Talk:Gamergate (harassment campaign)/Archive 25

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia
Archive 20 Archive 23 Archive 24 Archive 25 Archive 26 Archive 27 Archive 30

Don't use disambiguation titles as actual names

The title of the article has an added "controversy" to disambiguate it from gamergate. It's a basic dab title, like English language, which makes perfect sense. It makes no sense bolding it, though. This is usually referred to simply as "Gamergate" by almost everyone. "Gamergate controversy" is a description, not an actual name.

Peter Isotalo 22:51, 30 January 2015 (UTC)

That isn't the only reason. Another reason was that the only noteworthy part about Gamergate is the surrounding controversy - almost the entire article is about that. 109.152.100.135 (talk) 23:18, 30 January 2015 (UTC)
Or rather in the early days of the article people fought hard to move the goalposts in order to keep out things they didn't like. Rhoark (talk) 23:28, 30 January 2015 (UTC)
The IP user is correct. It used to be "Gamergate (controversy)", the consensus was to remove the brackets. The problem was that the article was about Gamergate, but the only coverage was some controversy around it. There was very little discussing the movement per se. So it was decided to make it about the controversy. HalfHat 23:33, 30 January 2015 (UTC)
Brackets don't really matter here. Again: English language. There are plenty of similar examples. Like Springfield, Massachusetts.
I'm not exactly sure how you're reasoning here. Focus is still on the notable aspects (controversy, not movement). News articles refer to this as "Gamergate", not "the Gamergate controversy". If the latter, it's descriptions like "the Gamergate death threats" or "the Gamergate media coverage". Or why not "the Gamergate movement"?
Peter Isotalo 00:03, 31 January 2015 (UTC)
I've put arguments (though not forced the issue) that we could really make this article about the movement since it is really about them and the actions they have caused (directly or indirectly) and the criticism that resulted. The term is used enough in sources (equal or more to "gamergate controversy"). However, there is a hesitation there because while sources will say that GG is a self-described movement, they will follow that up claiming they aren't a movement due to their lack of organization, etc. As such, calling it a movement in the prose in the past has been controversial by some editors. But I will argue that nearly every source recognizes that it is a self-described movement even if they criticize it harshly in the same article. --MASEM (t) 00:34, 31 January 2015 (UTC)
any coverage of the "so called movement" is entirely secondary to the coverage of the harassment. -- TRPoD aka The Red Pen of Doom 00:44, 31 January 2015 (UTC)
This is strictly an issue of wording and WP:COMMONNAME to me. The scope of the article is quite neatly defined by the actual article title.
Peter Isotalo 11:46, 31 January 2015 (UTC)

History

"Quinn began to receive hate mail over the game upon its release and criticism from some parts of the Steam user community, receiving enough harassment to cause her to change her phone number. This elicited further outrage from others and by September 2014,"

  • What elicited further outrage and by whom? Changing her phone number? I'm sure that's not what the fuss is all about, but it reads that way. This sentence is poorly constructed. --Two kinds of porkMakin'Bacon 12:34, 31 January 2015 (UTC)
I removed it. It was a pretty fuzzy statement and it's unclear why it needed to be mentioned in a paragraph about the plight of Quinn.
Peter Isotalo 20:17, 31 January 2015 (UTC)
I believe the trailing phrase was attached to a statement that noted that DQ was about to be released when Robin Williams committed suicide, and Quinn opted to continue the release commenting on the issue of depression. That would lead to additional harassment as left over. --MASEM (t) 21:57, 31 January 2015 (UTC)

Appeal of Protection Level

The current protection level on this page is being appealed at Arbitration Enforcement for being in violation of the current Protection Policy. Those Editors who wish to make a comment on the appeal may do so there. --Obsidi (talk) 22:03, 31 January 2015 (UTC)

4chan largely coordinated the attacks

This is directly stated in the Washington Post. Are there other sources that state otherwise? If not, to imply others had more influence is OR. Regardless it should be noted that 4chan is mostly responsible for getting this harrassment ball rolling.--Two kinds of porkMakin'Bacon 12:56, 31 January 2015 (UTC)

Gamergate has been banned on 4chan since mid-september, while august attacks may have been coordinated by 4chan Gamergate has been going on for 4 months since then. See this washington post article for details on more recent Gamergate coordination.Bosstopher (talk) 13:12, 31 January 2015 (UTC)
I don't think this changes that the organization started on 4chan, though. We should clarify that they were banned from 4chan, surely, but it definitely started there.192.249.47.186 (talk) 18:49, 31 January 2015 (UTC)
@Bosstopher: that article only supports the fact that 8chan made attacks. Absent a source being provided that contradicts the first source stating that 4chan was the stated and was largely responsible for organizing the attacks, I intend to restore the text you reverted.Two kinds of porkMakin'Bacon 19:55, 31 January 2015 (UTC)
@Two kinds of pork: if you did it would be a major case of WP:RECENTISM (but a past version, there should be a better way of phrasing it). None of the sources since mid September have mentioned 4chan being involved at all (unless referring to the beginning of the controversy, and instead focus on the involvement of 8chan and other forums. Since that was only a one example i'll give you the other sources too. Here are articles noting KiA, Escapist forums and 8chan are the new hubs of GG since 4chan shut it down.[1][2][3][4][5][6] also pretty much all articles about Brianna Wu mention that the doxxing happened on 8chan not 4chan, and if you read the wiki article on 8chan the sources used there also point out that it's now a central hub of GG activity instead of 4chan. I'd suggest if you want to include it in the lede, use the phrasing the IP editor suggested noting it started in 4chan. Bosstopher (talk) 22:25, 31 January 2015 (UTC)

I think the lede is wrong

I'm pretty sure "Gamergate" is described in the RSes as a movement or a campaign. I get this article is focused on the controversy around it, since the actual structure and workings or the movement/campaign gained little attention, but the current lede is just misleading. HalfHat 23:41, 30 January 2015 (UTC)

I see above there being discussion of it up there and I'm not sure if this was linked to a change so I made this a separate section, if in fact it was, please merge this in. HalfHat 23:45, 30 January 2015 (UTC)
As a "movement" its non notable. If you want to call it a "harassment campaign" I am good with that being supported by the sources. -- TRPoD aka The Red Pen of Doom 00:42, 31 January 2015 (UTC)
12,500 google news hits pretty much proves the movement is notable. It's just also so tied to the controversy that it does not make sense to consider it a separate topic. --MASEM (t) 00:46, 31 January 2015 (UTC)
WP:GHITS and there is zero evidence that it is notable outside of being a harassment movement. -- TRPoD aka The Red Pen of Doom 13:13, 31 January 2015 (UTC)
The sources in the article (which are what we have to go by) are ambiguous; some describe it as an amorphous movement, some note that it describes itself as a movement but qualify that, and some describe it purely as a controversy or purely in relation to the hashtag. I think the simplest thing for us to do is to focus on people relative to the hashtag in the lead, since that is universal among sources -- any other characterization should probably be qualified by citing it to someone (eg. so-and-so describes it as such-and-such.) --Aquillion (talk) 01:02, 1 February 2015 (UTC)

Arbcom article

Mentions of the wikipedia edit warring have been appearing in news articles about GamerGate for a while, but usually just in passing. Although more focused articles about Jimbo's role exist [7] [8]

Today Guardian has published a piece on the Arbcom case, which i think given that events on wikipedia are very much part of the Gamergate controversy avoids MOS:SELFREF. But there are two problems. The first one is that the article is incredibly inaccurate. First of all its based only off of the original version of the Proposed decision, and doesnt note any of the changes made. It claims bans have already been put in place, this is incorrect. It claims 5 anti-GamerGate editors were proposed topic bans regarding all feminism topics, this is incorrect only 4 were, (and only 3 have had the topic ban pass). While Masem who claims to be anti-GG was proposed a topic ban later, it did not mention prohibiting edits to Also given that the article is pretty much just based off of a blog post by MarkBernstein who's accused Masem of running an evil cabal of 8channers, it seems unlikely the author was referring to Masem. So given how much of a wreck the article is and how much of it is just plain wrong, is it useable?

The second problem is: if it is usuable who's allowed to put information from it into the article? Would it mean that everyone who took part in the Arbcom case would have to keep WP:COS in mind when editing anything referring to it? Would it just be parties to the case? Bosstopher (talk) 17:00, 23 January 2015 (UTC)

Do we have a reliable source saying that The Guardian is inaccurate? If The Guardian writes about it, with their history of Pulitzer prizes, reliability, and editorial oversight, than I am pretty sure we should report it exactly as Guardian says. Obviously we have to mention this is the Guardian's opinion, and cite it as such, unless you think that's WP:UNDUE. I'm not completely certain. In all seriousness though its probably just undue unless we have more articles writing about the ArbCom case in particular. I vote for passing the buck. Ries42 (talk) 17:06, 23 January 2015 (UTC)
We know the Guardian article is wrong because the ArbCom case is still open and all remedies have yet to be voted on completely. Trending towards the conclusions made, yes, but not there yet and some remedies can still flip on votes. Even given that, there's still not a heck of a lot appropriate to the GG topic here from these articles - a topic about the criticism of Wikipedia in general, possibly, but I yet to see any real point of substance of what WP's role has been in GG to date. --MASEM (t) 17:09, 23 January 2015 (UTC)
We are under no obligation to use reliable sources that still get things wrong. If the Guardian chooses to publish something asserting the moon is made of cheese, we wouldn't include that in an article either. Thargor Orlando (talk) 17:17, 23 January 2015 (UTC)
I believe everyone here missed the fact that Ries42 was being sarcastic. — Strongjam (talk) 17:20, 23 January 2015 (UTC)
Sorry. I thought the "In all seriousness" afterword would have indicated that. I was indeed being cheeky. Ries42 (talk) 17:22, 23 January 2015 (UTC)
My apologies. To be fair, it's been impossible to tell as of late here. Thargor Orlando (talk) 17:28, 23 January 2015 (UTC)
I don't think the arbcom really merits inclusion in the article based on one source (which printed prematurely.) Argument might be made for including something about Jimbo based on multiple sources, but I'm not convinced. Seems like a minor detail at the moment. — Strongjam (talk) 17:09, 23 January 2015 (UTC)
The arbcom case is definitely not a minor detail. Copulative (talk) 18:04, 23 January 2015 (UTC)
At this point, no aspect of how Wikipedia has handled GG - short of Jimmy Wales' statements and even then to a small degree - has affected the GG situation, or at least what has been reported in RS. The case is important to WP, but from the topic of GG, it has little immediate relevance. --MASEM (t) 18:07, 23 January 2015 (UTC)
I would disagree with you, considering how much controversy the Wikipedia article itself has generated. Copulative (talk) 18:10, 23 January 2015 (UTC)
I know in GG circles there is a lot of controversy, and here, I'll just point to the archive page count, but we have nothing from reliable sources that say that WP's article has had any role in the GG controversy. If anything, the stuff about Jimmy talking about the GG's try at their own version of a GG article, that's all about criticism of Wikipedia that involves the GG issue, but not a topic of the GG controversy. --MASEM (t) 18:13, 23 January 2015 (UTC)
What about this portion from the The Guardian article? "The conflict on the site began almost alongside Gamergate" Copulative (talk) 18:18, 23 January 2015 (UTC)
How has the actions we have done here in the long-running debates and through ArbCom have had any direct influence on the GG controversy/movement? The only thing I think that even approaches a point of use is an explanation of how this and other GG-related articles have seen an influx of people likely siding with GG to try to change the way these articles are written, as part of the activities of the GG group -- but there's zero sources for that at all. To GG, this is a non-story, but it is a possible story in discussing WP in the media or criticism of WP. --MASEM (t) 18:24, 23 January 2015 (UTC)
It has direct influence because when someone doesn't know what Gamergate is, they're going to Google it and the first thing that pops up will be a Wikipedia article called "Gamergate controversy." Copulative (talk) 18:27, 23 January 2015 (UTC)
Yes, that's definitely the issue the GG side has with WP, that's very clear, and that's why we've had lots of new editors trying to participate. But this is not at all documented in any reliable sources. --MASEM (t) 18:29, 23 January 2015 (UTC)
As I noted on the Arb page, the article is essentially correct; all they whiffed on was quoting Mark Bernstein saying "No sanctions at all were proposed against any of Gamergate’s warriors, save for a few disposable accounts", as there was only one; TDA. — Preceding unsigned comment added by Tarc (talkcontribs) 17:34, 23 January 2015‎
A diff of Tarc's comment for the interested. — Strongjam (talk) 17:37, 23 January 2015 (UTC)
I'm not seeing any big problems in article accuracy, sadly. Artw (talk) 18:18, 23 January 2015 (UTC)
I literally listed all the factual errors in the article, at the top of the section. There's a substantial number even excluding the Bernstein quote.Bosstopher (talk) 18:51, 23 January 2015 (UTC)
I'm seeing those as pretty minor quibbles or subjective on your part, TBH. Artw (talk) 20:01, 23 January 2015 (UTC)
Anyone who didn't take part in the ArbCom case can add info from the article without breaking Wikipedia rules. Copulative (talk) 18:08, 23 January 2015 (UTC)
It's really sad a source as respected as The Gaurdian published something that poor. Anyway it seems undue to go into any of the specifics anyway while there is only one source. HalfHat 18:19, 23 January 2015 (UTC)
  • You know what, I actually think this article is very good for something. Specifically, its factual errors point out that its author may not be deserving of the amount of weight this article has given him, and it seriously brings into question his reliability. Currently the Draft has three articles written by Mr. Hern. I believe there may not be anything directly questionable about those articles in a quick review, but perhaps they are not as rock-solid reliable as previous discussions have led us to believe. Ries42 (talk) 19:26, 23 January 2015 (UTC)

I wrote the following in a separate section before I noticed this discussion:

Some people are saying it's inaccurate but I think that, nitpicking aside, this is as fair an account of the political dimension of the case as I could expect.

