTC)
It totally depends on the article. If we remove everything from the timeline that's not mentioned in the prose and we're left with a bunch of trivia, then the timeline doesn't really serve any purpose. So I think remove them if they don't add anything of value, leave them if they do, per article based on consensus as usual JeffUK21:50, 26 March 2023 (UTC)
Thanks for the ping. I did see your request for feedback earlier but couldn’t find the time to give a proper response. Overall, I think it’s an improvement, and I encourage the move towards normalising the year articles. They are not special - all normal content policies apply. So organising and formatting them more like a normal article is healthy.
However, I still have one fairly large problem with the year articles, in that I’m not sure we have a grip on how WP:DUE should be applied. Per WP:INDISCRIMINATE, verifiability of a fact isn’t enough. We must have a reason to include the fact, and to include it with due weight. So it’s not enough to provide a citation that an event occurred in the year. We have to show that sources consider the event to be significant to that year - and how significant. Stated another way, when a source describes the year 2001, what events does it talk about? Note that talking about an event that happened in 2001 is not the same as talking about the year 2001!
I personally think that 9/11 was a highly significant part of the year 2001, and therefore due weight would be to devote significant “column inches” to it. But my opinion is subordinate to what sources say, which leads us to the question: what counts as an appropriate source for a main year article?
In my view, citations on these pages should be to sources that talk about the year. Articles such as a Year in Review, or maybe a study identifying the big trends of that year, or anything where the subject is the year. Because that’s what our article’s subject is.
My concern if we don’t take this approach is that inclusion or exclusion becomes arbitrary and we have no way to resolve content disputes over what is actually significant. Barnards.tar.gz (talk) 18:18, 29 March 2023 (UTC)
I've actually been working on 2002 with this issue in mind, using only overview sources rather than sources that describe a specific event. I intended to ping you specifically once I had written a few more sections, because you had raised this point before. I'm hoping to get all of the main content sections for 2002 written in the next few days, and then maybe some comparisons can be done between 2001 and 2002 to see what works and what doesn't. I do think a certain amount of "here's a source saying this happened" will be inevitable, but hopefully we can figure out best practices for an article like this. Thebiguglyalien (talk) 19:02, 29 March 2023 (UTC)
Talk pages more like Discord
Wikipedia talk page discussions could be in a discord conversation format. So that you don't get a long thread of replies and it's hard to track who has replied to whom.
I find it often difficult to understand who is replying to what in a flat Discord conversation from reading it, while Wikipedia's indent levels at least provide some degree of information. —Kusma (talk) 10:57, 11 March 2023 (UTC)
I don't think this is a particularly difficult example to track replies, compared with much more convoluted cases I've seen. In any case, though, a linear set of posts would be worse for the purposes of tracking replies, requiring the reader to manually backtrack and thread conversations together. (I have in the past pointed out how a linear format makes it easy to catch up on discussions, since all you have to do is find the last post you read and continue from there, which is why it's so popular in places like online bulletin boards. But tracking replies is not one of its selling points over threaded conversations.) isaacl (talk) 21:53, 11 March 2023 (UTC)
I do most (*) of my viewing of changes in either wikitext view or the beta feature visual diff view, but they have their drawbacks. Wikitext diff view requires you to interpret the source in your mind and has relatively narrow columns. Visual diff view is more friendly for many types of changes. In talk page threads, though, often someone will change one of the list types (that is, change a * to a : or vice versa), and then visual diff view will flag the entire list as having been removed and replaced with the new type. Also, for numbered lists, it always shows the entire list (presumably as a way to preserve item numbering), which can add a lot of unchanged text to scroll through as the list gets longer. With both modes, you're relying on enough context being shown by default to understand the diff (it's generally not a problem with visual diff view; getting more context than needed is more frequently an issue). While of course bulletin boards offer no context by default, since the commenters know this, they will provide quoted context to help understand their response. (This of course leads to a lot of text being quoted repeatedly, and sometimes way more text than necessary.) isaacl (talk) 17:44, 12 March 2023 (UTC)
(*) Using diffs to view changes is simplest for pages on my watchlist, since then Wikipedia will track the last seen revision for me. As I don't have high-traffic pages on my watchlist, for those I'll resort to scrolling through the thread, or subscribing to it, which very handily lets you view the current conversation with the new changes highlighted (though you still have to scroll through to find any after the first one). isaacl (talk) 17:56, 12 March 2023 (UTC)
Although I'm not a fan of this idea, I do think the fact that the basic structure of talk pages hasn't changed in almost 20 years (wow) is indicative that something might need to be done. (The WMF's various attempts haven't been liked by really anyone, unfortunately.) casualdejekyll12:34, 16 March 2023 (UTC)
Based on the 2019 talk page consultation, the talk pages project was started, which has produced the reply tool, section-based subscription, and the new discussion tool, all of which have been well-received by those who use them (and are easily avoided by those who don't want to use them). For better or worse, there are a lot of editors who are vested in the current format, and so layering tools on top of it seems like the best compromise approach for now. (There are Phabricator tickets open such as phab:T230683 for making wikitext syntax more expressive when writing nested responses, but it wouldn't alter the commenting workflow. It would just enable more types of content to be nested at the desired list level, or make it easier for someone to specify the desired nesting level.) isaacl (talk) 16:17, 16 March 2023 (UTC)
I love the reply tool. I also just found out about deindenting. I would support a set of features that optionally make conversations more like on reddit desktop, where vertical lines are shown next to each comment, and if you click the line, it collapses the comment. I believe this makes a threaded conversation overall easier to navigate. Born25121642 (talk) 19:58, 16 March 2023 (UTC)
Oh my god-- I'm not the biggest fan as that could be designed better, but that is definitely an improvement :3 F4U (they/it) 19:24, 23 March 2023 (UTC)
Having a bot which could quickly and accurately update population templates across Wikipedia would be beneficial. The bot would use the census data from each country to make its edits (the United States Census Bureau for US pages, for example). I remember there was a discussion a couple months ago involving the bot, CenPop, but the request expired after little interest. I think it's worth a shot to make a bot which can update population stats on municipal pages. This would minimize the work needed and pages missed by manually updating these stats. To add insult to injury, all this work would have to be done again when the data is inevitably superseded by the next counts. Thank you for your time, have a good day! DiscoA340 (talk) 18:12, 4 April 2023 (UTC)
It is probably easy to update the infobox by bot, but the body and any tables will likely require manual updates. —Kusma (talk) 18:44, 4 April 2023 (UTC)
It would likely have to be semi-automated, but I could imagine it working by searching for the old number and citation in the article, and replacing them with the new ones. Snowmanonahoe (talk) 18:47, 4 April 2023 (UTC)
Liberate notifications for article creators
If one expands a redirect, or creates an article from a redirect, I believe the one that created the redirect will be the beneficiary of the notifications. At least this was like this when I created articles from a redirect.
Then also I do not create certain articles because I know I am not interested in the notifications and prefer some other editor with a greater interest in the subject and expanding the article creates it. The notifications are anyway at some point useless for the article creators, when they stop editing wikipedia. Couldn't we somehow develop a tool or button like the watchlist star, to subscribe and unsubscribe to notifications from an article? Paradise Chronicle (talk) 05:40, 5 April 2023 (UTC)
When you get a notification for a link to an article, it now gives you an option to mute link notifications for that article (on desktop site at least). Joseph2302 (talk) 13:52, 5 April 2023 (UTC)
Ok, thank you for the information about the muting. For information about a tool with which one can subscribe to those articles someone else has created, I'd still be grateful. Paradise Chronicle (talk) 17:10, 5 April 2023 (UTC)
@Paradise Chronicle, I think you're referring to notifications such as when an article is proposed for deletion/merge, nominated for deletion, tagged for speedy deleting? Notifications of prod/AfD/CSD don't happen automatically; the nominating editor notifies the article creator. (If the nominator uses Twinkle, the tool posts the notification on behalf of the nominator.) It would be a technical hurdle to create a class of editors "interested" in an article that could be recognized by a manual editor or Twinkle to notify in addition to the creator. I think your best recourse is to watchlist those articles. (If I've misunderstood your intent, I apologize.) Schazjmd(talk)17:27, 5 April 2023 (UTC)
I was more referring to Wikilinks to and from the article so one gets aware of some potential expansion. But yes, of course such notifications would also be helpful for many editors. I actually get the notifications of two articles I initially created but then someone merged them and since I do not appear as their creator below the article title anymore (if one activates the xtool preferences) but still both are mentioned as their creator at the general xtools see here and here for 1967 Basel Picasso paintings purchase referendum. The other one is Baris Atay. So, I believe somehow it is possible to receive notifications of articles one has not created, I just don't know how it can be done. Paradise Chronicle (talk) 17:56, 5 April 2023 (UTC)
Yes, subscribing to these for pages you haven't created would be very useful. As others have said, you can already opt out for pages you have created. —Kusma (talk) 16:43, 5 April 2023 (UTC)
Pending Changes for talk pages?
