User:Acroterion/PG

Response
This may get a bit lengthy and episodic, since it touches on several issues or editing trends that I've notice that are wrapped up together in this topic, and this is as good an opportunity as any to set them forth.
 * There has been a growing trend toward jumping on virtually everything that appears in the news at the moment it's published without waiting to see if it's of lasting significance or even real. While we used to see WP:NOTNEWS quoted a lot, it seems to have been ignored in recent times as the news cycle has contracted and as social media have become major sources of content. Some technically -oriented breaking news, like for instance the Fukushima Daiichi nuclear disaster is handled extremely well by Wikipedia. Other things, like social-media-fueled rumors such as the topic at hand, are not handled so well and descend into BLP violation magnets in short order. Since it's no longer an option to sit down and write a new article on Blue (color), those who want to create new content gravitate toward the thing of the moment and write whatever they can find on the Internet. This trend circumvents both journalistic fact checking and fact-checking by other Wikipedia editors through flooding a potentially obscure and lightly-watched topic with new editors and agenda-driven accounts. Wikipedia really is set up to deal with fact-checking on the color blue, and is terrible at what amounts to rumor.
 * Lots of new editors appear on topics like this with the notion that "the free encyclopedia" means you can post any damned thing you want, or that they're entitled to free speech on a private website, or that it's sourced if you just provide a link to something somebody said somewhere on the Internet. The concept that this is an encyclopedia project rather than a platform for free expression, axe-grinding, making shit up, or just general mischief gets lost. It's also a by-product of Wikipedia's success as a widely-regarded source of information, whose word on something is considered a matter of record. This is not news, but it goes with the crowdsourcing model - the result is only as good as the crowd. Sometimes the crowd that is missing the point of the project is louder, more numerous or more determined than the editors who are trying to stick with the project and its rules.
 * This all came to a head with GamerGate, where a flood of new or inexperienced agenda-driven editors flooded in to document an evolving event that was being driven by social media and influenced by people who were at best unconcerned with the harm they might be doing and at worst actively interested in causing harm, all under the veneer of WP policies like NPOV adjusted to individual preferences in information sources. The difference between Reddit and the chan boards and Wikipedia is (somewhat) stronger governance, which seems perplexing or frustrating to the new arrivals, who are inclined to argue. Most who have been here for a while find the topic of Wikipedia "governance" risible, but it exists to a certain extent.
 * A similar trend is happening with the Comet Ping Pong topic.
 * The disambiguation has become a target because of the nature of disambiguation pages. They don't lend themselves to nuanced presentation, sourcing and such. It's easy to swing the one-line description to the summary of one's choice. Some of the same issues are endemic to infoboxes, by the way. They are also at a slight distance from the parent topic, which is usually more closely watched.
 * If there was just one "Pizzagate" we could just redirect and full-protect the redirect, full stop. Honestly, I think the old Pizzagate fails the test of time as a redirect - sources mention soup too, so should it be "Soupgate" in the article? Sourcing for the specific term w/r/t the Battle of the Buffet seems a bit scanty, and the BotB seems to be the primary title. Turning the disambiguation into a protected redirect to Comet Ping Pong is an option, but maybe not a good one. I suspect that the new Pizzagate will also fade once a newer, shinier conspiracy theory appears and the amateur investigators on Reddit turn elsewhere. It bears examination, though.
 * Now for the "false," "debunked" and so on. I believe that we should take the shortest path to a concise statement that summarizes the subject in a manner supported by standard WP RS and verifiability policies. This is usually not so hard in an article bodies where a paragraph or two can set forth the issue and its sources. Where it often fails is in ledes, infoboxes or disambiguation where it must necessarily be compressed, while at the same time not leave out the critical point that something is a fringe belief or just plain false. Time and distance help - there is no need to state that the earth is not flat in a single concise line, because it's widely understood to be an outlying notion. We can just say "Flat Earth - the belief that the Earth is flat" rather than "the false, erroneous and badly mistaken notion that the Earth is flat, resting on the backs of an infinite number of turtles, who are themselves fictitious." Where this fails is when there is a rumor in the air, and people are looking for material on this novel idea without really knowing in the back of their minds that the Earth is not in fact flat.
 * There is an obvious precedent on Wikipedia for a positive assertion of the falsity of a fabricated document: The Protocols of the Elders of Zion. It has the benefit of hindsight and scholarship, and is a good example of how whispering campaigns were handled before we could manufacture them electronically. It's a frequent target for drive-by attacks by anti-Semites, but it endures (the last time I checked) because there's a broad consensus for a positive statement of falsity, and because those who insist that it be given plausibility through omission are transparent in their motivations.
 * I strongly believe that when a rumor whose existence (as opposed to its substance) appears to cross the bar of GNG is reported (see below for that business) it must be made clear from the beginning that it's either doubtful or just plain wrong. Where the rumors are allegations of misconduct, criminal activity or depravity by living people it must either be flatly removed or an integrated statement of its doubtfulness or falsity be included. Newsworthiness is not truth nor is it necessarily verifiable as policy in general or BLP in particular demand. I'm aware of the whole verifiability-not-truth argument. A rumor's existence as a verifiable phenomenon is separate from the verifiability of its content, and a line must be drawn between the two.
 * The place this comes up most often is in a place where space is at a premium or in a prominent place. It's less contentious where there is a wall of text in which it can be given context or ignored, depending on the reader's point of view.
 * Finally getting to Koncorde's concerns. I don't view "false" in a summary as an infallible shield against BLP issues, rather the reverse. The issue arises from a conflict between those who view anything that passes the GNG bar as trumping its omission over those who express BLP concerns. That is the tension here, and perhaps an amendment of GNG or a strengthening of BLP is in order. In my view it's better gone, but the wrangling from those who insist that if it can be sourced (and here the Times and Post sources can be used by notability partisans) keeps it there, or it pops up again accompanied by an earnest argument and/or accusations.
 * Attempts to remove this sort of thing by administrative fiat tend to end badly once somebody can slap a reliable source on it. We are reduced to whacking new and inexperienced editors attracted to the rumor with sticks while they earnestly quote their version of policy at us, and then it happens again, and again ...