This one-off article probably doesn't merit mention in our article yet, because in the grand scheme of things Wikipedia hasn't been a big part of the controversy. That may change if many reliable sources start discussing the arbitration case. We'll cross that bridge when we come to it. --TS 19:39, 23 January 2015 (UTC)

It's not an accurate news piece at all. If you have particular feelings about Gamergate and this article, you might think the whole thing is being handled in a fairly one-sided manner, but as far as I can tell they're handing out various bans, admonishments, or other punishments to both sides pretty readily. I think Halfhat is on the topic ban list, the Devil's Advocate is on the ban list, Loganmac is there, and other users have had their existing bans confirmed. Anti-GG editors seem to be getting sanctioned, but there's politicking about it - especially for Ryulong. It looks like the article is being written based on Mark Bernstein's statements - and correct me if I'm wrong, but wasn't he involved with this whole mess? It's like if the Guardian did an article after an interview with Ryulong or NorthBySouthBaranof, which honestly would have made more sense because those two have done a lot of editing to the article.
My guess? It's politics. They're sounding the word that the article needs more anti-GG editors because a few of the current ones may get banned. Emphasis on the "may" part. YellowSandals (talk) 22:52, 23 January 2015 (UTC)

I think it goes without saying that this article is extremely insulting to the arbitrators, who are volunteers, not Chief Justices, and have worked very hard to pursue a just resolution. If anything this is good evidence that newspaper sources are untrustworthy, and that this article won't really be decent until a greater number of neutral, uninvolved accounts are available. Shii (tock) 01:31, 24 January 2015 (UTC)

Here's an article about it from the beloved Gawker: http://internet.gawker.com/wikipedia-purged-a-group-of-feminist-editors-because-of-1681463331/+cushac Copulative (talk) 03:26, 24 January 2015 (UTC)

If enough sources report on it, we'll have to cover it here eventually, but for now I think it'd be best to wait until the ArbCom decision is finalized, at least (which should produce most of the coverage if there's going to be more) -- it doesn't seem like something pressingly relevant or high-profile enough that we have to rush to add it on these few sources, anyway. --Aquillion (talk) 04:13, 24 January 2015 (UTC)

I haven't been keeping up with the drama, but I just read the news and it's disappointing to hear that a determined group of trolls is pushing away a good group of editors. Sorry I don't have a solution to propose, it's just sad to hear. --Frybread (talk) 06:27, 24 January 2015 (UTC)

Please do not call arbitrators "trolls." They have had to deal with this case for several weeks and are trying to find the best solution to multiple groups of editors whose actions might be deemed problematic. This case isn't a win for anyone, so I will apologize if that sounds snippy. To get back on track, do you have an opinion as to what we should do with the article? --Super Goku V (talk) 06:48, 24 January 2015 (UTC)
It's not the Arbitors he's calling trolls, it's certain editors involved in the dispute. Just to clarify for him. HalfHat 18:42, 24 January 2015 (UTC)
Whose to say he wasn't. And whose to say he might not be right?Two kinds of porkMakin'Bacon 07:14, 28 January 2015 (UTC)

My opinion is the following: I do not believe that the article should be included as a source at the present time. It appears to make several mistakes: Implies that the case is over, implies that there was a preliminary decision, states that arbcom has sanctioned anyone, etc. It fails to disclose that Mark Bernstein has an indefinite topic ban on discussion and edits to Gamergate related articles. I feel that the author of the article should have confirmed a few things before this could be a reliable article. --Super Goku V (talk) 07:10, 24 January 2015 (UTC)

I'm in total agreement with this statement by Super Goku V. We can dismiss even reliable sources articles on a case by case basis and this is one of them. The article has several factual errors and sources statements from an editor who has been topic-banned for his involvement in the dispute. There is little doubt in my mind that at this point this Wikipedia article itself, ArbCom, and the history of this article have become a part of the controversy, but this source is not accurate enough for inclusion about it. Reporting about a controversy over a Wikipedia article on the article itself is also a tricky situation that I think would require more community oversight before it is attempted, if a factual reliable source comes up about it. Weedwacker (talk) 09:49, 24 January 2015 (UTC)
"Wikipedia does not then further investigate whether a source provides "evidence" — we are encyclopedia editors, not investigative reporters. If reliable sources say something is true, for our purposes it is true." So which is it? Do we have to check if a source is telling the truth before using it or do we not? Or do we decide that based on which narrative we're pushing now? Akesgeroth (talk) 23:25, 24 January 2015 (UTC)
You are conflating two different issues. On this topic, we have only a single reliable source thus far presented and it is reasonable to discuss whether we should write an encyclopedia section based upon only a sole source, or whether we should wait to find out if other reliable sources chime in on the issue and what they have to say. We don't have any reliable sources saying that that source is wrong, but we don't have a consensus of reliable sources to say it's right, either. The more sources we can cite on a particular issue, the more likely we are to avoid issues of undue weight and the more likely we are to appropriately reflect the prevailing mainstream viewpoint of an issue. On the other hand, the issue you cite in your post is not a matter of a sole source, but of a clearly-established and overwhelming consensus of multiple reliable sources. NorthBySouthBaranof (talk) 23:32, 24 January 2015 (UTC)
So if a lie is repeated often enough, it becomes the truth? Akesgeroth (talk) 23:39, 24 January 2015 (UTC)
We are not truth finders. — Strongjam (talk) 23:43, 24 January 2015 (UTC)
And we are not obligated to put falsehoods in the article simply because traditionally reliable sources decide to promote them, either. Thargor Orlando (talk) 23:51, 24 January 2015 (UTC)
And the inclusion of any material is subject to editorial discretion, and using a single source for anything this controversial is not best practices at any rate, thus I agree that we should avoid including this issue until we have a wider array of reliable sources from which to support an encyclopedic section. NorthBySouthBaranof (talk) 23:53, 24 January 2015 (UTC)
So if you have any reliable sources to support your claim that something is a lie, then you'd present them here. That you haven't suggests that you don't have any reliable sources to support your claim, and that it is little more than your unsupported personal opinion. Wikipedia is not a platform for personal opinion, nor is it a platform to right great wrongs. I believe that you have a deep-seated and good-faith belief that it is a lie, but that's not how we write encyclopedia articles. It just isn't. NorthBySouthBaranof (talk) 23:44, 24 January 2015 (UTC)
NBSB is correct. Even if normally reliable sources are wrong, Wikipedia operates on verifiability and not truth. If sources continue to get the facts wrong and the notability of the ArbCom decision in relation to this article rises enough, we will have to include the factually incorrect information in this article. Lignos (talk) 13:50, 26 January 2015 (UTC)

IT KEEPS HAPPENING. [9] Bosstopher (talk) 00:25, 25 January 2015 (UTC)

Seriously, this is funny. I think at this rate the GG ArbCom case will become so notable, that we'll have to report on it. And then... wait, do we have to post information that is factually not true because its been reported on so much. I mean, this is The Guardian we're talking about. Even if we KNOW its wrong, for instance, the Arbitrators themselves can say it was factually wrong, are our hands tied? Ries42 (talk) — Preceding undated comment added 02:18, 25 January 2015 (UTC)
Given that the Guardian article utilizes only one source on the GamerGate article, that being a person with multiple bans re: the GamerGate controversy, it would almost be pro-GG to use the article itself as a "reliable source". Indeed, it almost comes across as a means to inject that person's PoV back into the article by way of getting a news organization to quote him about it. Calbeck (talk) 02:44, 25 January 2015 (UTC)
Pro GG in what sense exactly? Mr. Bernstien, the quoted source, was not Pro GG. If you mean "it proves the Pro GG narative," well, that doesn't really mean anything. Plus, if The Guardian felt it was reliable, we are supposed to not question that. This falls out in one of three ways. 1) Its not notable enough, and thus, no inclusion. 2) It gains enough notability and it is included as the Guardian says. WP:VNT takes precedence, where despite some things not being completely "correct" we report it as The Guardian sees it. 3) We agree that the falsities in the article outweigh its notability, and we challenge the reliability of it. Most likely the target of the unreliability would be the author, not The Guardian itself. Ries42 (talk) 04:08, 25 January 2015 (UTC)
In the sense that histrionics tend to actually go against a person in the public arena. Bernstein's claim that this decision is "worse than a crime" is an example of something that would tend to set an otherwise neutral reader wondering why such hyperbole was in play. I concur, however, that challenging unreliable sources should go to including sources used by otherwise reliable sources, such as the Guardian as an institution.Calbeck (talk) 02:31, 29 January 2015 (UTC)

Has there ever been a precedent where the press is so obscenely lazy, political, and unwilling to do research on a subject that there was no way to write to accurate article? For example, imagine James Cook comes back from Tahiti and and a scholar on his boat says that everyone in Tahiti is beautiful and willing to have romance with any European. Afterward everyone began reporting on that and spreading that lie because it was juicy and sexy, and for years people thought Tahiti was some island of nymphs. Is it possible that maybe a lot of the news doesn't actually care that much about video games or video game culture and they just want to report on the juicy stuff? We've had people both pro and anti-Gamergate say they were harassed and sent death threats, but the press has only talked about the women - we can't really report anti-Gamergate harassers because it's only being discussed in social media and not on the BBC. Maybe there runs a possibility a lot of sources are just off-handedly reporting a sexy narrative that's vaguely based on reality, but as with the stories of Tahiti, have much more mundane, banal origins. YellowSandals (talk) 07:06, 25 January 2015 (UTC)

A thing to keep in mind: all the articles that I've seen about the ArbCom decision all extend/source directly the Guardian article and then subsequently MB's blog posts. It is effectively the same single article, just in the telephone-game of slight variations from message to message. Let's not yet throw the baby out with the bathwater though we should be aware that the article and its small mutations are creating a stir on the social media that we might have to deal with. --MASEM (t) 07:13, 25 January 2015 (UTC)

I am not "involved", nor am I a regular Wikipedia editor (I was active in some specialized and unrelated articles about a decade ago, but I have completely lost the login details and can't recover them as I no longer have the associated email address). I have, however, been closely following the issues associated with this article, and have often been stunned at the disconnect between what was actually happening, and what was reported (the "GamerGate is dead/over/old news" drumbeat every week was particularly amusing). Now, I realize that Wikipedia has to use reliable sources, but it is a delicious irony that the same style of reporting (using only one biased viewpoint for sources, and presenting their allegations as facts without further research) that has characterized the entire episode has now, in a beautifully meta moment, wrapped around to the Wikipedia article about the whole mess and the ArbCom case it spawned. Now, for all that the overwhelming majority of RS being of a particular tone and presenting a particular group of facts, you may want to consider, how many of those "facts" were independently derived, and how many of those sources were sourced from each other. 150.167.144.14 (talk) 04:24, 27 January 2015 (UTC)

Even if the information in the article is 100% accurate (I haven't read the article, and I haven't cared enough about this issue in several months for it to be fair for me to comment) and even if we view the source as reliable, we still can't use it in this article because doing so would amount to our article indirectly citing itself (albeit through an article quoting a participant in an arbitration regarding the Wikipedia article citing). I forget the acronym, but I seem to remember it being a rule here that we can't cite reflexively. If it isn't, it should be. Quodfui (talk) 14:26, 25 January 2015 (UTC)

I'd also just like to point out that all of this goes to the point I made October 23rd. We're being used by the Guardian as a primary source in a controversy we're ostensibly reporting on. Quodfui (talk) 14:26, 25 January 2015 (UTC)
That said, I think the Guardian article is important and should be included on Wikipedia, even if not here. Is there an article on discussion of Wikipedia in the media? Quodfui (talk) 14:29, 25 January 2015 (UTC)
As the Guardian article is factually inaccurate, I don't see it getting any play here currently. Thargor Orlando (talk) 14:34, 25 January 2015 (UTC)
The article reflects very real concerns about the effect of the ArbComm on Wilipedia and the claims of falsehood are massively overblown. Artw (talk) 16:08, 25 January 2015 (UTC)
And as those concerns are based on very real falsehoods, including who is getting banned and the attitude of Wikipedia toward specific points of view, it's not worth discussing at this point. Thargor Orlando (talk) 16:19, 25 January 2015 (UTC)
Pretty much every sentence about the Arbcom decision contains some factual error. The headline of the article is "Wikipedia bans five editors from gender-related articles " which is factually inaccurate in that only 3 (possibly 4) are being banned, and from articles on gender related disputes not on gender altogether. This means the 3 (not 5) editors have not been banned from making corrections to feminism articles. Also none of the editors have actually been banned yet. Then quoting Mark Bernstein it says only throwaway GG accounts were sanctioned, which is incorrect, and that there will be no feminists left editing the article, which is also incorrect because I consider myself to be a feminist (although Bernstein admits this is by his own rough count). Everything from that point onwards about wiki-politics being super toxic seems fair enough. While whats written written may sound vaguely true all the specifics of fact are completely wrong. Bosstopher (talk) 17:12, 25 January 2015 (UTC)
Argh...please read all of what I wrote before responding to part of it. I do not think the Guardian article should be used for a source here, and I said as much above. I have no reason to doubt that the Guardian article is factually inaccurate, but factual inaccuracy isn't in and of itself a reason not to use a source - this article is on the #Gamergate controversy and the vehemence of the debate here and the article's wide use as a source elsewhere on the internet say it has become part of the controversy. We can't use it because its content derives from the article we'd be citing it on (WP:CIRCULAR). I think it could be included in a different article on controversies regarding Wikipedia's editing policies. I'm sure such an article exists. I think I've read it before. Quodfui (talk) 20:33, 25 January 2015 (UTC)
Sorry that was meant as a response to Artw's claims of factual innacuracy being overblown. Bosstopher (talk) 21:05, 25 January 2015 (UTC)
Assuming that the case itself becomes notable enough to spawn an article, the place you're probably thinking of is the Criticism page, on a new subpage thereof like other past incidents that became notable. If consensus is that it should be referenced here somehow, we could link to the specific new criticism subpage page from here to avoid WP:SELFREF issues. Sappow (talk) 06:06, 28 January 2015 (UTC)
Ah, I knew there was a better spot. The list of controversies is where you want to be. There's already an entry for this event there, and I'm sure there will be a fair number of additional reliable articles about it whenever the ArbCom case closes. Sappow (talk) 06:58, 28 January 2015 (UTC)

So I've heard it go back and forth between not using the source because it's factually inaccurate, but then there are some voices saying that factual inaccuracy does not mean we should not use the source. So I have 2 questions in regards to this.