Has there ever been any discussion of the possibility of enabling pending changes for article talk pages? I've noticed an uptick in purely disruptive edits on some talk pages, especially those dealing with contentious subjects where the article is already protected. Currently pending changes is not an option for talk page protection. -Ad Orientem (talk) 14:28, 25 March 2023 (UTC)
@Ad Orientem This actually sounds like something I'd support, but maybe at a different level.
I know the Abuse Filter can tag edits for further review, but can it hold edits for review? i.e. the edits are not published until an admin/PCR approves them? That might be a good solution for disruptive edits to talk pages. Some forums like Reddit actually allow for this. Aasim - Herrscher of Wikis ❄️ 05:26, 27 March 2023 (UTC)
@Awesome Aasim no, there is no where for the abuse filter to put something pending review, regarding an edit it can TAG an action, it can interrupt an action with a warning, and it can prevent an action. — xaosfluxTalk20:04, 27 March 2023 (UTC)
@Ad Orientem One of my interests is analyzing reasons/location/processes for abuse, and trying to find causes to solve and prevent it. Most of disruptive/abuse edits I have seen have been on user talk
Can you provide some examples of some contentious topics where this is happening? ( I looked at you user page and your stated interests don't spring to mind as inspiring conflict). And what type of editorWikipedia:IP_users, experienced , SPA, very new troll, .... etc
@Xaosflux Do some abusive/disruptive edits get removed from history? And would the abuse filter tag/stop an issue that is only to do with the edit summary? I ask because I went through a dump of a day's edit summaries, and there weren't as many offensive ones as the comments on village pump and by WMF concerning the toxic culture indicate Wakelamp d[@-@]b (talk) 07:20, 28 March 2023 (UTC)
@Wakelamp disruptive edits may be deleted or oversighted. The abuse filter actions can happen alone, or in combination. Abuse filters can trigger on action summaries. An abuse rule that makes a "log" or "tag" does not stop an edit unless it also is set to warn and/or disallow. If an abuse filter is set to private, and it stops an edit it won't be publicly reviewable. You can see all public abuse entries here:Special:Abuselog. — xaosfluxTalk09:37, 28 March 2023 (UTC)
@Wakelamp The incidents I typically encounter are a result of reports that show up at WP:AIV as auto-reported by the edit filter. As an admin I try to keep an eye on AIV when I can. IPs are the most common, but sometimes we get newly created accounts that are in most cases socks of already blocked users. In a few cases I've seen some that I was pretty sure were some specie or other of LTA trying to get around the protection on the main article page. Good old-fashioned trolls also pop up now and then. -Ad Orientem (talk) 21:01, 28 March 2023 (UTC)
I would prefer not to extend pending-changes to talk pages for two reasons. Talk pages are already much less visible than Article space, so the harm of leaving disruptive content there is far more minimal. And my second reason is user-interface wise, I find pending changes incredibly confusing when there are multiple edits in a row. I never quite fully understand what accepting a change means, if there's disruptive content followed by productive content. An edge case question, do we have Article talk pages that are protected? (I only found User talk pages that were protected), and if so, how are editors supposed to request edits to said Article talk page? {{Edit fully-protected}} can only be transcluded on a talk page for example. ~ 🦝 Shushugah (he/him • talk) 22:24, 28 March 2023 (UTC)
do we have Article talk pages that are protected? 129 are semi-protected (see cat, but it includes archives, FAQs, and Wikipedia: namespace pages too), 0 full-protected (cat); mostly happens in WP:CTOPs, mostly very temporary, and: how are editors supposed to request edits, they're not! DFlhb (talk) 22:39, 28 March 2023 (UTC)
In addition to HouseBlaster's correction below, I'll add that there are in fact quite a few more protected talk pages that are not categorized as such, a few of which are indeed fully-protected (see Special:ProtectedPages). These are largely subpages and redirects, but there are some exceptions (e.g. 123) 74.73.224.126 (talk) 03:10, 3 April 2023 (UTC)
A new edit filter now prevents some talk page edits which clearly have no useful content. It should solve some but not all of the problem described above. Further details: WP:EFR#Talk page junk. Certes (talk) 14:42, 6 April 2023 (UTC)
Show number of thanks in edit history?
As a follow up to my other proposal for improving/reducing discussions on edits... (really hope that first proposal will be picked up someday as it would help SO much)
Why don't we show the number of thanks an edit has received in the edit history?
It might help show "consensus" with other editors when editing certain pages and avoid edit wars. A user might be less inclined to just revert an edit if he notices that others support it but might be more inclined to start a discussion. {{u|Gtoffoletto}}talk17:23, 21 March 2023 (UTC)
I think that if implemented this would pretty soon be gamed by spammers to such an extent that it would be meaningless. I get what seems to me to be a lot of thanks from other editors, and I also give some out. Let's all take private satisfaction in that rather than make this public. Phil Bridger (talk) 23:00, 21 March 2023 (UTC)
Looking at the code (I'm not a coder so forgive me if I'm wrong), it would suggest that Certes is correct in saying that it would not be technically possible at the moment. - L'Mainerque - (Talk - Sign) -23:04, 22 March 2023 (UTC)
I'm saying that the data stored in the obvious place only has thanker, thankee and time; no indication of which edit or even which page was edited. That information may be stored elsewhere, but I doubt it. There's a sample extract here. Certes (talk) 23:07, 22 March 2023 (UTC)
Actually this can't be right: when I get a notification of a "Thanks" it does show which specific edit received it. So that information is available somewhere. {{u|Gtoffoletto}}talk00:40, 23 March 2023 (UTC)
When you send a thank, you do send it for a specific edit or action (e.g. Special:Thanks/1146124820) - that could pass on to notification system and not actually "store" that part though, I haven't read all the code on this. — xaosfluxTalk00:51, 23 March 2023 (UTC)
I wasn't involved in that project, but I remember hearing that this was a deliberate decision. There are privacy implications in publicly revealing that Alice thanked Bob for a particular edit, because that means that Alice was reading a particular page or agreed with a particular contribution, which Alice might not wish to be public information.
The Echo/Notifications separately stores all of this information (thanker, thankee, timestamp, and diff). However, that is temporary (two years or 2,000 notifications, whichever comes first). The permanent log stores only the thanker, thankee, and the timestamp. This combination allows admins to check for certain kinds of problems (e.g., interaction bans) but doesn't publicize certain other kinds of problems (e.g., how many times someone got thanked for being rude). Whatamidoing (WMF) (talk) 01:53, 23 March 2023 (UTC)
Since you are thinking about Thanks, this might be interesting:
This week, the Growth team completed a quick analysis of Thanks usage.
In early January, the Comumity Wishlist proposal to Enable Thanks Button by default in Watchlists and Recent Changes was fulfilled. Since the Growth team has been working on a Positive Reinforcement project, we were curious to see if this change increased the number edits that were thanked. As is often the case, the answer isn't totally clear and seems to vary by wiki. But perhaps more interesting is to see that the ratio of edits Thanked varies considerably by wiki, and seeing the low percentage of newcomer edits that receive Thanks. I'm not necessarily surprised by either finding, but I'm always thinking about ways to ensure newcomers receive the support and encouragement they need to continue to contribute, so feel free to chime in here or on the Positive Reinforcement talk page if you have any ideas. :) KStoller-WMF (talk) 20:50, 24 March 2023 (UTC)
@KStoller-WMF I think the consensus here is that we shouldn't rely on thanks for content disputes. Which I think is a good point. However I think it's great that you are making it easier to give thanks to other users.