NPOV vs BLP

 * WP:NPOV contains the unfortunately-phrased line "This policy is non-negotiable, and the principles upon which it is based cannot be superseded by other policies or guidelines, nor by editor consensus." This implies that one's chosen NPOV is immune to BLP and consensus. We know that's not what it means, but I've seen this quoted to support BLP violations and POV editing in recent times. Answering such an objection with the fifth pillar, IAR, leads to trouble, even though that's what most people dowithout explicitly invoking IAR. There was an RFC to remove it that failed a few weeks ago in which there was some sentiment for rewording, but nothing came of it. The page is written for newcomers' understanding, or at least should be, and it shouldn't need ten years of absorbing Wikipedia's editing ethos to parse NPOV correctly.

Overlap with enthusiasm for conspiracy theories

 * This kind of topic attracts editors who tend to be interested in general in what might be termed alt-media, and who are either interested in or are actual partisans in favor of conspiracy theories. This is another area where Wikipedia fails: topics that aren't current news, but which attract fringe POVs that either find conspiracy theories fascinating in their own right, or which have appeal to those who are convinced that there's something up and that great truths can be revealed on Wikipedia's convenient platform.
 * Adherents of conspiracy theories can't be pigeonholed into specific political categories, but politics plays a role and CT enthusiasts tend to congregate at either end of the political spectrum.
 * Because CT enthusiasts find the mainstream medium either complicit or unhelpful, they view Wikipedia as a mutable, widely consumed publication-of-record where they might be taken seriously, or where they can wrangle the rules to get a hearing.
 * As with more current topics, peripheral articles and disambiguations sometimes receive as much attention as the parent. Texas School Book Depository gets regular drive-bys from Kennedy assassination enthusiasts who place "alleged" before "Kennedy assassin Lee Harvey Oswald," often never appearing at the parent article at all.