1. If we use sources that we know to be factually in accurate, what does that say about WP? Is it just a reflection of popular opinion or potentially even just propaganda? I know that WP doesn't consider the "truth" of sources. But if we know an article to not be true, then don't we have a responsibility to avoid using that material? All of the guidelines of WP:RS AND WP:BLP are in regards to material being used in WP. But does that mean we can't use facts to avoid using overly falsified or misleading articles? In other words, we can't use facts to including unreliable material to an article, but can and should we use facts to exclude seemingly reliable source material from an article?

2. Reliability is based on accuracy, honesty and/or achievement. Does this not mean that factual inaccuracies detract from reliability? If an article is factually inaccurate, would that not mean, at the very least, that this article is not reliable and therefore not a proper source of material for WP?

I do realize though that there are some times when facts don't matter in a source, but it seems like those are usually just going to be opinion pieces or other special circumstances. And I guess there's some question to the autonomy of editors too. For WP not being a bureaucracy it sure seems very bureaucraty... Thanks.TyTyMang (talk) 04:25, 26 January 2015 (UTC)

This is a fair enough argument and we should trim down our use of the Guardian on this specific topic. Shii (tock) 16:18, 27 January 2015 (UTC)

The article should not be included, even after the proceedings are done. Wikipedia's stance on this is that the decisions have nothing to do with wikipedia taking a side in the gamergate issue. Since it has nothing to do with it, we can't now say it DOES have something to do with it by including it. The article is a bit of a trap anyway. If we DON'T include it, some people will take that to be confirmation that we have a bias. If we DO include it, we could be seen as using Wikipedia as its own source, as the article references a wikipedia editor. It's circular. As a result, leaving it out of this page and placing it (if at all) in the criticisms page is the most intellectually honest thing we can do. Vygramul 01:15, 1 February 2015 (UTC)

Survey

Wikipedia, and to a lesser extent the Arbcom decision are not part of the Gamergate controversy and don't belong in the article at this time.


I hesitate to stick my head above the parapet. But it seems to me that this is what everyone has said. The Guardian and Mary Sue and others have decided that ArbCom (and hence Wikipedia) are part of GamerGate - and the nasty "misogynist" GamerGate at that, not the nice "ethics in gaming" GamerGate. And of course the dominant narrative is either that the nice "ethics in gaming" GamerGate doesn't exist (it's part of the nasty Gamer Gate", pretending to be nice - just like Wikipedia and ArbCom), that it's vanishingly small, or (per User:Jimbo Wales) that they should "leave GamerGate" and start something troll free (Good luck with that!).

All the best: Rich Farmbrough16:26, 29 January 2015 (UTC).

Eh, I don't think they're making that claim, more just that the ArbCom (and absolutely not "hence Wikipedia") buried its head in the sand and allegedly valued politeness over the quality of the wiki or protecting its editors from harassment. The articles aren't advocating that Wikipedia is like, plotting on KiA or anything, more just that its overseers tried to irresponsibly sweep things under the rug.192.249.47.186 (talk) 17:11, 29 January 2015 (UTC)

1RR self-report

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.


I was the one who reverted Altenmann's pending change mentioned above. I reinstated the Intel 2020 date change right afterwards, though.[10]

I also reverted[11] an edit that I viewed as a rather unnecessary skewing of wording soon afterward. I realized moments later that this could technically be a violation of the active discretionary sanctions. I've never edited an article with these kinds of restrictions, so I'm not really familiar with how strictly 1RR is interpreted. Does it literally mean one revert of any kind per 24-hour period? Does it apply to pending changes?

Peter Isotalo 01:46, 1 February 2015 (UTC)

@Peter Isotalo:. Yes it applies to pending changes as well. Anything that isn't BLP or vandalism. Probably best if you self-reverted yourself. I don't see those changes sticking anyway as violations of WP:SAY. —Strongjam (talk) 01:51, 1 February 2015 (UTC)
How about I accept revert of my text as a non-controversial, and Peter is off-the-hook? -M.Altenmann >t 01:53, 1 February 2015 (UTC)
Not an admin, but if everyone is reasonably happy I'm sure admins are reasonable people. — Strongjam (talk) 01:56, 1 February 2015 (UTC)
Shadowrunner has already edited the relevant section so I think it's up to that user to decide whether to re-add the previous wording or not.
Peter Isotalo 02:02, 1 February 2015 (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.

Cherrypicking

I think this long discussion ran its course in productivity. I'm started a alternative related proposal. starship.paint ~ ¡Olé! 06:33, 1 February 2015 (UTC)
The following discussion has been closed. Please do not modify it.

This finding of principle has unanimous support in the proposed ArbCom resolution: "The contents of source materials must be presented accurately and fairly. By quoting from or citing to a source, an editor represents that the quoted or cited material fairly and accurately reflects or summarizes the contents and meaning of the original source, and that it is not being misleadingly or unfairly excerpted out of context. Failure to accurately reflect sources, whether by accident or design, is a serious matter as it undermines the integrity of the encyclopedia."

We presently cite this article: http://www.newsweek.com/gamergate-about-media-ethics-or-harassing-women-harassment-data-show-279736

to support this statement: "Newsweek/Brandwatch performed an analysis of about 25% of two million Twitter messages with the Gamergate tag from September 1, 2014 onward, and suggested from the results that "contrary to its stated goal, GamerGate spends more time tweeting negatively at game developers than at game journalists".

Reading the article, one finds they based this conclusion on tweets directed at a whopping six twitter accounts. It was also unable to determine the sentiment of more than 90% of tweets in the sample. This analysis is obviously utter garbage, and falls under the remit of my post directly above in that reliable publishers are not reliable when making gross errors. Failing that, its rampant cherry-picking and against the above ArbCom finding to cite the conclusion of this farcical analysis without mentioning how the analysis directly contradicts itself - no synthesis required on WP's part. Rhoark (talk) 22:46, 27 January 2015 (UTC)

That the algorithm used was not, on its own, able to automatically categorize 90% of the tweets without further human analysis does not imply that the overall analysis is "obviously utter garbage". You are taking lines from the study grossly out of context.192.249.47.186 (talk) 22:49, 27 January 2015 (UTC)
The context is they have basically no idea who is tweeting what at whom. They don't know which tweets come from Gamergate. They don't know if they were positive or negative. They don't know who was tweeted at outside of 5 individuals and 1 corporation. The context is that they have no clue, and any citation of their conclusion without that fact is out of context. Rhoark (talk) 22:57, 27 January 2015 (UTC)
The fact remains that it is a cited, attributed, noteworthy analysis. We could poke holes in it all day, but, in this case, I don't think there's a problem with how we're treating it in the article. This aspect of the analysis (and we call it a study in the article elsewhere and that's misleading, and I'll change that) is getting one sentence and seems like the proper weight and play. Thargor Orlando (talk) 23:05, 27 January 2015 (UTC)
Again: the algorithm did not know who was tweeting what at whom. The analysis went on from there, and followed standard methods of assessing relevance from a statistical perspective. If you would like to constructively refute them, and add to the RS against their result, perhaps you could help author a peer-reviewed refutation of their method.192.249.47.186 (talk) 23:11, 27 January 2015 (UTC)
That would likely be original research and thus inappropriate. The Newsweek methodology is sound, statistically speaking, even if we can quibble with some of the choices. Thargor Orlando (talk) 00:02, 28 January 2015 (UTC)
The process of editorial decisions, like determining a claim does not reflect the content of its citation, is outside the scope of page-content policies like WP:OR. Rhoark (talk) 02:58, 28 January 2015 (UTC)
I feel like people must be reading an entirely different article than me. They said what the algorithm spat out, showing their big grey bars with slivers of red and green on the ends. Then they said it means Gamergate is bad. That's it. No analysis, no statistics. Definitely not anything like a p-value. Unless there's some kind of link to a Part 2 that I'm missing, they do nothing of the kind. Rhoark (talk) 02:35, 28 January 2015 (UTC)
Hm, you actually have a point I wasn't considering here. What we have here is factually true, but also misleading based on the data. You're not wrong. I'd love to get some more feedback on this from other observers. Thargor Orlando (talk) 02:58, 28 January 2015 (UTC)
Currently the observations this article is cited for are not in fact "Gamergate is Bad". They're the pretty uncontroversial points that examination reveals a lot of new accounts taking part in the discussion, and that game developers are being tweeted at much more often than journalists. Those are points based on the data gathered that aren't influence by the whole grey bars neutral or unclassified confusion. Additionally the original post on this topic is incorrect. The 6 twitter accounts displayed in the graphs were not the only ones examined, they were the 6 most mentioned. The "gross error" that supposedly invalidates its use as a source is an imagined one. Lord Lion Lad 03:14, 28 January 2015 (UTC)
You might be thinking of the medium piece that examined tweets. This says nothing about new accounts. It says nothing about how the 6 accounts were selected. Even if they are the top recipients, by the numbers given they account for about 5% of their sample of 500,000 #gamergate tweets. That enables them to draw precisely zero valid conclusions, not limited to the one we cite them for, "GamerGate spends more time tweeting negatively at game developers than at game journalists". (Edit: dropped a 0, its 5% rather than .5%. conclusions unaffected) Rhoark (talk) 03:37, 28 January 2015 (UTC)
The article is cited twice. Once for each of the claims I mentioned. Taking another look it does seem like the attached claim is nowhere to be found in that particular article, though I do remember it being a Brandwatch where I first saw that info presented. Maybe in some kind of additional graphs left out of the main article? I can't seem to find anything on short notice though. May be something worth looking into all on its own.
Outside it being cited for something it doesn't seem to say though, none of the things you've brought up are enough to discount it being used as a source. It still doesn't contradict itself anywhere that I can see and the conclusion it's cited for in the Wiki article seem to be pretty self-evident from the pure numbers. It doesn't really matter how small a proportion of direct @ mentions make up the tweet numbers, there were still far more directed at one group than the other. This isn't as clear-cut a case of inaccuracy as something like the Guardian's Arbcom coverage. Until we have a reliable source rebutting it other than you, there's no reason to remove it. Lord Lion Lad 04:06, 28 January 2015 (UTC)
I open a bag of M&M's and remove 3. Two are yellow, and one is brown. Do they contain peanuts? Rhoark (talk) 04:24, 28 January 2015 (UTC)
I'm finding it hard to understand exactly what you're trying to say here, Rhoark. Try again? PeterTheFourth (talk) 07:13, 28 January 2015 (UTC)
Can you tell, based on the color of 3 M&Ms from a bag, whether or not it is a bag of peanut M&Ms? Can you do that? You might take advantage of the fact that different flavors of M&Ms have different color distributions, but there would still be absolutely no way to confidently answer this question. A sample of just 3 M&Ms does not have statistical significance to tell you what the color distribution of the bag is. I'm asking anyone who wants to say they feel the article's statistical methods are valid to actually look at the (lack of) statistical methods they are talking about. OR does not apply to editorial decisions. You don't need OR anyway - you can just look at the graph that demonstrates in a very visual manner that they don't know what the tweets are. They could be cited for saying Zoe Quinn was tweeted at more than Nathan Grayson. It can be cited for saying their algorithm was able to find more negative than positive tweets. To cite it for saying developers were tweeted at more than journalists, or that there were more negative than positive tweets, without also mentioning its statistical failures, is to misleadingly excerpt the article. Rhoark (talk) 13:34, 28 January 2015 (UTC)
So thinking about this more, do we just introduce a more direct analysis into the article? Perhaps going with something better wordsmithed along the lines of "and suggested from the analysis of tweets showing a positive or negative tone, 'GamerGate spends more time tweeting negatively at game developers than at game journalists'. Thargor Orlando (talk) 14:11, 28 January 2015 (UTC)
To avoid being misleading, the level of uncertainty needs to be explicit. I suggest, "A Brandwatch analysis of 500,000 #gamergate tweets beginning September 1 found significantly more tweets were personally directed at Zoe Quinn than at Nathan Grayson or Stephen Tolito. More still were sent to the official Kotaku account or "Gamers are Dead" author Leigh Alexander. Receiving the most individual mentions were Anita Sarkeesian and Brianna Wu, though all of the above represented less than 5% of #gamergate tweets. Algorithms were able to determine the sentiment of fewer than 10% of these tweets." Rhoark (talk) 14:36, 28 January 2015 (UTC)
Or more concisely, "Based on Twitter mentions of seven selected accounts for which automated algorithms were able to determine a sentiment, constituting about 0.125% of #gamergate tweets over a period during September 2014, Newsweek concluded that Gamergate tweeted negatively at game developers more than at journalists." Rhoark (talk) 14:54, 28 January 2015 (UTC)
I understand what you're attempting to do here, but the first option is too much and the second is far too caveat-heavy. What we have is one sentence in the article about this specific point from the analysis, is there a way we can tweak that sentence to be more accurate? Thargor Orlando (talk) 15:14, 28 January 2015 (UTC)
The sentence has a lot of caveats because the analysis has a lot of caveats. To leave it out is misleading. Rhoark (talk) 16:20, 28 January 2015 (UTC)