Another great place would be right here in the discussion page. I have a quick "reply" button but I don't have a quick way to send "Thanks" for your message. I wanted to do that for some of the messages received in this discussion such as yours and I would need to find them in the history log! {{u|Gtoffoletto}}talk15:29, 28 March 2023 (UTC)
Yes, I can see how Thanks could get problematic in content disputes if publicly connected to an edit.
I would love to see Thanks added to discussion pages! I know the Editing team was considering that work as part of the DiscussionTools work: T249893. My understanding is that the Editing team is hoping to shift to another project soon, so I'm not sure if that work will move forward. KStoller-WMF (talk) 20:50, 28 March 2023 (UTC)
I am happy that thanks are low-key. If we try to use them to resolve content disputes, I expect the collegiate informal nature of them would be lost, and we would start to see things like gaming the system, or creating an expectation/duty to “vote” on edits. It might also encourage users to eschew discussion in favour of just bumping the thank count. We already weigh edit count too highly when judging other editors; we don’t need another high score or anything approaching social media upvote/downvote buttons. Barnards.tar.gz (talk) 21:22, 24 March 2023 (UTC)
I think not. At best this will just be interesting trivia, at worst it could encourage WP:NOTVOTING on edits, "You can't revert my edit because 100 people liked it..." JeffUK21:54, 26 March 2023 (UTC)
I like this idea, though it's too easily to use this to make Wikipedia a popularity contest, and I think it's obvious that we aren't Reddit. I'd suggest keeping it low-key similar to the position of @Barnards.tar.gz's idea above. InvadingInvader (userpage, talk) 20:51, 5 April 2023 (UTC)
The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
Good day guys; every time I look at WP:ANI, I feel like that every other complaint or thread is on something which has to do with a legal threat. I think that discussion can be much more streamlined if editors who notice legal threats being posted are specifically redirected to a noticeboard dedicated to legal threats. Therefore, I would like to bounce around the creation of Wikipedia:Administrators' noticeboard/Legal threats (or WP:ANLT).
With a new noticeboard dedicated solely to legal threats, a dedicated team of admins can find every reported ;legal threat in one place rather than having to sift through ANI discussions often longer than many articles. Additionally, discussions not affected by legal threats would benefit as there would be less potential for edit conflicts (though this problem is starting to become better handled with new WMF technical developments like better talk pages). InvadingInvader (userpage, talk) 20:58, 5 April 2023 (UTC)
Most legal threats I see pop up at ANI are handled very quickly. I don't think splitting their handling off to another board with far fewer watchers would be as effective. ScottishFinnishRadish (talk) 00:10, 6 April 2023 (UTC)
I'm with SFR - I'm not aware of any particular issues with the status quo. In fact, it could be a nuisance for cases where we have a NLT and other issues overlapping. This is more relevant because a purely legal threat block is usually undone when retracted and awareness shown, while a mixed case may not. Nosebagbear (talk) 12:38, 6 April 2023 (UTC)
Legal threats rarely come in isolation, and moving them elsewhere would likely result in less scrutiny of the underlying issues. A bad fix for a non-existent problem. AndyTheGrump (talk) 13:10, 6 April 2023 (UTC)
Meh ANI is not overwhelmed with NLT discussions. Someone posts an NLT violation, the person in question gets blocked rather quickly, and we move on. It's working under the current system just fine. --Jayron3218:01, 6 April 2023 (UTC)
The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
A way of resolving "No consensus"
In an RfC, when there are two choices and no consensus can be reached, we are left with no decision. If no consensus generally means we stay with what we had before, then at least we have something, but sometimes there is no established rule to fall back on and we are left with mixed usage. In that case, Wikipedia is inconsistent, looks sloppy, and nobody's happy.
However, at that point we could ask another question to see if there is a consensus that either option is better than no decision, then a coin could be flipped or the closer could make a decision based on a majority rather than a consensus.
I proposed this in one case and the answer was that we don't do that here. I've been involved in several such RfCs on matters of style and Wikipedia remains inconsistent. I think either choice is often better than no choice.
I cannot imagine why choosing a result that is not supported by consensus makes sense. A perfectly consistent Wikipedia is a mirage that will never exist. Forcing agreement won't help anyone. — Ixtal( T / C ) ⁂ Non nobis solum. 13:33, 3 March 2023 (UTC)
@Ixtal: No hope of forcing agreement or becoming perfectly consistent. Right you are. But we could hope to stop endless arguments between pedants who seem to prefer to argue rather than solve a problem. SchreiberBike | ⌨ 23:03, 3 March 2023 (UTC)
It is fine for Wikipedia to be inconsistent, because the real world is. WP:ENGVAR is a great example of Wikipedia choosing to be inconsistent over one of the several options. This is a feature, not a bug, as it helps Wikipedia align with the real world. —Kusma (talk) 13:54, 3 March 2023 (UTC)
Should Wikipedia say "the Gambia" or "The Gambia" midsentence. There are good arguments for both. A recent RfC closed as "No consensus", so we have usage in articles and in article titles that goes both ways. For some people, especially some Gambians, this is a very important issue. It's been a low-key edit war for years.
Several books worth of writing and editor time has been spent trying to reach agreement on the capitalization of universe in an astronomical context. Again, reasonable arguments can be made both ways and no consensus has been reached.
Flipping a coin or picking an odd or even number in some lottery would resolve these and people would stop yammering about them for a while. Maybe people shouldn't care so much, but they do. SchreiberBike | ⌨ 22:12, 3 March 2023 (UTC)
I found the RFC for (T/t)he Gambia, but I'm not finding the one for (U/u)niverse. I can see Gambia being quite contentious. For (U/u)niverse, however, I would think it would be more context dependent. For example "There are billions of galaxies in the Universe" seems like it's being used as a proper name, thus should be capitalized, whereas in, "Imagine a universe where the laws of physics were different..." it is not being used as a proper name, thus should be lower case. I think your proposal might help provide clarity in a situation like (T/t)he Gambia, but it could also end up inflaming the situation, especially if Gambians feel strongly about it and do not feel their voices are being heard in whatever decision gets made. I do not think your proposal would be appropriate for something as context dependent as (U/u)niverse, but I could be misunderstanding the debate there as I have not found the actual RFC. ~ ONUnicorn(Talk|Contribs)problem solving16:45, 6 March 2023 (UTC)
The obvious question is, what happens if there is, in turn, no consensus for either option being better than no decision? Gnomingstuff (talk) 16:05, 3 March 2023 (UTC)
@Gnomingstuff: Then we are no worse off than we were before, but if there is a consensus that either option is better than continued inconsistency, it's a problem solved and Wikipedia is better, more consistent, more credible. SchreiberBike | ⌨ 22:16, 3 March 2023 (UTC)
@Rosguill: As an example, imagine Wikipedia as a new country with a new transportation system. I doubt we could reach consensus that we should only drive on either the left or the right; would we continue to let people drive as they please with the resulting accidents and inefficiency? Neither right nor left is inherently better. I'm wondering if people are interested in finding a new way to solve long-standing problems that have not been solved by our consensus system, even if that system usually works. SchreiberBike | ⌨ 22:25, 3 March 2023 (UTC)
The obvious solution is for everyone to slow down enough that it does not matter which side they are driving on… sure it takes longer to get where you want to go, but there are no accidents. Blueboar (talk) 02:13, 26 March 2023 (UTC)
Of course it's hyperbolic, but the examples I give above of The Gambia and universe correspond closely. We let people edit as they please resulting in argument and inefficiency. It makes Wikipedia look less professional, less consistent and less credible and it wastes our time. SchreiberBike | ⌨ 23:14, 3 March 2023 (UTC)
It is a waste of time to attempt to enforce uniformity in stylistic matters that have no real importance. Given how poor our sourcing is in many areas, our credibility is arguably too high already. Note also that WP:ENGVAR, our deliberate inconsistency in using all kinds of different spelling systems does not seem to have caused any real world issues that I am aware of. Making decisions that go one way and possibly violate WP:NPOV is much worse than inconsistency. —Kusma (talk) 10:55, 4 March 2023 (UTC)
There are times when it would be preferable to settle a dispute rather than continue it via no consensus, but they're few and far between. The recent WP:ACAS is a good example, and in that case the closers did basically as you suggest: reduce one of the questions to a pure headcount. My [well-documented at this point] position is that the headcount question was the wrong one -- that we should've come out of that slog with at least a definition of what we mean when we call about "mass creation" or the like instead of a strange, arguably redundant rule applying to an ill-defined concept. What makes it hard in a case like that are the entrenched camps and drama carried over from specific cases/people, coloring an important policy debate. There are specific cases when I feel like it would be a net positive to the community to either grant the closers slightly more leeway than usual to glean important conclusions from an otherwise messy discussion, or to find some other approach like you're describing. I certainly wouldn't support and policy/proposal that included relying on coin flips or headcounts, but there are other possibilities. For example (I tossed this out at the arb motions board), run the same RfC again but don't allow anyone who participated in the first RfC to !vote (talk page only). I think that would be worth a try sometime, despite the inevitable complaints. YMMV. — Rhododendritestalk \\ 23:07, 3 March 2023 (UTC)
@Rhododendrites: There are plenty of problem solving methods we haven't tried yet. What I'm suggesting is that if there is a consensus that people are willing to accept the results of some method besides consensus, we should specifically allow that. SchreiberBike | ⌨ 03:52, 7 March 2023 (UTC)
Given that we've been unable to reach consensus on the capitalization of universe in many discussions [which I will list], is there a consensus that the results of a coin flip (or similar random choice) would be better than what we have now?
I think continue the discussion here, but change the title/create a new topic similar to- What should we do when Consensus cannot be reached when there are strong opinions on small changes? Wakelamp d[@-@]b (talk) 15:00, 27 March 2023 (UTC)
In that sort of example, it depends on the question!
"Do we capitalise [u/U]niverse in all cases", may get 'no consensus'. "Should we have a consistent capitalization for Universe" may get a strong consensus for 'Yes'. You can then build on the conversation to define /how/ we determine which to use (i.e. "having agreed we should be consistent, let's use whatever the BBC uses"), and finally let that discussion determine the answer. If the consensus is 'No' or 'Use the local consensus', 'Use whatever the sources use' etc. then you also have a consensus. Ultimately, it sounds like a lot of effort, and we don't require absolute consistency in matters of style anyway! JeffUK16:00, 28 March 2023 (UTC)
@SchreiberBike Alterantes to random choice (which is a common Agile (or Fail-fast/learn fast approach) for low cost/low risk reversible decisions, these three wiki articles are relevant Consensus_decision-making, Group_decision-making, and Consensus-seeking_decision-making]]. We also have some extra constraints in terms of WMF unweildiness due to complexity (300+ languages and 800+ wikis) and long term technological under-investment
For instance, some group process chnages
- the Internet RFC approach which is a of requirements gathering, problem defintition, analysis, consensus, then summarise current arguments, and rinse and repeat. In this case I think the problem should be defined as "How we can allow people to display alternate content"
- Extend the consensus - If connsensus can not be reached then expand the electorate (consensurate??) involved in a non canvassing fashion. Start with article ->RfA-> Pump-> RfC-> Community-> Readers
@JeffUK Maybe we should interpret your phrase that "we don't require absolute consistency in matters of style anyway" in its broader sense, Why not allow readers to specify Wikipedia preferences for the style of words and phrases and then locally preprocess (or pre-process in the UK :-)) the wikitext before display (I just found [[5]]
A personal bugbear is when personal preferences in minor matters are imposed on groups (especially without evidence). Some issues with Vector 2022 were great examples as it stops evolution of systems and groups, and wastes time on the trivial but annoying/high friction)).
- specify dialect fot spelling (UK/US/Indian/Australian/..) There is some 2019 research in the 90% for identification of dialect words.
- Use O/S or user preference for display of dates, numbers,
- Spelling "The Universe" as "the universe" off standard or private lists.
- Use different style (Wiki's own , Strunk, Oxford commas etc)
- Automatic conversion of amounts into currency of choice, and
- Display of prefered Nationality for those endless fights over BIOs
- or even showsing editor names as their preferred Avatar
- Display preferred language if available (as non auto translated) and then current page language (eg Spanish, Simple English, English, autotranslated into Spanish) by section
Some of these cases have already been mentioned for [[6]] @User:Sannita_(WMF), @WhatamIdoing, but it would be nice to implement some parts of Abstract earlier,
Do you mean something like allow the user to set their language to en-gb and then wrap every word that has a variant in a template like {{i18n|color}} that renders as colour? Barnards.tar.gz (talk) 06:49, 30 March 2023 (UTC)
The below is a tortured attempt (with I am certain many errors) to show punctuation (quotation marks, Oxford comma, hyphenation), language differences, and spelling. The parts that woul be coloured would be The Gambia, camdy floss, crisps, and clag.
"He dreamt that night that he had apologised to his humourless neighbour from The Gambia by singing. ‘My accusation on the 1st of October 2012 AD at 8 AM (GMT) was unfounded; your well travelled llama did not eat candy floss coloured purple, or the crisps, or the pre-school’s clag labelled "toxic".’
"They dreamed that night that they had apologized to their humorless neighbor from Ghambia, by singing, “My accusation on October the 1st 2012 CE at 2000 UMT was unfounded; your well traveled llama did not eat cotton candy colored purple, or the potato chips or the preschool’s glue labeled 'toxic' ”.. Wakelamp d[@-@]b (talk) Wakelamp d[@-@]b (talk) 12:47, 30 March 2023 (UTC)
Immediate problem with that is that if it requires each word which has a possible variant to be marked up then it would make editing impossible for anyone who's not a computer programmer, and it couldn't be applied automatically, because you'll end up with "Color: spelling differences: Brits and the rest of the world spell color as 'color' while Americans spell color as color" either way, its' a hell of a lot of work simply to achieve consistency across articles, and there's a strong consensus that consistency across articles doesn't matter. Also, what about when Brits disagree with each other. e.g. "Dispatch Department" and "Despatch Department" are both in common use in the UK. JeffUK19:37, 30 March 2023 (UTC)
Maybe no markup editing change is needed, and definitely not consistency - without our quirks Wikipedia becomes soulless and dull.
Mediawiki- Could you tell me where the dispatch department is at The Gambia's main airport?
New pre-processor - Has the reader specified a preferred dialect? Yes - British UK version 14 (despatch, Gambia)
Wikipedia displays : Could you tell me where the despatch department is at Gambia's main airport?
Displayed on Edit ; Could you tell me where the dispatch department is at The Gambia's main airport?
Displayed on Visual edit (for preference British UK version 14; Could you tell me where the despatch department is at The Gambia's main airport?Wakelamp d[@-@]b (talk) 05:53, 7 April 2023 (UTC)
One-word talk page contributions
I'm seeing more of these than I used to, talk page edits where the heading and content are just one or two words each:
Some pages seem to attract these repeatedly. Is there some aspect of the talk page (re?)design that's making IP users think that they're typing in a search field? Or has this always been a thing?
I'm not sure what the solution is as I don't really know what's causing it.
When starting a new thread, the box to type a message in just says Description, which doesn't make it obvious what the box is for, or what the user is expected to type there. If that's the issue, would it help to change the message in the text box to something like Enter your comment here? Or maybe some kind of filter where a new thread consisting of just a single word in both heading and subject is rejected as almost certainly an erroneous or test edit? Belbury (talk) 16:08, 5 April 2023 (UTC)
I think the message probably should be updated to "Enter your comment here" which is a pretty good description of what the box is for (or something else as suggested). "Description" really isn't enough of a, uh, description. Galobtter (pingó mió) 23:41, 5 April 2023 (UTC)
Or 2, 3 or 4 word edits. There are undoubtedly far more of these than in the past. I assume they are test edits, probably encouraged by some training stuff somewhere. I don't believe they (or most of them) are malformed searches. They are especially common on India-related articles, I find. Johnbod (talk) 14:17, 7 April 2023 (UTC)
Parallel guideline to WP:PLOTSOURCE, but for non-fiction texts
Articles on non-fiction texts, for example on philosophical works, generally cite their summaries of the texts' contents directly to the original text, rather than using secondary sources.