Clearly Wikipedia editors attempting to use their own analysis to discredit a reliable source that has a decades long history of analyzing popular culture and who hired a professional social media analysis firm is PRECISELY the the type of disruptive behavior the ArbCom has noted is not acceptable. Drop the stick or you will be the next to go. -- TRPoD aka The Red Pen of Doom 12:58, 28 January 2015 (UTC)

Determining the reliability of sources is a core function of a Wikipedia editor. No publisher becomes so reliable they get an automatic free pass. What is disruptive behavior is using threats to try to shut down discussion aimed at improving the article. Didn't ArbCom just personally admonish you about that? Rhoark (talk) 13:38, 28 January 2015 (UTC)
@TheRedPenOfDoom: this sort of response is more disruptive than analyzing whether or not to use a source. This is a civil conversation that may result in some positive improvements, so if you believe this analysis to be worthwhile as handled, a better tack might be to explain why you support it as used as opposed to using a better explanation. Thargor Orlando (talk) 13:53, 28 January 2015 (UTC)
If you want to keep going down this path, do not say you were not warned. -- TRPoD aka The Red Pen of Doom 14:03, 28 January 2015 (UTC)
If you think that the analysis, as offered, is worthwhile, please share why. Discussion about the article is what we do on talk pages to build consensus. Thargor Orlando (talk) 14:08, 28 January 2015 (UTC)
The "analysis" as offered is not worthwhile as it inappropriately substitutes Wikipedia editor's opinions over the analysis of a reliably published source and a professional social media analytics firm. Please identify any manner in which the article content is misrepresenting the clear analysis presented in the source. -- TRPoD aka The Red Pen of Doom 14:41, 28 January 2015 (UTC)
I'd start with what I posted above. Would you have a problem with that adjusted language? If so, why? Thargor Orlando (talk) 15:05, 28 January 2015 (UTC)
Does this count as a reliable source? An Actual Statistical Analysis of #GamerGate? UPDATED by CaineJW It debunks the findings of the Newsweek / Brandwatch "study" fairly effectively -- to be specific, the Brandwatch study did not say what Newsweek claims it said. KiTA (talk) 14:55, 28 January 2015 (UTC)
That would be an excellent source of great value to the article, except that Medium is a self-publishing platform. Rhoark (talk) 14:57, 28 January 2015 (UTC)
A quick google search of site:en.wikipedia.org "medium.com" suggests that it is used as a reliable source on many, many other articles. KiTA (talk) 15:00, 28 January 2015 (UTC)
Then we need to clean those articles up, as most usages of Medium don't fall within our policies. Thargor Orlando (talk) 15:05, 28 January 2015 (UTC)

To note, I originally had problems with the Newsweek article but as tRPoD has pointed out and as I came to recognize, to counterargument that does require SYNTH and OR we cannot include. As such, we have to accept that this is what the editors that analyzed the tweets and wrote the article concluded. But I will also note that that sentence in our article was written by me (or at least its start), and that it properly puts their conclusion (which they did state in with "seemingly"-type language) as just their conclusion, and not as an absolute fact, thus keeping what is included simply stating "this is their study, this is what they concluded", and letting the reader determine if the study is good or not. --MASEM (t) 14:57, 28 January 2015 (UTC)