For fiction texts, there is a clear guideline, WP:PLOTSOURCE, that clarifies the acceptability of this custom. It seems as if there should be a parallel guideline for non-fiction texts. I am not advocating for what this should say, only suggesting that this gap be addressed.
Is there a problem to be solved here? Have you encountered any problems because of the perceived lack of this parallel guideline? I had no big issues getting the non-fiction A Voyage Round the World through FA despite (or perhaps because of) an extremely long "Content" section sourced only to the book... —Kusma (talk) 16:58, 11 April 2023 (UTC)
I've definitely encountered issues with this, and I've thought about making a post like this myself. Current guidelines are really unclear about best practices for non-fiction works. Even just a few sentences in a guideline somewhere would help. Thebiguglyalien (talk) 17:16, 11 April 2023 (UTC)
I'm not a fan of the idea. Fiction plots are generally pretty straightforward. Other than attributed quotes, what is said in a work of nonfiction needs discussion. If it isn't discussed in RS, maybe it's not worth including in the article. If it is, let's summarize what those RS are saying. Valereee (talk) 19:13, 11 April 2023 (UTC)
Plot summaries of works of non fiction should be limited in the same manner as works of fiction, but I also think that fictional and non fictional narratives that have received extensive, nearly scene by scene coverage, can go long in the tooth as long as that additional analysis is included, and the longer plot section helps to guide our reader through the analysis. Eg, the plot of any Shakespeare play or Kubrick film could be longer die to the extensive coverage of each work. Same would apply to Around the zworld... above. Masem (t) 20:38, 11 April 2023 (UTC)
Can you provide some examples? I would be surprised if primary-sourced summaries of philosophical works did not veer into OR. BilledMammal (talk) 01:58, 12 April 2023 (UTC)
There are many different kinds of non-fiction works, though. A narrative biography of a famous figure is vastly different from a scientific thesis, or an examination of art history, or a math book. Independent sources existing to demonstrate the notability of something like The Making of the English Landscape, it is fine to cite the work itself for a summary of its contents. BD2412T03:00, 12 April 2023 (UTC)
One tricky area… figuring out how to appropriately summarize and cite a notable non-fiction work that advocates for a WP:FRINGE theory or topic. It can be done, but it takes skill. Blueboar (talk) 13:17, 12 April 2023 (UTC)
Or even just a personal interpretation. I recently had a discussion at Talk:No, Ma'am, That's Not History about a summary of a nonfiction book. The summary had stated He presented many counter-arguments to Brodie's arguments, using logic to show inconsistency or casting doubt on her sources. He also criticized her psycho-historical method, stating that historians can't know what hidden emotions Joseph or Emma Smith felt unless they have a source that says so. I felt that this was the article creator's take on the nonfiction book's arguments and that we needed to find someone else saying these things. Valereee (talk) 13:30, 12 April 2023 (UTC)
it should be readily possible to summarize a non fiction work on a FRINGE position without giving any impression the fringe position is "right". As a poor example 2000 Mules doesn't give the work any time to brief speak for itself, which given the amount of criticism for the work, should only be about 3 to 4 sentences. The details if the presentation which are wrong/disproven/fringe can be discussed with sourcing as appropriate in a criticism section (though with what I am seeing at 2000 mules, there are a number of synth/OR insertions, which can't be included). But the amount of valid criticism that film got would post most of the details next to the reality counterpoints. Masem (t) 16:03, 12 April 2023 (UTC)
Embedding a Wikidata-Infobox within Wikipedia articles
Hi, the current method of Wikipedia for representing structured data of an article is by redirecting to its Wikidata item and doing a URL redirection from an HTML page to another HTML page. But this method is not consistent with Web 3.0. According to the article Semantic Web:
Specifications such as eRDF and RDFa allow arbitrary RDF data to be embedded in HTML pages.
So these structured data (from Wikidata) should be treated as metadata and should be embedded within HTML page of each article.
Hence, I have an idea for placing a template named "Template:EmbedWikidataInfobox" in all existing Wikipedia articles, that is by default hidden, but users can show that by clicking a button. This way all related structured data is embedded within that article as some metadata.
I should note that this new extracted and embedded "Wikidata Infobox" should have three columns to be consistent with the semantic triple format: one column for subject, one column for predicate, and one column for object. And in each row a structured data item should be placed.
This Wikidata-Infobox has benefits for both machines and for humans. For machines is discussed, but for humans a user can "rapidly" see what structured data "is not inserted" or "is not correct" and somehow refer to Wikidata to correct that data or add new not existing item.
Finally, I should note that services like ChatGPT and many other newly created bots certainly benefit from this embedded Wikidata Infobox to extract structured data more accurately, and to respond users more precisely. Thanks, Hooman Mallahzadeh (talk) 05:45, 10 April 2023 (UTC)
@WhatamIdoing Hi, implementing the Web 3.0 is like placing airbag in cars. Without an airbag, a car works perfectly without any problem, but in a special situation (an accident) this airbag is very vital and important. Here, without this metadata, Wikipedia articles work perfectly and users benefit from Wikipedia without any problem. But in the special situation that users would ask questions from a bot, Wikipedia does not work properly. So I really believe that for implementing something that is somehow ««standard»» we can use the word "should". Hooman Mallahzadeh (talk) 05:02, 11 April 2023 (UTC)
The actual standard says that this is permissible ("may"). It does not say that is recommended ("should").
Consider the difference between "You may drive a white car" and "You should drive a white car". You have gone from "The standard allows you to make a choice about whether to do this" all the way to "The standard recommends that you do this".
It's perfectly fine if you want to recommend that we do this, but it is incorrect to claim that the standard recommends it. The standard permits it. The standard says that you may do this. It does not recommend it or say that everyone should do this. WhatamIdoing (talk) 15:58, 13 April 2023 (UTC)
Organizing potentially another Wikipedia blackout over RESTRICT Act
In short: these bills would allow for US to shut down platforms that it deems to be a risk to national security. There is a lot of coverage about this in the news about a possible ban of TikTok; however, the way the bill is worded can be very very arbitrary, goes beyond TikTok, and can result in serious infringement on the same First Amendment rights that allow Wikipedia to exist in the first place.
Wikipedia successfully was able to protest the Stop Online Piracy Act back in 2011, but this may actually be something even bigger.
I want to raise this in idea lab before an RfC is made about this. Of course, any RfC to organize a blackout in the United States would require a very strong consensus. We would need to come to consensus on how long the blackout should be and what the blackout message should be.
This is relevant because Wikipedia is hosted in the United States and is subjected to US laws, and Wikipedia is committed to rights to freedom of speech. More than 150 million Americans use TikTok, which means that more than 150 million Americans could have their rights to freedom of speech restricted if such a ban was implemented, as well as dozens of small American businesses. As a project, Wikipedia is committed to uncensored, free, open speech. And as such, I believe it is crucial that that right remains open, and that social network platforms used for free speech, even those that may be considered "adverse", are not taken down.
That said, I want to figure out what the structure of such hypothetical RfC should be. A blackout should be considered a last measure, but it would be a very effective tool at getting a large number of Americans to understand what is at stake with this legislation. Aasim - Herrscher of Wikis ❄️ 05:23, 27 March 2023 (UTC)
If you want any hope of an RFC getting support, you need to laser focus on how the law would be an existential threat to Wikipedia, written with the same care and sourcing you might bring to a featured article. You're not going to be able to get enough support based on a general "Wikipedia is committed to uncensored, free, open speech" argument to overcome the "no politics ever" faction. Anomie⚔11:53, 27 March 2023 (UTC)
As a project, Wikipedia is committed to uncensored, free, open speech. - this is... not exactly true. I agree that Wikipedia doesn't care much about "national security", but we certainly exercise censorship of other kinds, for our own reasons. Wikipedia:Hate is disruptive talks about it a bit. 199.208.172.35 (talk) 15:01, 27 March 2023 (UTC)
I think a lawyer and/or WMF legal should give a proper analysis of this Act to determine whether this actually is a problem for Wikipedia. Even if it isn't, if it could permanently change the notion of freedom of speech in the US, as said in Verge, then it is something that WP should absolutely protest with a blackout.
Hi @Awesome Aasimthis is a really interesting discussion. I'm Ziski from the WMF Global Advocacy team. I can make sure you get the information you're after from our team. What would you like us to provide? Generally how we're monitoring the RESTRICT Act and our sense of how it does/doesn't threaten Wikipedia..or something else? Please let me now if you have specific questions and we'll work on getting a response to you. Cheers! ~ FPutz (WMF) (talk) 17:48, 6 April 2023 (UTC)
Hi @FPutz (WMF), I am wondering if WMF thinks that this bill can be potentially disruptive to Wikipedia and/or the open Internet it supports, and if this bill might have affects about as big as SOPA. Also any reliable secondary sources that might back up this idea. What you are saying about "Generally how we're monitoring the RESTRICT Act and our sense of how it does/doesn't threaten Wikipedia..or something else?" is kind of what I am at.
If there is, then it might be a catalyst for a blackout RfC. I haven't had time because of university to dive more into the impacts of this bill, but I do want to see what WMF thinks. :) Aasim - Herrscher of Wikis ❄️ 18:53, 6 April 2023 (UTC)
Thanks for clarifying! That shouldn't be a problem. I'll start putting together a response to those points and I'll dig around for some further readings in case you are interested. I hope to have it for you soon - good luck with Uni in the meantime. FPutz (WMF) (talk) 19:15, 13 April 2023 (UTC)
@Awesome Aasim see my comment at the end of this discussion!
@Thebiguglyalien are you talking about this? Right to free speech just means government can't do anything to block the free flow of information. Private entities like WMF can choose who can speak or not to speak; that is them exercising their right to freedom of speech by letting who they will listen to. Aasim - Herrscher of Wikis ❄️ 04:17, 29 March 2023 (UTC)
I'm talking about the fact that Wikipedia's mission is not to babysit free speech in the United States. It is to build an encyclopedia. Unless you have incredibly strong evidence that Wikipedia is going to be imminently banned, I'm going to assume that this is an attempt to disrupt the entire website over a political issue. Thebiguglyalien (talk) 04:53, 29 March 2023 (UTC)
Any blackout on its own will be disruptive at some level to readers. The consensus last time there was such a government overreaching potentially free speech violating law (SOPA) was that a blackout would be appropriate. I do think any measure done to protest RESTRICT should be done in a manner that is as least disruptive to the project as possible.
That could mean having a full or partial blackout, having a full page banner (which a user could click dismiss to), or running a banner campaign on users in the United States to get people to tell their representatives that RESTRICT can permanently fracture the open Internet. That can certainly be raised in the proposal.
As Anomie said, any such proposal has a high chance of success when written with the same rigor as featured articles, otherwise it would be considered a waste of time to most. Aasim - Herrscher of Wikis ❄️ 12:22, 29 March 2023 (UTC)
I support your proposal to protest the banning of online platforms, including TikTok. Even though I detest the despotic regime in China, I think there are other methods to deal with spying than banning foreign media. It's incredible that Congress is considering such authoritarian step, copying behavior of the Soviet Union and Eastern Germany. Thinker78(talk)04:24, 29 March 2023 (UTC)
I Oppose as the legislation does align with us in part. Some of the reasons for the ban (privacy, stopping espionage and threats to the individual, social media is not a social good ) do line up with the beliefs of some wikipedia editors, but others do not (removes a platform for free speech). I do think a company or government should be able to ban an app for their employees or contractors on any BYOD or provided device.
We monitor editors and readers [7], and we monitor donors intensively.
We have banned some admins for being government stooges, but I assume they had access to check user tools in their language wikipedias before the ban, and
the WMF work with Big Tech (which collect an enormous amount of information) on projects or creates apps that are part of the android/apple/ ... infrastructure. These apps are built on code libraries that could be leaking information.
Also will the legislation do anything? Or will it just become Whac-A-Mole with new apps or literal moles (espionage within big tech, inserting code into chips, creating surveys, buying card purchase and voter databases). Regardless of the legislation, the internet at all levels is monitored by the National Security Agency and equivalents. Wakelamp d[@-@]b (talk) 06:57, 7 April 2023 (UTC)
I support such a blackout, but freedom of speech really isn't the issue here though. It is still a massive issue for the internet though. It's how this affects the freedom of international communications on the internet. Under this law the US could theoretically enact bans on any internet services affiliated with a country deemed adversarial. While this shouldn't apply to Wikipedia in the most strict reading, neither would SOPA. This law could easily be manipulated by the US government to restrict practically any internet service that it chooses to.
It has an incredibly broad scope that could easily affect Wikipedia and the entire internet as a whole. This act is just as big a threat, if not bigger (due to its severe restrictions on foreign communications), to Wikipedia as SOPA was, and I believe that Wikipedia should take similar actions to protest the act as it did with SOPA. Serafart (talk) (contributions) 04:27, 29 March 2023 (UTC)
I believe strongly that this is a disastrous piece of legislation, and that it should absolutely not be passed. I believe it represents an existential threat not just to Wikipedia, but to the internet as we know it, and the legislation is also fundamentally in conflict with basic U.S. constitutional principles. I also think it is not the place for Wikipedia as an organization to make any political statement, no matter how noble. That's the first step on a very slippery slope that, while we've made the grave mistake of doing so at least once in the past, we should never do again. Or, more succinctly "Please, no." --Jayron3215:35, 30 March 2023 (UTC)
Look, if there's consensus to do so, it wouldn't the first time I would be on the wrong side of consensus, and it definitely won't be the last. My contribution towards developing a consensus one way or the other is, and I quote, in case you missed it earlier, "Please, no". I hope that clarifies the matter for you. --Jayron3217:13, 30 March 2023 (UTC)
@Blueboar it's a bit more complicated than that. While the content of our articles do not WP:RIGHTGREATWRONGS, as a platform we have participated in a number of advocacy, from the SOPA Blackout in 2012, other language editions Spanish and German Wikipedia participated in a blackout in 2018 over the EU Copyright Directive and of course WMF itself does a lot of lobbying for freedom of expression. If you think this specific issue or tactic is wrong venue, I respect that, but to say we're not politically active, feels imprecise to me.
I think the text/purpose needs to be reworked a bit, but I support something like an informative banner, and think reaching out to Meta:Main would be useful to hear their perspective, and then we can make an RfC here. ~ 🦝 Shushugah (he/him • talk) 18:03, 30 March 2023 (UTC)
I am aware of our past advocacy, and I disproved of it at the time. I still disapprove… Strongly. I feel that our responsibility to maintain a Neutral Point of View extends beyond just article content. Our job is to REPORT- not to participate. Blueboar (talk) 18:21, 30 March 2023 (UTC)
Wikipedia is an encyclopedia and its articles should be neutral but Wikipedia is also a project with its own set of needs in order to fulfill its objective. From time to time if those needs are imperiled, Wikipedia should raise its voice. Thinker78(talk)03:51, 31 March 2023 (UTC)
Wikipedia doesn't have a voice. It has neither vocal cords to speak, nor hands to sign, nor fingers to type. It is an abstract concept and a collection of code stored in computer servers. You have a voice, and if you feel strongly that your work on Wikipedia is imperiled, you should feel empowered to raise your voice by contacting your elected representatives and make your voice heard. --Jayron3211:44, 31 March 2023 (UTC)
Here are some more sources about this Act, related Acts, and a possible TikTok ban: [8][9][10][11][12]. While it is not our place to decide the fate of something unrelated to Wikipedia, it is our place to protest legislation that can be twisted in a manner to impede Wikipedia's goals of providing a free, accurate encyclopedia. Also please remember that this is not for straw polling or !voting but for developing this idea into something that might be ripe for an RfC. Aasim - Herrscher of Wikis ❄️ 14:12, 31 March 2023 (UTC)
It may be your place to do so. It may be other people's place to do so (insofar as they want to). That is the "we" that matters here, you (as an individual person) and other people (as individual people) who care and want to enact change. Wikipedia, as an organization, should take no public stance on anything. Instead, feel free to organize and use your own voice how you see fit, there's no need to involve Wikipedia in it. --Jayron3215:05, 31 March 2023 (UTC)
I respect the principle of neutrality and not activism of Wikipedia that you are advocating but I don't fully share it. Although I share your concern about politicization of Wikipedia, I think there is certain times Wikipedia should take a stand, specifically in legislation that could affect its ability to fulfill its objective. That is, to provide free encyclopedia knowledge to humankind.