There is a way to word it, however, to note that the distinction is made about those with a clear tone in either direction. Are you opposed to that language? Thargor Orlando (talk) 15:05, 28 January 2015 (UTC)
I find it impossible to parse this section on cherrypicking. I'd say that cherrypicking is when you dig deep into an article to find a sentence that supports what you want it to say, as opposed to reading the first sentence that defines the conclusion and use that to summarize the source. In this case, it appears people want to use the source to say "Using an algorithm that looks for positive and negative words, BrandWatch found most tweets were neutral in sentiment," which appears approximately 9/10ths down the page as opposed to "an analysis by Newsweek found that Twitter users tweeting the hashtag #GamerGate direct negative tweets at critics of the gaming world more than they do at the journalists whose coverage they supposedly want scrutinized." which is the first statement summarizing the study made. Is that accurate? Hipocrite (talk) 15:12, 28 January 2015 (UTC)
Essentially, the analysis shows that more negative than positive tweets occurred, but more neutral tweets occurred than either positive or negative, and by a large margin. Our sentence on the article instead suggests that the analysis is only showing more negative than positive tweets, which is misleading but not totally inaccurate. Thus my suggestion that we reword it to ensure that we're talking about the tweets with a specific tone. Thargor Orlando (talk) 15:16, 28 January 2015 (UTC)
I guess I'm confused, because when I read the article, I go to the conclusion (which is at the top of inverted pyramid article writing - which is what almost all news-analysis articles are). Could you explain why you are focused on the deep body, and accusing others of cherrypicking, if that's what you're doing? If you're doing something else, could you explain what you're doing, in a section titled appropriately, as opposed to "cherrypicking," which implies that other people are finding single sentences deep in the body of the article to cast doubt on the conclusion? Hipocrite (talk) 15:24, 28 January 2015 (UTC)
"Data scientists refer to these tweets as 'undetermined' because the algorithm did not classify the mention as either negative or positive." Neutral is not the right word to use here, as out-of-context it means something they did not intend for it to mean. — Strongjam (talk) 15:19, 28 January 2015 (UTC)
I believe your quote is below the bottom of the article - I'd call it 11/10 of the way through, correct? Why are you digging for that quote and accusing others of cherrypicking by using a quote from the introduction? Hipocrite (talk) 15:24, 28 January 2015 (UTC)
I'm just pointing out that calling the bulk of the tweets "neutral" would be cherry picking. Not suggesting we add that quote to the article. Quote from the intro is fine with me. — Strongjam (talk) 15:28, 28 January 2015 (UTC)
Would it? I mean, is it really cherrypicking to see the graph, see mostly undetermined tweets, and then wonder if saying "more negative than positive" is really the most accurate reading? Maybe we shouldn't be using this article for this claim if the intent is what I'm seeing here. Thargor Orlando (talk) 15:30, 28 January 2015 (UTC)
Newsweek is a reliable source. You appear to be using the graph to reach a conclusion not stated in the article. I think that's a very exciting novel work that you've created, and I'd be excited to consider including it as soon as you get in published in a reliable source. Hipocrite (talk) 15:36, 28 January 2015 (UTC)
"Essentially, the analysis shows that more negative than positive tweets occurred, but more neutral tweets occurred than either positive or negative, and by a large margin." -- this is absolutely false. If you read the analysis, these "neutral" tweets are actually tweets unclassified by the algorithm -- basically, they had an automatic sorter, and the sorter said "you're going to have to do these ones by hand". Classifying them as "neutral" is a gross failure to understand the classifications of the analysis.192.249.47.186 (talk) 15:59, 28 January 2015 (UTC)
No one is saying Newsweek is not reliable, nor is anyone reaching a conclusion that isn't in the article. As editors, we need to use discretion in what sources we use and how we use them. If you believe my claim to be novel, please explain why as opposed to this sort of response that brings more heat than light. Thargor Orlando (talk) 16:02, 28 January 2015 (UTC)
"I mean, is it really cherrypicking to see the graph, see mostly undetermined tweets, and then wonder if saying "more negative than positive" is really the most accurate reading?" -- yes, it is definitively cherrypicking. You're choosing to base an analysis on the graph alone, setting aside the rest of the study that explains what the graph means, and the newsweek article that provides further context. That's pretty much the definition of cherrypicking, as given in definition 1 and 1a. To be frank, you're also choosing to set aside basic methods of statistical analysis, and how they're used. I would strongly advise that you leave opinions on statistics to actual statisticians, because making false implications by misunderstanding the significance of statistics is a huge issue for lay people. Again: if one wants to dispute the findings of the brandwatch analysis, or its coverage by Newsweek, I strongly advise them to author a published, peer-reviewed rebuttal so that we can cite it.192.249.47.186 (talk) 16:29, 28 January 2015 (UTC)
@Masem If we wanted a counter-analysis that would be OR. Including more detail from the source to avoid being misleading is not OR. It is not synthesis to notice the axis markings on a chart. Enabling the reader to decide is precisely the goal. Deciding editorially that it is misleading at present is outside the scope of OR.
@Thargor It doesn't even show that there were more negative than positive tweets. It just shows the algorithm was able to classify more negative than positive tweets.
@Hipocrite Where something appears on the page should not be a dominant indicator. If I presented a source with the headline "Most women worldwide support GamerGate" and supplied a pie chart showing 10,000 support and 4 billion were not polled I'm sure people would have a problem with that. Just taking the headline can still be cherrypicking. Rhoark (talk) 16:20, 28 January 2015 (UTC)
Rhoark, I don't think that presenting an article that would never exist is really telling me much about things - if such an article were written, (and ignoring that headlines are separate from articles, so let's assume the headline is repeated in the body), I'd support this article using it, and not engaging in a novel work about why the article was wrong. The reason position in the article is relevant is because of the inverted-pyramid structure of news articles - the important stuff goes at the top, the filler below. The author highlighted the main conclusions - as we should - not an irrelevancy about process. Hipocrite (talk) 16:28, 28 January 2015 (UTC)
I had my doubts on this article; I asked at the OR board if a reasonable argument that any person skilled in statistics could demonstrate would be OR and yes it was considered that. We cannot state, in WP's words, highlight the fact the bulk of the tweets were neutral if this is not what the study concluded. The sentence as I have written is fully fair game to include, as well as its conclusion as sourced. --MASEM (t) 16:24, 28 January 2015 (UTC)
Can you link that discussion? My point is different from that, in any regard. Thargor Orlando (talk) 16:26, 28 January 2015 (UTC)
[12]. --MASEM (t) 16:36, 28 January 2015 (UTC)
Reading this now, the question/consensus is about whether Newsweek is reliable. Of course they are. I see no consensus there regarding OR claims or even any significant addressing of the issue. Thargor Orlando (talk) 16:47, 28 January 2015 (UTC)
As we're attributing the analysis appropriately, what they're saying isn't entirely wrong. This is like arguing that political polls shouldn't be used since they use a small sample. The issue is not the methodology, it is how we're presenting the conclusion. The conclusion, as we have it in the article, is "more negative tweets than positive tweets." The conclusion of the analysis is "more negative tweets than positive tweets out of those that could be determined." My desire is for the latter to be what's reflected in the article, as it matches up with the study. If we can't do that, we shouldn't have a mention of that aspect in the article at all. Thargor Orlando (talk) 16:26, 28 January 2015 (UTC)
But as you notice, the discuss does say, effectively, we cannot throw doubt on the results as they report in WP's voice otherwise that's OR. I know there's bloggers that went through and did their analysis but these are effectively SPS so unusable for us. We certainly could add about the determination of tone of tweets, that's fine, but we have to be careful in going too far beyond that. --MASEM (t) 16:50, 28 January 2015 (UTC)
The article doesn't say that. Hipocrite (talk) 16:28, 28 January 2015 (UTC)
It absolutely does: From a measurement of "the sentiment of GamerGate tweets...using an algorithm that looks for positive and negative words, BrandWatch found most tweets were neutral in sentiment. 'If our algorithm doesn't identify a tweet as positive or negative, it categorizes it as neutral,' a Brandwatch representative told Newsweek. 'Data scientists refer to these tweets as 'undetermined' because the algorithm did not classify the mention as either negative or positive.'" The conclusion we've pulled from the data is instead simply what Newsweek has opted to use from what was determined, rather than what the data says. Thargor Orlando (talk) 16:45, 28 January 2015 (UTC)
The source does not say "more negative tweets than positive tweets out of those that could be determined." It says "more negative tweets than positive tweets." It also says "If our algorithm doesn't identify a tweet as positive or negative, it categorizes it as neutral." It also says "Data scientists refer to these tweets as 'undetermined'." It also says "most tweets were neutral in sentiment." One statements was given prominent weight in the source. Three of them were buried at the bottom. You want to take the prominent statement, stich it together with the buried statement to reach a conclusion not in the source. You are creating a novel statement that is not in the piece you are citing. This is WP:OR, and a violation of WP:UNDUE. Hipocrite (talk) 16:49, 28 January 2015 (UTC)
Hipocrite, I have pasted exactly the part that is of importance to the section. I do not want to create a novel statement, but rather use the relevant statement. Note that most tweets had an undetermined tone, but that the ones that could be determined were more negative than positive. Why are you opposed to that? Thargor Orlando (talk) 17:01, 28 January 2015 (UTC)
I oppose all policy violations. The source states what it believes to be important up in the summary of the source - that more tweets were negative than positive . It is not our job to second guess our sources. I'm stepping out of this circular discussion now, as you're obviously never going to change your mind, and I'm entirely unimpressed by your argument. Hipocrite (talk) 17:04, 28 January 2015 (UTC)
So you'll be okay with pulling the comment entirely, as it's a NPOV violation? Thargor Orlando (talk) 17:29, 28 January 2015 (UTC)
We are opposed to the statement because the tweets of undetermined tone are as irrelevant as the 75% of tweets that weren't sampled, or the tweets pre-9/1 and post-10/24, or tweets that didn't include the hashtag "#GamerGate". There are a near-infinite number of statements that could be made about data that was not included in the analysis of their results, but we're not going to include them because they are irrelevant. Woodroar (talk) 17:24, 28 January 2015 (UTC)
The entire article is about the tweets. Are you saying it isn't misleading to say that, when most tweets are undetermined, "more" do X than Y? Maybe this is a reason to simply remove that sentence. It's literally one sentence. Thargor Orlando (talk) 17:29, 28 January 2015 (UTC)
The article states that undetermined tweets are not included in their final analysis, so any mention of them by us is misleading. It is equally as misleading as adding a ridiculous statement like "Newsweek and Brandwatch also found no positive tweets prior to Sept. 1st". We know that is factually true by looking at their data samples?—or, in this case, their lack of data samples—but it is a fundamental misrepresentation of the source. Woodroar (talk) 18:38, 28 January 2015 (UTC)
Right, which is why I think we, if we use this sentence at all, clarify that the tweets that could be determined are what were measured. Right now, we don't do that, and it provides a misleading sentence as a result. Thargor Orlando (talk) 18:42, 28 January 2015 (UTC)
But it's obvious that they are only ever going to base their analysis on tweets that can be determined. Mentioning that some tweets were undetermined only introduces ambiguity and confusion when Newsweek and Brandwatch are unambiguous and clear in how they present their interpretation. Woodroar (talk) 19:10, 28 January 2015 (UTC)
Then what value is this to the article at all? Thargor Orlando (talk) 19:16, 28 January 2015 (UTC)
I'm not sure I understand your question. It's a reliable source with findings relevant to the article. Why wouldn't we include it? Woodroar (talk) 19:29, 28 January 2015 (UTC)
If clarity and specificity are not valuable and relevant to this information for the article, why have the sentence at all? Thargor Orlando (talk) 19:57, 28 January 2015 (UTC)
Because your "clarity" and "specificity" gives a false impression of a violation of normal statistical procedure, which this is not. The Newsweek analysis gives no indication that the study authors have misunderstood statistical significance. Introducing your language would give that indication, in contradiction to the RS.192.249.47.186 (talk) 20:06, 28 January 2015 (UTC)
Technically, I think it does. The Newsweek article states that most tweets were "neutral in sentiment", which it quickly clarifies means "not identified by the algorithm" (and I think all of us here should be fully familiar with the concept of being negative to someone without using words that are specific red flags). It gives a fairly good sample size -- 25% of all #gamergate tweets --and makes the following analyses based on the data:
  • Tweets were overwhelmingly more focused at female game developers or commentators who had done no more than criticize Gamergate, rather than at journalists or males who Gamergate had made specifical allegations of ethical violations against
  • Most tweets were unable to be classified for sentiment by the specific algorithm used
  • Tweets directed at the male journalists are generally angrier than at the females
  • However, negative tweets directed at the females critics were far greater in quantity, indicating that the females recieved both more positive tweets mentioning gamergate, to even out the average, but also that a greater majority of negative tweets mentioning gamergate were sent to harass female developers who disliked gamergate rather than male journalists who gamergate had accused of ethical violations.
The overall conclusion of the article is that gamergate supporters, by a wide margin, prefer to harass female non-journalists who criticize gamergate or sexism in gaming, in place of harassing male journalists who have been alleged (truly or falsely) of violations by gamergate. The article restates forms of this conclusion several times throughout, so some rewording of it should be how we represent the article. The specific percentages of Positive/Negative/Undetermined tweets for each player are not the point of the article, and it feels undue to focus on them.192.249.47.186 (talk) 16:51, 28 January 2015 (UTC)
It's far worse than that. While at some point they had in hand 25% of tweets, they then whittle down to the 5% that mention one of the chosen 7 accounts. Figures like Eron Gjoni, Phil Fish, Greg Tito, Ben Kuchera, Milo, Totalbiscuit, etc that could skew the results greatly are left out. Then they further restrict to the 0.5% that the algorithm can classify in order to draw all their conclusions. This is misleading. What is at question is not whether Newsweek is generally a reliable source, or whether we can say in the article that the methods are bad. What is at question is whether we can knowingly abet the intent to mislead by leaving out conflicting information that can be found in the article itself. The ArbCom principle cited at the start of this says we can not. The article states both in words and graphics that the intent of most tweets could not be classified. It clearly shows only 7 accounts were considered. This is not original research. We don't need another reliable source to tell us which parts of the first source we can include. It doesn't matter if this information is at the top of the page or the bottom of the page. Leaving it out is misleading. Rhoark (talk) 17:47, 28 January 2015 (UTC)
Again, your personal interpretation of the data is irrelevant. What matters is what the reliable source who hired a professional social media analyitics firm decided. That is the starting point. The ending point is: Are we appropriately reflecting what the reliable source determined. -- TRPoD aka The Red Pen of Doom 18:14, 28 January 2015 (UTC)
The reliable source determined how many tweets they examined, how many tweets they did not examine, and that most of them could not be classified. The reader's opportunity to know and interpret this information for themselves is very relevant. Rhoark (talk) 18:18, 28 January 2015 (UTC)
No it is not. Particularly in the overall scheme of the article. If the reader cares, they click on the reference and find the details. -- TRPoD aka The Red Pen of Doom 23:42, 28 January 2015 (UTC)
The only objection to my proposed wording was that they included too much description. It should be possible to convey the skepticism this article is due using only a single figure: the percentage of tweets in the sample that were used to derive their conclusion. Arriving at this figure is simple calculation that does not constitute original research. (See WP:NOTOR). They initially sampled 25% of 2 million, so 500,000. Eyeballing their chart: http://s.newsweek.com/sites/www.newsweek.com/files/styles/embedded_full/public/2014/10/24/sentimentvolume.jpg?itok=eySg1iIE approximately 2400 tweets were categorized. There's some fuzziness here, so chime in if you have a better estimate. 2400/500000=.0048 so just under one half of one percent of tweets their sample were used to inform their conclusion. Incorporating the 75% they discarded at the outset, they used 0.12% of #gamergate tweets in the sampling period. Rhoark (talk) 19:03, 28 January 2015 (UTC)
Slotting this right in to the existing grammar of the statement, we could say: Newsweek/Brandwatch performed an analysis using 0.12% of two million Twitter messages with the Gamergate tag beginning September 1, 2014, and suggested from the results that "contrary to its stated goal, GamerGate spends more time tweeting negatively at game developers than at game journalists." Rhoark (talk) 19:14, 28 January 2015 (UTC)
We're not going to question a source with your own eyeballed fuzzy napkin math. The source specifically calls it a "reflective amount of data". Woodroar (talk) 19:27, 28 January 2015 (UTC)
WP:NOTOR "You may round to appropriate levels of precision." If you have a problem with my estimate, give your own. Rhoark (talk) 19:36, 28 January 2015 (UTC)
Not when you're using it to suggest that a sample size is insufficient, when the actual experts state that it is. Woodroar (talk) 19:39, 28 January 2015 (UTC)
Does the proposed wording suggest that? Show me. Rhoark (talk) 19:46, 28 January 2015 (UTC)
If not for that reason, then why the complicated breakdown of their sample? Woodroar (talk) 19:54, 28 January 2015 (UTC)
To enable the reader to decide. We can't tell them there's a problem with the statistics, but we are under an onus to indicate in some manner that there is content in the article we cite that conflicts with the conclusion we cite it for. The only acceptable alternative is to exclude the citation. WP:EP "on Wikipedia a lack of information is better than misleading or false information" Rhoark (talk) 20:10, 28 January 2015 (UTC)
Rhoark, inserting "the reader's opportunity to know and interpret this information for themselves is very relevant," would, in this case, amount to pointing at a number, tacitly implying it's an overly large number (without explaining, as the RS does, that this does not detract from the mathematical significance of the result), and putting it, contextless, in front of a reader that is well-established to have a poor understanding of how statistics work. There is ample understanding in the academic community that this kind of fuzzy napkin statistics is usually misleading if not outright dishonest. The main thrust of the Newsweek article is very clear: Gamergate supporters make more tweets harassing females/female developers unrelated to ethics allegations for criticizing gamergate, than they spend harassing males/journalists they've accused of violations. Any tweets the computer-programmed algorithm classified as "undetermined" are literally that -- tweets not taken into the analysis. The statisticians know how to deal with having sample sizes less than 100% (or even less than 25%), and you've failed to provide any evidence that they neglected to apply the well-understood accomodations for less-than-total sample size.
In order to insert that "context" here (which conveniently ignores the frequently-repeated analysis of gamergate preferring to harass female developers over male journalists) would amount to woowoo noises unless we went through and inserted "and here's everything that wasn't measured" into basically every sentence. The study, nor Newsweek's summary of it, is not out of the bounds of normal statistical procedure.192.249.47.186 (talk) 20:06, 28 January 2015 (UTC)
Reducing the initial sample to 25% is not the problem here. That is the one valid step they take. The rest is, to use your words "misleading if not outright dishonest". I am thoroughly aware of the kinds of mathematical tools that can control for sample size and support reasoning under uncertainty. Newsweek does not use them. They use only raw counts of Twitter mentions. We cannot use WP's voice to give their analysis the thrashing it deserves, but we have not just the justification but the responsibility to provide information from the citation that undermines its own conclusion. Rhoark (talk) 20:25, 28 January 2015 (UTC)
Might i remind you of "The contents of source materials must be presented accurately and fairly. By quoting from or citing to a source, an editor represents that the quoted or cited material fairly and accurately reflects or summarizes the contents and meaning of the original source, and that it is not being misleadingly or unfairly excerpted out of context. Failure to accurately reflect sources, whether by accident or design, is a serious matter as it undermines the integrity of the encyclopedia." You do not get to assert that the reliable sources have "gotten it wrong". (Well, you can, but no one needs to take such assertions seriously and repeated actions of that type will be subject to the GG sanctions)-- TRPoD aka The Red Pen of Doom 20:30, 28 January 2015 (UTC)
Discussions remain okay, and threatening people over a content dispute is not okay. Please stop doing that. As for your quote, we all agree with you. We just disagree as to whether the information from the source is being fairly or accurately reflected, and some of us believe that the source is being misleadingly and unfairly excerpted. This is directly relevant to the article. Thargor Orlando (talk) 20:36, 28 January 2015 (UTC)
Thargor, Rhoark -- do you feel that the intent of the Newsweek article is not "Gamergate supporters were found to spend more tweets harassing female developers and critics of gamergate more than they harassed male journalists that they had alleged violated journalistic ethics."? If we all agree that that's the intent of the Newsweek article, than any attempt on our part to "provide clarification" or "give it the thrashing it deserves" would be solidly SYNTH and OR.192.249.47.186 (talk) 21:46, 28 January 2015 (UTC)
That is clearly the intent of the Newsweek article, but its intent is at odds with its own "contents and meaning". Thats why citing it for its intent alone is misleading excerptation. It would have been nice to refute it directly, but that would be OR. I have never proposed to do that. What we can do is include additional specific information sourced from Newsweek itself so that it is no longer misleadingly excerpted. Rhoark (talk) 22:05, 28 January 2015 (UTC)
I believe we should be interested in what the article says, not what the article intended to say. If those two are at odds, then maybe we shouldn't be using the Newsweek article at all. Thargor Orlando (talk) 22:08, 28 January 2015 (UTC)
I know I said I was done with this never-gonna-happen section, but to rephrase, you believe we need to include a few specific sentences from the article so that readers can figure out on their own that the main point of the article is wrong, correct? Hipocrite (talk) 22:07, 28 January 2015 (UTC)
Close. We need to include some additional information, which is probably easier to do in a summary tone than by quotation. The justification in doing so is for the way in which it is cited to not mislead about its content. Rhoark (talk) 22:15, 28 January 2015 (UTC)
Again, you have failed to establish 1) that we ARE misleading and 2) that any "misleading" is relevant to the subject of the article. It seems that it is abundantly clear what the source found relevant and it seems equally clear that we are reporting what they found relevant. -- TRPoD aka The Red Pen of Doom 23:40, 28 January 2015 (UTC)
What they found relevant is not the entirety of the content and meaning of the article. If you prefer option A below to B or C, that's fine. If you can't accept additional well-sourced information being used, the burden of proof is on you to explain why. Rhoark (talk) 23:46, 28 January 2015 (UTC)

These are the options as I see it:

  • A: Cite the conclusion along with specific caveats drawn from the article.
    • A1: Note that only mentions against hand-picked accounts were included, and that the sentiment of most of the tweets is unknown.
    • A2: Skip over the nature of the sampling problems and encapsulate them with the figure that only 0.48% of the tweets were used to draw the conclusion. (Since the initial cut to 25% was a valid move, this is more fair than the 0.12% figure)
  • B: Don't cite the conclusion, just numerical results such as X combined tweets at Kotaku/Tolito/Grayson vs Y combined tweets at Zoe Quinn/Brianna Wu
  • C: Don't cite the article

Rhoark (talk) 23:43, 28 January 2015 (UTC)

Your basis that there is a "sampling problem" makes this discussion moot. The reliable source has professional social media analysts look at the data. There is no mention of any "sampling problem" mentioned in the article and no reliable third party has identified any. We cannot as Wikipedia editors proceed as if there were. Doing so is a blatant violation of the ArbCom's : "The contents of source materials must be presented accurately and fairly. By quoting from or citing to a source, an editor represents that the quoted or cited material fairly and accurately reflects or summarizes the contents and meaning of the original source, and that it is not being misleadingly or unfairly excerpted out of context. Failure to accurately reflect sources, whether by accident or design, is a serious matter as it undermines the integrity of the encyclopedia." -- TRPoD aka The Red Pen of Doom 23:46, 28 January 2015 (UTC)
We are not, as editors, bound to proceed only on the basis of citable sources. You may find it helpful to reflect on Wikipedia:Inaccuracy. The citation as it stands is misleading and does not reflect the totality of its source. We do not need citable sources in order to establish that. I'm not going to continue to rehash why it is misleading. You can look at what I've already said, what Thargor has said, and the medium.com analysis to help you understand. We cannot use those resources in the article itself, but fortunately the situation can be resolved using only material from Newsweek. The ArbCom quotation with which I began this section and that you keep repeating describes the problem with the status quo, not the proposed remedies. Rhoark (talk) 01:20, 29 January 2015 (UTC)
The quote as it stands does indeed reflect the totality of the source: Newsweek a reliable source hired a firm qualified to analyze social media and here are their conclusions. The details are are not necessary to understand the Newsweeks overall story about their study. Including such details, particularly for the purpose of attempting to throw the study into question are completely inappropriate. -- TRPoD aka The Red Pen of Doom 01:34, 29 January 2015 (UTC)
What's inappropriate is to allow known falsehood to remain in the article. WP:EP "on Wikipedia a lack of information is better than misleading or false information." You have developed a very strident notion that a publisher at any point considered reliable for any single claim thereby becomes infallible for all purposes, not only in articles but talk pages. There is no support for this in any Wikipedia policy, essay, or community consensus. Rhoark (talk) 03:05, 29 January 2015 (UTC)
So now your position is that the Newsweek article is false? And you are basing this on? -- TRPoD aka The Red Pen of Doom 03:25, 29 January 2015 (UTC)
Newsweek's conclusion is devoid of any relationship with the data they collected. It is perfectly admissible to cite them for their opinion that they figured something out, but it has to be tempered with some indication that avoids misleading people into accepting that Newsweek's conclusion bears even a semblance to their data. Rhoark (talk) 03:44, 29 January 2015 (UTC)
So you are second guessing a reliable source based solely on your interpreation . Cant do that. -- TRPoD aka The Red Pen of Doom 04:10, 29 January 2015 (UTC)
I'm not "second guessing". I'm finding unequivocally that the source is inaccurate. An editor is not only enabled by policy but mandated to consider the accuracy of a source, whether or not it has been previously considered reliable. Your continual insistence that this is not a function an editor can perform is unproductive. Rhoark (talk) 04:47, 29 January 2015 (UTC)
"The contents of source materials must be presented accurately and fairly. By quoting from or citing to a source, an editor represents that the quoted or cited material fairly and accurately reflects or summarizes the contents and meaning of the original source, and that it is not being misleadingly or unfairly excerpted out of context. Failure to accurately reflect sources, whether by accident or design, is a serious matter as it undermines the integrity of the encyclopedia." Without anything other than your assertion that it is inaccurate, we are done. -- TRPoD aka The Red Pen of Doom 12:58, 29 January 2015 (UTC)
Sometimes a good argument is enough. We're trying to write an encyclopedia, and part of that includes editorial discretion. Why you're arguing to use a misleading-to-incorrect source in the article, I remain puzzled by. Thargor Orlando (talk) 13:33, 29 January 2015 (UTC)
Because there's nothing to demonstrate that it is "misleading-to-incorrect" beyond the fuzzy napkin math of a non-expert editor. That is the entire reason WP:NOR is a policy. Seriously, can you not see how it would be a problem for, say, an article stating the government was running up the debt to be "discarded as an RS" simply because an editor said "Nah, that doesn't look right"? I mean, we're getting into Trutherism here.
I guess neither of you are taking me up on my request to try to write an analysis that could get published and actually cited, so at the very least, have either of you actually [i]read the Brandwatch study[/i]?192.249.47.186 (talk) 15:32, 29 January 2015 (UTC)
Simple calculation is WP:NOTOR. A valid analogy would be an article with the headline "Government is running up the debt" where later in the article it has a chart showing increased taxes and no information about spending. I have only read what Newsweek says about the Brandwatch study, but if this study itself is available in a more raw form it might be a preferable source entirely. Rhoark (talk) 17:56, 29 January 2015 (UTC)
That would be true if you were proposing a simple calculation. You're not. You're proposing that numbers you decide (which, let's be honest, rely on what are almost certainly fuzzy numbers to begin with) are significant be mentioned as significant, when the RS you are taking them from specifically portrays them as not significant. You are asking that we interpret data in specific opposition to how the RS interprets it, and to cap it off, you've even made it very explicit above that your intention is to "thrash" the RS. Your whole argument, in fact, hinges on the assumption of Newsweek's incompetence in interpreting data based on Newsweek's interpretation alone, when you admit you haven't looked at the actual original data they draw their conclusions from.
I'm frankly bewildered here, as this seems to be the exact behavior that WP:OR is intended to prevent. When and if you decide to take a look at the original study, and can demonstrate from actual evidence that Newsweek's analysis is misleading, that would be absolutely worth discussing. But a citation to Newsweek for what would essentially be "Hey, Newsweek is stupid" seems the very worst at misrepresenting a citation.192.249.47.186 (talk) 18:45, 29 January 2015 (UTC)
The actual data they draw their conclusions from are summarized in the charts they publish, which is the same basis for my arguments. There is no reason to believe any other unpublished information informed their conclusion. Material taken directly from the article, with or without simple calculations, cannot be OR. An editorial decision to make that inclusion is perfectly permissible to support using original research, and is strengthened for it. The fact that Newsweek considers these matters not to be significant only deepens the case for their Wikipedia:Inaccuracy. Guidelines on potentially inaccurate material are to inline attribute it and provide balancing material, or if this is not possible then to drop the citation. Rhoark (talk) 19:10, 29 January 2015 (UTC)
The WP:NOR article explicitly disagrees with you: "Similarly, do not combine different parts of one source to reach or imply a conclusion not explicitly stated by the source." You have yourself stated that the conclusion we've been talking about "is clearly the intent of the Newsweek article", and that your intent is to "provide information from the citation that undermines its own conclusion". The WP:NOR policy explicitly says that what you are trying to do is a misuse of sources, and absolutely forbidden. If you are unfamiliar with the details of the NOR policy, as you seem to be, please go read it before continuing to edit, but right now you seem to be arguing as if the NOR policy doesn't exist. If you still dispute this interpretation, I strongly advise you to take it up with the noticeboard, making sure to link back to this thread.192.249.47.186 (talk) 21:52, 29 January 2015 (UTC)
The sentence you quote is from WP:SYNTH Synthesis does not occur if no new thesis is stated or implied. WP:SYNTHNOT gives further guidance that mere juxtaposition is not synthesis, summary or explanation is not synthesis, a conclusion an educated person could reach from reading the cited source is not synthesis, and anything that is not OR is automatically not synthesis. Neither OR or SYNTH applies to the reasoning for an edit given on a talk page. I am in the habit of reading every applicable policy, guideline, or essay in the process of preparing an edit, rather than stopping with the first convenient phrase that seems friendly to my position. I advise you to avoid calling the WP:KETTLE black. Rhoark (talk) 22:31, 29 January 2015 (UTC)
Furthermore, WP:Cherrypicking Rhoark (talk) 22:38, 29 January 2015 (UTC)
Wow. Okay, first off, yes, that sentence is in the SYNTH section, which is on the NOR page. Second off, you have explicitly stated that your intent in adding this juxtaposition is to "provide information from the citation that undermines its own conclusion". "Mere juxtaposition" is allowed, certainly, but I don't see how you can claim, with a straight face, that you are attempting "mere juxtaposition" while simultaneously claiming that you intend to "provide information from the citation that undermines its own conclusion". This is also distinct from summary, as you are, in your own words, attempting to introduce a new thesis -- that the Newsweek interpretation is inaccurate. This is not a thesis present in the Newsweek article, and is not a valid summary of their article.
I am honestly unsure whether productive discussion can continue on this, if you're now accusing me of arguing in bad faith. Would you prefer to take this to dispute resolution?192.249.47.186 (talk) 22:48, 29 January 2015 (UTC)
I accuse you of nothing, but want you to know your condescending implication that I might not have read a policy I cited while simultaneously misapplying it yourself is unappreciated.
Yes, there is information in the source that undermines the conclusion. It is my intention to include this information in the article, as prescribed by Wikipedia:Cherrypicking. "The main information from a source, insofar as stated in Wikipedia, must be accompanied by any contradictory and qualifying information from the same source." "As to contradictory information, if, for example, a source says 'Charlie loves all blue coats and hates all red coats', to report in Wikipedia that according to the same source 'Charlie loves all ... coats' is cherrypicking. It is cherrypicking even if the source is precisely cited." "Qualifying information is information that might not contradict the main information but that alters how the main information should be understood. For example, to quote a source that says that most Americans sleep late and skip work but to ignore that the source limits that by saying "on weekends" is to omit qualifying information and misrepresent the source on Americans' work customs." "Either contradictory or qualifying information may be found anywhere in a source, not necessarily adjacent to the main information. For example, while the main information may be in a middle chapter of a book, contradictory or qualifying information may be in an endnote, in an introduction, or on a cover. Many sources are well organized and make finding everything you need relatively simple, but not all sources are so helpful."
The logical structure of these example fallacies is almost isomorphic to those exhibited by Newsweek. Replace coats with twitter accounts and replace "on weekends" with "when the algorithm could decide".
Just because a source has at some point been considered reliable does not grant it protection from skepticism. On the contrary, one of the reasons given that cherrypicking is a a problem is that it could cause information to be missed, that could otherwise point to the unreliability of the source.
There is no synthesis in the article if there is no new thesis stated in the article. If information is juxtaposed in a way that a reasonable reader will draw a valid conclusion themselves, that's just good editing. It is mere juxtaposition if the relationship between two pieces of information is not characterized by a connective such as "because" or "however", but uses that are not mere juxtaposition may still not be synthesis. It's not synthesis if its not original research. Sums, averages, approximations, and other simple mathematical operations combining separate verifiable numbers are not original research or synthesis. The burden of proof is on someone asserting a claim is synthesis.
Wikipedia:Cherrypicking and Wikipedia:Inaccuracy have the status of an essay rather than guideline or policy, but as essays they are well regarded and ultimately traceable to reasoned application of WP:EP and WP:V. If you do want to have a productive discussion, I suggest you express a preference for one or more of options A, B, and C that I presented previously. Rhoark (talk) 03:44, 30 January 2015 (UTC)
If I may reply to long with short: "Yes, there is information in the source that undermines the conclusion." This statement isn't accurate. Identifying this false premise solves your dilemma, I should think. Darryl from Mars (talk) 08:55, 30 January 2015 (UTC)
Prove it. Rhoark (talk) 14:41, 30 January 2015 (UTC)
The burden of proof is on the person making the statement, Rhoark, not the person calling it out. Would you like to preset proof that your premise is not false? PeterTheFourth (talk) 15:06, 30 January 2015 (UTC)
This thread currently runs to six pages on my screen, and I have a big screen. I will not recap it all to do the work for an editor who tacks "nuh-uh" on the end. A summary for the benefit of external commentators may yet be necessary, but in the meantime the analogy of the coats in the strand directly above this interjection is a good starting point. Rhoark (talk) 16:49, 30 January 2015 (UTC)
The NOR, SYNTH, and Cherrypicking guidelines are exceedingly clear that their purpose is to prevent editors from misrepresenting the message of a source by taking specific pieces from a source, whether you're taking more or less than what the article already uses, in order to make an implication or thesis that is not an accurate summary of the source. Your stated intent is to take a point from the article in order to "give it a thrashing" and "undermine the message of the article". I'll say it again: you have explicitly stated that your intention is to use the source to make a message opposing the spirit of the source. This is a bald-faced admission that you intend to flout the spirit of the guidelines. If you truly, honestly, sincerely believe the claims about the purpose of the guidelines that you are making, take it up with the OR noticeboard. It's there for exactly that reason. But otherwise, there's nothing else to say, as consensus and policy are clearly against you.Not even Mr. Lister's Koromon survived intact. 17:29, 30 January 2015 (UTC)
You make an argument that supports inclusion of Newsweek's thesis so long as the source is cited. To do otherwise would equally be cherrypicking. Consider option B off the table. The purpose of the policies and essays I have cited is not to enshrine the "spirit" of a source but to ensure the material on Wikipedia is verifiable. If the motives of editors are to come in to play, the first one to question would be the motive for misleading the reader. The OR noticeboard is not an appropriate venue for matters that do not involve original research. Rhoark (talk) 18:57, 30 January 2015 (UTC)

Candidate language

Before making a contribution to the effect that "The status quo is fine" or "This looks like OR" please review the preceding section.