Undue censorship of Wikipedia is always a problem. Imagine the US government blocking Wikipedia as the government of China does. The prospect of the government having the power to ban online websites it doesn't like actually imperils the very existence of Wikipedia. Regards, Thinker78(talk)05:21, 1 April 2023 (UTC)
Also, regarding Also please remember that this is not for straw polling or !voting but for developing this idea into something that might be ripe for an RfC It is also the place to say "this is a bad idea, don't do that". I'm saying that now. --Jayron3215:07, 31 March 2023 (UTC)
If Wikipedia the organization wants to do something, the corporation contact Congress. Wikipedia the site should just do our job. A blackout is the opposite of our job. --User:Khajidha (talk) (contributions) 23:03, 31 March 2023 (UTC)
Except there is actually some precedent for having Wikipedia do advocacy like this. If there wasn't I'd agree, but there is.
I think these could be the options for a hypothetical RESTRICT Act protest:
Full blackout - the entirety of Wikipedia would be inaccessible for the time of the blackout, with a notice that Wikipedia is protesting the controversial RESTRICT Act, similar to the SOPA protest
Partial blackout - only essential pages such as medical topics would be visible
Full-screen click-through banner - a notice about the Act would be visible to readers and readers would be able to act on it by contacting their local reps
Banner at top of page like in a similar manner to all the donation banners, but would be dismissable
Nothing if it is determined that this Act is not protest-worthy.
But whether it is protest-worthy or not is what the focus of this RfC would be. Aasim - Herrscher of Wikis ❄️ 17:41, 1 April 2023 (UTC) (edited 14:22, 7 April 2023 (UTC) to correct misfired act I keep mixing up)
I don't think it was precedent so much as a cautionary tale, warning us about what happens when we let editors think that political activism is tolerated on Wikipedia. Thebiguglyalien (talk) 00:33, 3 April 2023 (UTC)
Disagree with the premise of 'E', we can believe that the act is protest-worthy and still not think that it's appropriate to use Wikipedia as a vehicle for that protest. If we protested everything that was protest-worthy, we might finally get a default 'Dark-mode'! JeffUK11:11, 4 April 2023 (UTC)
Support banner, oppose blackout. I believe the Wikimedia Foundation, as a prominent organisation committed to principles of free information and freedom of speech has some moral duty to protest when its host country (i.e, the United States) proposes terrible legislation like the RESTRICT Act that has significant potential to affect the free exchange of information worlwide. There has been past precedents with this, like with our SOPA protest, which AFAIK was broadly supported. A banner would be a good idea. That being said, IMO blackouts should only be used if there is direct potential to adversely affect Wikipedia/Wikimedia. The RESTRICT Act doesn't quite past that test. Obviously not a lawyer, but my reading of the Act is it can't be applied to the WMF as it is incorporated in the United States and is therefore not a "Covered Entity" within the meaning of Sec. 2 Cl. (4)(b) of the Act, but the Foundation's chapters in Russia/Venezuela/HKSAR have potential to be affected as being subject to the jurisdiction of a foreign adversary under subclause (ii). Satellizer el Bridget(Talk)07:43, 4 April 2023 (UTC)
Oppose - SOPA, and the other legislation that has been protested in the past, directly affected Wikipedia's ability to carry out its core mission. This law is not directly related to Wikipedia, this advocacy against government policy will put Wikipedia in direct conflict with the US government by choice, and nothing good will come of politicising Wikipedia unnecessarily. JeffUK11:08, 4 April 2023 (UTC)
SOPA would have made the WikiMedia Foundation liable for detecting and removing potentially infringing content identified on Wikipedia, potentially having a chilling effect on what would be allowed to be discussed or posted here. Also, the SOPA blackout was instigated and organised by a third party, Wikipedia joined a large number of other web companies in the blackout, that's different from us coming up with the idea ourselves. JeffUK16:25, 4 April 2023 (UTC)
Alternative
As an alternative, perhaps we need a policy statement that says “WP should not engage in advocacy” (no matter what the cause, or how “worthy” we may think it is). Blueboar (talk) 18:08, 1 April 2023 (UTC)
I think that the only time WP should be involved is when there is a clear legal threat to WP's existence - RESTRICT does not do that directly, but the COPA/SOPA bills did. Masem (t) 18:30, 1 April 2023 (UTC)
I agree with this @Masem. This is where what Anomie's statement of how "[I'd] need to laser focus on how the law would be an existential threat to Wikipedia, written with the same care and sourcing [I] might bring to a featured article" would come into play. I was also thinking it would be helpful to get someone with an expertise in law to actually make this determination if in their opinion this RESTRICT Act is as bad as claims on the Internet are making it. That or a WP:RS legal opinion analyzing the effects on the Internet if this Act were to pass. Either of those two would be a basis for an RfC. Aasim - Herrscher of Wikis ❄️ 13:43, 2 April 2023 (UTC)
To be honest, I’m not sure we should be engaging in advocacy on WP - even when a proposed law might directly impact Wikipedia itself.
There are lots of on-line venues that are appropriate for advocacy by Wikipedians, but I don’t think WP itself is one of them. Blueboar (talk) 17:18, 2 April 2023 (UTC)
I know… and those of you who don’t are the most evil people on the planet. I would organize a day of blackout on WP to protest, but I don’t think WP is the correct venue for that. Blueboar (talk) 00:06, 3 April 2023 (UTC)
Oppose per WP:NOTBURO, etc. I also think that we shouldn't be enshrining such things in policy. While I personally oppose using Wikipedia as a vehicle of advocacy, I also oppose setting rules to stifle discussion. Who know, maybe I'll be convinced in the future to change my stance on this. I want to allow other the opportunity to convince me. I don't want rules that say "we don't talk about such things on Wikipedia". That's also bad. --Jayron3211:55, 3 April 2023 (UTC)
Why are people bold !voting against the guidance of this page? - instead, just discuss thoughts. I am almost the only person on en-wiki who has started a blackout RfC that did not either get the full majority (a single case) or an absolute hammering/snow close (the vast majority). In effect, the outcome of that was to show that even a significant possible threat (I realise all threats are inherently "possible") was viewed as insufficient by en-wiki (but not by some sister projects). It takes a particularly drastic state of affairs. That threshold means that an RfC along these lines would need to set out why such a thing is needed. Either we'd turn it down naturally, or some grave affairs are occurring. That catastrophic threat or issue might not be within the wording of the policy, so we might as well not bother creating a rule that inherently would only ever be actually of importance when it was overruled. Nosebagbear (talk) 12:17, 4 April 2023 (UTC)
I didn't read that part of the guidance and will try to be better next time, but I think it shows the strength of feeling that this is not something we should do. JeffUK08:57, 5 April 2023 (UTC)
I do not think adding an explicit statement that Wikipedia shall not engage in advocacy is good for this project. As Masem states above, there are certain policies that, if enacted, are detrimental to this project and we should at least consider taking action. --Enos733 (talk) 16:39, 4 April 2023 (UTC)
I agree, there are some things that are so egregrious or important whereby saying nothing is advocacy! JeffUK13:11, 5 April 2023 (UTC)
A terrible and dangerous potential law. But it appears that it doesn't directly affect Wikipedia and hypothetical application to Wikipedia looks pretty farfetched. IMO such advocacy is not our mission. Sincerely, North8000 (talk) 20:40, 5 April 2023 (UTC)
If I am not mistaken, Wikipedia:SOAPBOX is meant to prevent editors from engaging in excessive self-promotion in articles, talk pages, project pages, etc. where such promotion is in of itself unencyclopedic. It does not apply to editors discussing whether or not the Wikipedia community should formally approve or disapprove of a piece of legislation (as was done with WP:SOPA back in 2012) or some other action by some other government or non-government entity. Also this forum is about workshopping ideas together, not for consensus polling. That is for when the idea gets formally proposed through one of Wikipedia's procedures. Granted, it has to be related or even tangentially related to the encyclopedia and/or its future for the issue to even be raised. One can also invoke WP:IAR but only if there are absolutely very good reasons to do so, not to engage in disruptive bad-faith edits. Aasim - Herrscher of Wikis ❄️ 20:38, 6 April 2023 (UTC)
Wikipedia:What Wikipedia is not § Wikipedia is not a soapbox or means of promotion covers all forms of promotion, not just self-promotion. The first numbered point lists "advocacy, propaganda, or recruitment". It doesn't apply to this particular discussion, for example, but would apply to a change to mainspace, such as placing some kind of banner message. (The community can decide by consensus to take a certain action in spite of this guidance, of course.) Regarding the specific shortcut used: standing on a soapbox to give impromptu public speeches was historically related to propagating opinions, rather than direct self-promotion. isaacl (talk) 21:27, 6 April 2023 (UTC)
Discussing the RESTRICT Act's potential effect on Wikipedia (put aside the protest proposal for a moment)
Put aside the idea of a protest for now. What is the potential for harm to Wikipedia or Wikipedians here? Off the top of my head, I can think of a few, though I would appreciate any corrections. This is not something I have studied in depth:
Effect on sourcing. Wikipedia relies on a large number of foreign websites/platforms for sourcing and external links. The language of the act appears sufficiently broad that it could conceivably apply to a variety of information sources based outside the US, and not just e.g. social media. This would primarily affect non-English Wikipedias, but remember that a significant number of the people who edit other language Wikipedias do so from the US. In fact, there are some language Wikipedias written in large part by people outside the country. With the parameters for the use of the law so vague, it remains unclear how frequently this could be used, so it's worth considering a wide range of potential sources in countries the current or any future administration decides are a threat.
Effect of its vague language on Wikipedians. The section of the law "SPECIFIC UNLAWFUL ACTS" includes a wide range of broad categories of acts, e.g. No person may cause or aid, abet, counsel, command, induce, procure, permit, or approve the doing of any act prohibited by, or the omission of any act required by any regulation, order, direction, mitigation measure, prohibition, or other authorization or directive issued under, this Act. What does it mean to "aid" or "counsel" in this context? The language of the law makes it seem that any individual on Wikipedia could be engaged in punishable illegal activity for, say, discussing how to continue to access information on a site that has been banned.
Effect on other sites we rely on. We should consider that Wikipedia is part of a larger knowledge ecosystem. How does this affect organizations we rely on? For example, will the Internet Archive be considered to be aiding a foreign adversary if it retains archives of material that came from now-banned websites (or website/platforms banned sometime in the future)? What kind of burden would that put on them?
Perhaps most importantly, it's hard to say how this will be used. This is meant for TikTok, but the actual scope of the law means we need to consider consequences that may be difficult to predict. It puts the United States in the business of banning entire platforms and entire technologies used for a wide range of First Amendment-protected speech, but also allows that the President may take such action as the President considers appropriate to compel divestment of, or otherwise mitigate the risk. That leaves ... a lot on the table. It creates a general purpose bludgeon and concentrates the ability to wield it in two people. First, we have to designate a "foreign adversary". There are a few named in the bill, but all it takes is a single unelected cabinet member (the Secretary of Commerce) to say an entire country is now a "foreign adversary" for the purpose of this bill. That cabinet member may have been appointed for their judgment and competence, or may be an incompetent but loyal friend of a mercurial and paranoid president. The secretary then refers those adversaries to the president who can do ... basically anything in the course of divestment. It's a concern in the abstract, at very least, for the internet as a whole. We have seen that governments, including the US government, have been willing to take sweeping action against entire foreign states as well as foreign citizens based on individual events or threats, stretching vague legal language as far as possible. This is a blunt, ill-defined instruments, and for that reason it's hard to say how it might effect Wikipedia.
Adding: Prior to the SOPA blackout, then-General Counsel Geoffbrigham provided a helpful overview. Perhaps the current General Counsel (User:Slaporte_(WMF)) or someone else from WMF Legal might be willing to do something similar for would be willing to do something similar here? (Or maybe Geoff is interested enough to offer pointers in an unofficial capacity). — Rhododendritestalk \\ 21:19, 13 April 2023 (UTC)
Comments from Wikimedia Foundation Global Advocacy team
Hi all! It’s great to see USA public policy being discussed. I’m Ziski from the Foundation’s Global Advocacy team. I’m happy to share more about how our legal and public policy specialists view the biggest threats to the Wikimedia movement and people in the USA right now, and how we can work together (if you’re interested).
RESTRICT is very different from SOPA in the threat it may or may not pose to Wikimedia projects. One of the main differences is that SOPA would definitely have applied to the projects, and it was on the verge of being passed into law when the Wikimedia community joined a massive grassroots movement to oppose it. In contrast, RESTRICT is still far from whatever eventual final form it might take, if it does eventually get passed by both houses of Congress and then signed into law. It is very unclear at this time how or whether it would ultimately affect Wikimedia platforms, editors, or readers. We keep an eye on a lot of laws that are in this category of development (see final comments for how to stay up-to-date on updates from our team about such developments).
On principle we believe that wholesale banning or blocking of websites by governments violates the right to freedom of expression. For that reason, we have joined a coalition of civil society groups, academics, investors and companies warning against government efforts to ban and block apps and websites. If U.S. lawmakers want to protect internet users against abusive corporate surveillance, they should get serious about enacting a strong federal privacy law. The Foundation has long supported efforts to enact a federal privacy law. We have also worked hard to protect the movement against abusive government surveillance including in the U.S.: We sued the NSA and we are actively working with a coalition to lobby for surveillance reform.
The biggest threats to Wikimedia projects and people right now come from efforts to limit the scope of Section 230 of the Communications Decency Act in ways that would make it harder for Wikimedians to moderate content without the Foundation being constantly sued. Currently, Section 230 protects website hosts from lawsuits for the content submitted by users. If it were weakened or eradicated, then Wikimedia’s volunteer-run content governance model couldn’t function the way it does today. That’s why our main U.S. advocacy priority is to defend our S230 protections (which SOPA would have wrecked). S230 is under challenge in two Supreme Court cases this year (Gonzalez v. Google; Netchoice v. Paxton) and we are expecting some politically-motivated S230 reform bills to emerge later this Spring. A D.C.-based North America Public Policy Specialist has just joined our team and will be leading our work on these policy developments. As they pick-up, we may host conversation hours and more. Below is information on how to watch what we’re doing.
How to keep working with, or talking with WMF Global Advocacy about USA public policy
If you want to get in touch and talk about running campaigns or activities in the USA, or you want to organize a conversation hour, then email me: fputz[AT]wikimedia.org. You can also email the Global Advocacy team: globaladvocacy[AT]wikimedia.org.
=== Responses to/Questions for Wikimedia Foundation Global Advocacy team ===
~~ Your comment goes here
Making diff the first parameter in the diff template
{{diff}} has title as the first unnamed parameter, despite it being optional. The diff ID is also the most important parameter and most of the times you can just link to a diff with only the diff ID, so I think the diff ID should be the first parameter. Putting this here as the template talk page doesn't seem to have traction Aaron Liu (talk) 16:03, 12 April 2023 (UTC)
I don't disagree with you about which field is more important, but as a practical matter, {{diff}} is a widely used template. Making a change like this now would be quite disruptive. -- RoySmith(talk)16:21, 12 April 2023 (UTC)
That was what I was thinking about. We could use a bot or autowikibrowser to do this, maybe like move the template to "/old" or something, add in the fixed version, and replace all mentions of it with the newer syntax. Aaron Liu (talk) 17:53, 12 April 2023 (UTC)
That would fix the template for future revisions of a page, but when you view an old revision of a page, the current code of the template is executed, so all the 30k+ uses in several millions of old revisions will be broken. —CX Zoom[he/him](let's talk • {C•X})07:16, 15 April 2023 (UTC)