I propose that the claim sourced to http://www.newsweek.com/gamergate-about-media-ethics-or-harassing-women-harassment-data-show-279736 which presently reads

Newsweek/Brandwatch performed an analysis of about 25% of two million Twitter messages with the Gamergate tag from September 1, 2014 onward, and suggested from the results that "contrary to its stated goal, GamerGate spends more time tweeting negatively at game developers than at game journalists".

be replaced with

Using a sample of tweets that mentioned seven specific accounts, Newsweek suggested, "contrary to its stated goal, GamerGate spends more time tweeting negatively at game developers than at game journalists." The number of game developers and journalists worldwide is significantly larger than seven. The algorithms Brandwatch used on behalf of Newsweek were unable to determine positive or negative sentiment in more than 85% of the tweets mentioning the seven accounts.

Rhoark (talk) 22:03, 30 January 2015 (UTC)

That is not an accurate summary of that source. 109.152.100.135 (talk) 23:15, 30 January 2015 (UTC)
It's not an accurate summary of the source, and it's also an attempt by an editor to put the source on trial, apparently for no reason other than that they disagree with the source's conclusions. The edit to add that change to the article is entirely inappropriate, in my opinion. AtomsOrSystems (talk) 23:21, 30 January 2015 (UTC)
WP:IDONTLIKEIT is not an argument. It's not supposed to be a summary of the source. It's inclusion of qualifying and contradictory information alongside the thesis of the source, as prescribed by WP:Cherrypicking. Rhoark (talk) 23:24, 30 January 2015 (UTC)
I'm beginning to get the idea that you really enjoy citing wikiessays. And my argument is not WP:IDONTLIKEIT, it's that you are making an unsourced statement regarding another source, based purely (so far as I can tell) on WP:IDONTLIKEIT. If you had a source that called the Newsweek source into question, that would be one thing. You don't, and your insertion of original language is inappropriate. AtomsOrSystems (talk) 23:28, 30 January 2015 (UTC)
I do like to maximally support my actions using existing project consensus. Based on that consensus, I do not need another source to support what I have done. It is within editorial discretion. All the information used is verifiable. Paraphrasing a source is standard practice. The reason to do it is because without doing so the citation is misleading, not just distasteful to me. Ultimately, the change is an improvement to the encyclopedia, and that is what matters in the final analysis. Rhoark (talk) 23:33, 30 January 2015 (UTC)
I would suggest that the many, many lines above on this exact subject indicate that we have no firm agreement, or any kind of agreement, among the editors of this article, that your change is an improvement. They also strongly contest your claim that WP:Cherrypicking applies. Rather the opposite, in fact. And instead of citing WP:DRNC in a reply, can I recommend that you do the responsible thing as an editor in a cooperative project, and revert your edit so we can reach an agreement? That, in my opinion, would be what is best for the encyclopedia. It's always preferable to not have one editor making unilateral changes against a chorus of disagreement. AtomsOrSystems (talk) 23:43, 30 January 2015 (UTC)
That language won't work. As others have said, it's not an accurate summary of the source; in particular, stating "The number of game developers and journalists worldwide is significantly larger than seven" is original research (in that you are making the implicit assertion that that is a relevant comment, and using that to present your personal views on the study in the article voice.) --Aquillion (talk) 23:40, 30 January 2015 (UTC)
No one has yet identified any way in which what I have done is OR. I expect they never will. They cannot. The "more than 7" bit is of an obviousness way beyond needing citation, but if that is really the locus of dispute, I don't object to omitting only that sentence until a citable source is found. That should be very easy. Everything else I said was sourced from Newsweek, and juxtaposition is not OR. It does not need to be a summary of the source. It does not need to agree with Newsweek's conclusion. It mustn't agree, because the conclusion is misleading. An additional source is not needed to confirm Newsweek is misleading. People want more time to form a consensus; they shall have it. If in this time no new arguments are advanced, then there will be no reason to do other than what I have done. Consensus is not unanimity or a vote, but based on reasons. Rhoark (talk) 04:47, 31 January 2015 (UTC)

@HJ Mitchell: Editing the page to your version when consensus was against was -way- out of line, rhoark, especially when multiple editors have given you reasons citing policy for why your vetsion is innapropriate. That you disagree with their interpretation gives you no right to ignore them.66.87.77.218 (talk) 05:08, 31 January 2015 (UTC)

Consensus is an important part of the Wikipedia project. However, the most important part is to build an encyclopedia. I have enlisted a ream of policies and essays in support of my position. The most central is WP:EP:"a lack of information is better than misleading or false information". No secondary source would be necessary to establish that Newsweek is misleading, but as it happens one by Andy Baio is already cited. Unless a consensus forms that it is justified to Ignore All Rules in this instance, any other consensus towards "no change" must be disregarded. Rather than fighting change, editors' attention would be better spent discussing the best wording for the change. Rhoark (talk) 05:37, 1 February 2015 (UTC)

Citations in the Lede

My previous stance was we don't need them as it's a summary and they can be found in the body, but I'm finally tired of the {{fact}} tagging. Any chance that consensus could change on adding citations to the lede? — Strongjam (talk) 03:06, 1 February 2015 (UTC)

I honestly don't care either way. It'd be fairly trivial to add them, thought it might impact readability a small amount. Are there any relevant policies about adding citations to the ledes of articles? PeterTheFourth (talk) 03:08, 1 February 2015 (UTC)
WP:LEADCITE. Which is best summed up with this bit "The necessity for citations in a lead should be determined on a case-by-case basis by editorial consensus. Complex, current, or controversial subjects may require many citations; others, few or none." — Strongjam (talk) 03:12, 1 February 2015 (UTC)
In the particular case of {{cn}} the statements in the lede are fairly simple, noncontroversial (whether some sources said something) and easily found in article by text search. -M.Altenmann >t 03:14, 1 February 2015 (UTC)
My opinion is the same from a few days ago: Since we're summarizing material based on 159 sources (at current count), we could theoretically require 159 sources in the lede. We probably wouldn't need that many sources, I hope, but I'd say about a third of them are cited multiple times, which means they're probably critical sources. So let's call it 50 sources in the lede, which for 4 sentences is 12.5 sources per sentence, all of which would have to be checked every time someone wants to add some nuance to a sentence or tighten the language, maybe juggle some phrases around to make it flow better. To me, that's the single best reason. The lede is currently more than 4 sentences, but the sentiment remains. Every time we want to, say, rewrite three sentences into two, or move phrases back and forth, we'll have to check each source and move it accordingly. It's going to end up being a huge pain in the ass. Just my $0.02. Woodroar (talk) 03:20, 1 February 2015 (UTC)
The whole point is that the lede is a summary of our article, kinda "quaternary source". And each time you "juggle some phrases around" you must check whether the lede is faithful to our article. If for some phrase in the lede it is hard to find the corresponding article text, this means the lede sucks, references or not. -M.Altenmann >t 03:29, 1 February 2015 (UTC)
That's not quite what I'm getting at. Let's say our theoretical lead has 3 compound sentences: "[Statement A], while [Statement B].[12 sources] [Statement C], but [Statement D].[12 sources] [Statement E], though [Statement F].[12 sources]" But maybe someone suggests that the lede would flow better if we wrote it as "[Statement A], even though [Statement D]. But [Statement F] and [Statement C]. This means [Statement B] leads to [Statement E]." We've had some fairly drastic restructuring of the lede lately, so that's not even a "what if?" situation. It's going to happen. And right now there's not a big issue with that, because all we're doing is summarizing the article itself. But with a fully-cited lede, every time we rephrase it, we have to go through and move each source with its associated claim. Changes start to become prohibitive because of the work involved. Maybe that won't happen, but the history of this article has been citing and backup citing and backup backup citing to the point where controversial statements have a string of citations on them, and I see the lede just compounding that issue. Just my $0.02, though. Woodroar (talk) 06:31, 1 February 2015 (UTC)
Considering the lead will have few or no specific stats or anything like that, it will always have to be some sort of aggregate summary. Anyone who wants to edit it should be expected to read through the entire article, or at least the relevant sections. Citations will not make this process easier.
Adding fact tags to a heavily cited, highly controversial article like this amounts to careless editing. Good faith should be assumed, but all such attempts should be reverted since communicating through fact tags in a case like this is pretty disruptive. Competence should be required in regard to this.
Peter Isotalo 12:37, 1 February 2015 (UTC)

explanation of lede change

I've made this change: [13], which basically means to address this statement which is not supported as a fact in sources: The leaderless, amorphous group that acted under the hashtag became known as the "Gamergate movement". The movement came simultaneously to the harassment attacks and did make it clear in their statements that harassment was bad. Of course there's a great deal of opinion that the movement was a cover for the harassment attacks, but we don't have sources that affirm that as a fact. As such, rewording this Simultaneous to these events, a leaderless, amorphous "Gamergate movement" grew, using the #gamergate hashtag to identify issues of ethical concerns they had towards these video game professionals. points out the various Occums razor reasons that many think that GG movement is just a cover (happened at same time, targetting the same people). (and of course avoids introducing the BLP issue that started this in the lead) Further, this is a point that the article and our sources supports, in that many many sources doubt the true nature of the GG movement, and have criticized the group for enabling the harassment. --MASEM (t) 22:36, 31 January 2015 (UTC)

It seems to me that Simultaneous to these events, a leaderless, amorphous "Gamergate movement" grew is worded in a way that supports the idea that the "Gamergate movement" had no involvement in the harassment, which seems equally unsourced (and in some cases, opposed to sources). Can I suggest an alternate (and simpler) wording change, such as The leaderless, amorphous group that formed under the hashtag became known as the "Gamergate movement"? It seems this is a simpler change, and it removes the earlier implication that the movement was formed by those "acting" in regards to the harassment. AtomsOrSystems (talk) 22:45, 31 January 2015 (UTC)
Also, I notice you added the sentence These commentators have expressed doubts to the true intentions of the movement, and claim that the movement's unorganized nature has created an environment that has enabled the ongoing campaign of harassment. I'm not certain a review of our sources from the listed commentators show that all, or even most, of them have expressed doubts of the movements true intentions. Unless you mean to say that they often say they doubt the movement's stated intentions regarding ethics and journalism. Either way, I'm not certain that the additional sentence is needed, or warranted. AtomsOrSystems (talk) 22:50, 31 January 2015 (UTC)
My difficulty with ... to identify issues of ethical concerns they had towards these video game professionals is that this is only half the story. Gamergate is about ethical concerns and anti-feminism/anti-social critique. Highlighting half the story in the first paragraph seems misleading - the second paragraph seems to better expand on their issues. - Bilby (talk) 22:55, 31 January 2015 (UTC)
Concur with Bilby above: what is meant by "ethical concerns" by those that use the term in this context is far broader than what common usage denotes. I would suggest substituting "ethical and ideological concerns" to communicate this breadth of meaning. CIreland (talk) 23:05, 31 January 2015 (UTC)
Which sources are you using to support "Simultaneous to these events"? Kaciemonster (talk) 22:59, 31 January 2015 (UTC)
I've reverted the change until a consensus is formed, since a few people raised concerns. I think that's the correct protocol, feel free to revert my revert if it isn't. Kaciemonster (talk) 23:22, 31 January 2015 (UTC)
Black Kite, may I hear your reasoning? It seems like most people so far are disagreeing with the language, which is why I reverted it in the first place. Kaciemonster (talk) 00:27, 1 February 2015 (UTC)
I don't believe our sources support the claim that there are two gamergate movements, one of which is completely harassment free and only concerned with ethics, and one of which is notable only for being made up of sexist harassers. When looking at amorphous, leaderless groups, we cannot label any who reliable sources claim are part of the group as separate to it- this would be original research on our part. PeterTheFourth (talk) 23:34, 31 January 2015 (UTC)
Agreed. The vast majority of sources treat GamerGate as one thing; there are some talking about how it is hard to determine who supports what, and about how some people pushing the hashtag are using it for harassment while others are not, but I don't think there are significant number of sources that would separate them into two distinct events or movements the way the "simultaneous to this..." wording does. Most sources seem to basically use 'GamerGate movement' to refer to everyone using the hashtag, including people who are responsible for harassment. --Aquillion (talk) 00:07, 1 February 2015 (UTC)
Some sources for the above, from the article:
  • Univerity of Wisconsin center for Journalism Ethics: "While many claimed this movement was about calling out ethical lapses in videogame journalism, I was astounded and appalled by the misogynistic and threatening nature of some posts. People — particularly women — were attacked for speaking out, often getting “doxxed” (slang for having your personal information documented or published online). I, like many, have had and still have hope that the participatory nature of digital media will help more people engage with news coverage, counter bias and correct errors. But GamerGate is challenging those hopes of mine. Much of the conversation — if I can even call it that — has been a toxic sludge of rumor, invective and gender bias. The irony comes from people who claim to be challenging the ethics of game journalists through patently unethical behavior."
  • Is GamerGate About Media Ethics or Harassing Women? Harassment, the Data Shows
  • Here: What does GamerGate want? We don’t know. Attempts to engage with those involved often turn into rambling, cyclical shows of defensiveness with no concrete reasoning. Reasonable party lines are drowned out in the conversation. One thing is clear: when members of the games industry are being driven from their houses and jobs, threatened, or abused, it makes GamerGate’s claim that it is engaged in an ethical campaign appear laughable.
  • (Describing why GamerGate behaves the way it does, here : There is also an identification with the mob and its chaotic, dynamic nature – consider the delight Anonymous took in being an inchoate, implacable enemy of whoever aroused its ire (“Because none of us are as cruel as all of us”). As the MIB says, “Chan culture considers personal reputation meaningless but collective identity sacrosanct”. Deliberately trying to stand out is in the eyes of this subculture and in the rules of its discourse uncouth. To claim that the consensus is wrong and that one’s personal experiences can refute a point breaks the rules of discourse. To have a mass of people respond with rude invective to any statement (including statements they actually agree with) is how you have an argument: if you do not want a wave of hostility, why did you invite it by making a claim?
  • Here: Okay, enough metaphor, Jon. What is #gamergate, literally and specifically? It’s a Twitter hashtag. What else? What else indeed. While various patterns of behaviour coalesce around the hashtag, #gamergate’s protean nature resists attempts toward summary and narrative. It readjusts and reinvents itself in response to attempts to disarm and disperse its noxiousness, subsuming disaffected voices in an act of continual regeneration, cycling through targets, pretexts, manifestoes and moralisms. Say that it began as a harassment campaign targeting a female indie developer, as reported by credible news sites, and you are subjected to contradictory objections - "No, #gamergate began after that, as a reaction to biased reporting" and "No, #gamergate has been building up for years" - as proponents jostle for the story that paints them in the best possible light
  • Here: "The abuse currently polluting Twitter and other social media is largely aimed at female devs – targets include Depression Quest creator Zoe Quinn and Giant Spacekat founder Brianna Wu – and often linked, though not always, to #Gamergate. The root of this self-professed‘movement’ are well-documented enough, with details of Quinn’s personal life leaked by an ex-boyfriend. But this incident has almost become irrelevant given the severity of threats levelled at victims." ... "22Cans’ head of production Jemma Harris adds that such behaviour has grown worse as the internet has seemingly bred “a faceless culture that thinks it’s okay to do this"
--and many more like that, none of which support your thesis that the two are separate. These are the overarching way in which GamerGate is covered and described by the reliable commentators we're using in the article; the harassment and generally caustic behavior they see as originating from GamerGate is described as an outgrowth of its origins in chan culture, the depths of the culture-war passions driving it, and the deep belief the movement has that their ideological opponents are monstrous and must be crushed at any cost. I think that the chan culture analysis is particularly pointed, because (as the sources who discuss it say) it describes why so many people in GamerGate say that they are against harassment and have spoken out against harassment while, in the same posts (and under the GamerGate hashtag!) they continue to do things that the rest of the world describes as harassment. There's a sense among them that vitriol is natural on the internet, and that confronting someone with a flood of angry invectives therefore isn't real harassment the way a death threat is. I can see where you're coming from with that, anyway; but we need to go with the majority of reliable sources, and they have described what the #GamerGate hashtag is used for -- that is, what the GamerGate movement is, to the extent that that's a discrete thing they can define -- as being inextricably entwined with harassment. --Aquillion (talk) 00:31, 1 February 2015 (UTC)
I would also add that even in the case of valid (or at least discussable) criticism, it is delivered in a highly vitriolic way, not only in the choice of the language, but full with personal insults. And if the tone objected, the answer is usually kinda "Well, lady, you stepped into a snake pit. Don't whine, deal with it or out" -M.Altenmann >t 03:36, 1 February 2015 (UTC)

What the larger issue has been, and discussion needed on that

Related to points made above, and with new eyes (due to ArbCom's request) on this, I think we need to (re)visit a core point of debate of how this article is written, in relationship to WP:V (particularly the verifyability-not-truth paradox) and issues related to NPOV.

Specifically, because the press has largely blamed the Gamergate movement for the harassment, there are parts of this article (particularly in the lead) that wholly blames the GG movement for directly harassing, etc. The issue is multifaceted but here are points to consider/discuss:

  • There is clearly different definitions of what the Gamergate movement is depending on who you ask. The press use a more encompassing definition that at its largest includes anyone that has used the #GG hashtag; on the other hand, if you look at the sites/forums that are the apparent home of the movement, they consider that much smaller, with more focus. Yes, the difficulty in defining this is a criticism of the movement to start, but point here is that we have a group that consider themselves part of the movement, while others not associated with the movement consider the inclusion much differently.
  • In relationship to this, the self-stated movement has clearly stated by their public materials that they condemn the harassment, do not condone its use as part of their communication tools, and actively try to stop/report the use of harassment done in the name of GG. This is mentioned in some sources but it's spelled on their self-published sources, which are poor/unacceptable sources, though would be the truth from their side. A majority of the press do not believe this, but some do point out the harassment partols by the movement. In otherwords, I know from my own research with sources that wouldn't be acceptable on WP that this is what they claim about themselves, very much counter to what the press claims they are.
  • Harassment, doxxing, swatting, and rape/death threats are criminal acts in many areas, and I'm certain at the end of the day there will be criminal proceedings on the specific individuals that did these once their identities are known. However, as best as I've read, no one yet has been positively identified for doing these outside their handles. Now, yes, BLP does not apply to pseudoanonymous identities (per a discussion on the BLP board about the articles that discussed the mods of the Reddit KIA board and their other questionable activities), but a principle of BLP - where applying criminal blame where there is no such blame - should still apply as we are a neutral, impartial source.

What this amounts to is that we have to be careful and cannot state as a fact (in WP's voice) that the whole of the Gamergate movement is directly responsible for the harassment. We can state the press believes this, we can state that the press indirectly blames them for the harassment due to their poor organization and the environment and poor attitude they have taken towards the targets of harassment, and a whole bunch of criticisms. But as a neutral source, we should be given the people that call themselves as part of the movement the benefit of doubt (even if that's not a lot of doubt) that they claim they aren't responsible and thus we cannot factually assign blame on the entire movement. This doesn't change the content for the most part of this article but it does require careful wording of our statements of fact, which was what the change above is. This is why I feel we need to be extra careful of calling the movement as a whole directly responsible for the harassment attacks as a fact, even if the press has made this claim. If anything, calling the entire movement is a contentious point (between the press and the movement itself) and we should treat it as NPOV says, present both sides of this point within the context of WEIGHT. --MASEM (t) 03:44, 1 February 2015 (UTC)

This is a lot of original research, Masem. Your objection to us stating what the overwhelming majority of the media says is that you believe what the media says is wrong- this is not enough, unfortunately. We cannot give a leaderless movement the 'benefit of the doubt' if doing so requires us to go against what our sources say. Again, I urge you to find sources that show your point of view if you disagree with the article rather than endorse original research. PeterTheFourth (talk) 03:50, 1 February 2015 (UTC)
The issue is not the "overwhelmingness". If any of reliable sources describe these "good gamegaters" in reasonable detail, you are welcome to expand the article. (BTW, Masem is not disagreeing with the article; he seems to say it is incomplete.) -M.Altenmann >t 04:16, 1 February 2015 (UTC)
I'm saying they are expressing opinion more than facts. They may be (and I suspect they are mostly) right, but we don't know. This is what NPOV says when it comes to contentious statements : "Avoid stating opinions as facts. Usually, articles will contain information about the significant opinions that have been expressed about their subjects. However, these opinions should not be stated in Wikipedia's voice. Rather, they should be attributed in the text to particular sources, or where justified, described as widespread views, etc. For example, an article should not state that "genocide is an evil action", but it may state that "genocide has been described by John X as the epitome of human evil." --MASEM (t) 04:08, 1 February 2015 (UTC)
re "genocide": You are joking, right? The proper wikpedian way to say "genocide is a crime against humanity", not just opinion of John X. On the other hand I did understand what you mean: instead or writing "the movie sucks" it is better to write "John X says the movie sucks". Still better, there are reviewers which aggregate the reveiews and say "65% reviews say the book sucks". Coming back to gamergate, in order to handle an alleged NPOV, you have to document several POVs. If you can document only a single POV, your point is moot. So, once, again, please find good sources which speak about "good gamergarters". -M.Altenmann >t 04:18, 1 February 2015 (UTC)
We do have sources that discusses the antiharassment parts of GG [14]. And note that we should not be considering if we are talking "good" or "bad" GGers (that's biasing the discussion). It is the fact their documentation that is out there for anyone to read (but has potential BLP issues so I will not link, but it is very public) that they say they aren't harassment-based. Even if this is really a lie, we have no evidence that this is the case, and per our polices, we should be given them the appropriate benefit of doubt in terms of assigning the blame of harassment as a fact. --MASEM (t) 05:18, 1 February 2015 (UTC)
re: "we have no evidence" - We have plenty of evidence from reliable sources which is why the lede correctly notes the consensus that the ethical concerns are unfounded, debunked, trivial, unrelated to ethics etc. What you seem to be suggesting is that the self-identification of any person or group must be granted a controlling interest in the article, and that is clearly not the policy on Wikipedia. Scientologists don't get to write the Scientology article. The editors who do write it don't have to give the self-description provided by Scientologists equal weight alongside the consensus view of Scientology. I'm sure you can think of countless other examples. In short it's reliable sources, not GamerGate itself, that determines content. So find reliable sources to support your suggested editing and stop relying solely on self-description. Emarkcd (talk) 17:23, 1 February 2015 (UTC)
re: 'expressing the opinions'. Some sources are 'expressing the opinions', while others summarize these opinions. Now what? "In opinion of John X. Random, the widespread opinion is that the opinion about gamergate is wrong"? While your generic quotation of policy is correct, you better be more specific and write clearly what is misrepresented in the article. -M.Altenmann >t 04:23, 1 February 2015 (UTC) -M.Altenmann >t 04:23, 1 February 2015 (UTC)
The "other" opinion here is for the most part what the GG movement says about themselves - that they are not involved and also condemn the harassment. When we state "The Gamergate movement is responsible for the harassment", we are stating something that has not been proven to any degree and is a highly contentious statement in considering the GG side of the situation. --MASEM (t) 05:18, 1 February 2015 (UTC)
@Masem: re:" GG movement says about themselves" - the CG movement shot themselves in the foot when they refused to self-organize (AFAIU). Now any moron or troll who uses the tag is automatically member of the movement. Therefore I am using the term "good gamergaters" lacking better description. By letting the horse out of the stable they lost the right to say that they are "true CG movement". Because everybody else sees a quite different "true CG movement". THerefore when we state "the Gamergate movement is responsible" we are stating something that is proven to the extent reasonable. Only when the "true Gamergaters" self-identify themselves to make it possible to weed out "impostors", we can speak of an identifiable "CG side" and protect its rights. Otherwise "what the GG movement says about themselves" is akin to the "No true Scotsman" fallacy. Right now "the CG side" is seen as 88% trolls. -M.Altenmann >t 17:25, 1 February 2015 (UTC)
Masem, you have posted "the press" over 55 times on this Talk page recently. "The press" is otherwise known to Wikipedia as RELIABLE SOURCES. You, once again, are twisting our policies for some strange reason. Stop it. Your collogues do not appreciate it, nor do they your flippant use of genocide, Nazism and the Holocaust. 192.187.100.10 (talk) 05:01, 1 February 2015 (UTC)
I've never mentioned these (outside the above requote out of NPOV), and that's also a personal attack. This is a very subtle but key detail that needs to be resolved that we have to look at the overall situation to understand how the tone is critical here. You cannot just wave away "oh, they are reliable sources, we should never question them", because our polices do allow for us to make sure RSes are used appropriately and to make sure we separate fact from opinion when contentious points are presented. --MASEM (t) 05:18, 1 February 2015 (UTC)
re "criminal blame" What the heck are you talking about? The threats are documented. It does not matter whether they were from pseudoanonymous identities or bot-generated. Are you saying that we must also mention the opinion that Sarkeesian fabricated these threats, or, even better, told her buddies to post these threats so that she may cry helpme? -M.Altenmann >t 04:26, 1 February 2015 (UTC)
They are documented, but who exactly carried them out is not known, nor their relationship to the self-described GG movement. They could be a part of it, they may not be. As such, we cannot assign that blame to the movement even if the press has expressed their opinion this is the case. --MASEM (t) 05:18, 1 February 2015 (UTC)
Once again, AFAIU the only way CG movement is "self-described" is the usage of hashtag. If I am mistaken and you have other way to distinguish "good" users of #gamergate hashtag from evil ones, you are welcome to expand the article with this info. Clearly, the primary sources which say "we are the good ones; they are trolls" is insufficient. -M.Altenmann >t 17:31, 1 February 2015 (UTC)
Masem, this has been discussed quite extensively. Do you have any specific suggestions for article changes?
Peter Isotalo 05:27, 1 February 2015 (UTC)
See the section this is in, where I present the change I made in the lede, as the current issue, but this remains a larger problem in trying to organize the article in an objective manner. --MASEM (t) 05:38, 1 February 2015 (UTC)
Masem, your interpretation of the reliable sources and how they should be presented has been discussed extensively on this talk page already, and a large number of people have disagreed and continue to disagree with your interpretation. It might be time to consider that, since it looks like this same conversation has happened at least three times over the past few weeks and the consensus does not seem to be in your favor, that your interpretation is faulty. Kaciemonster (talk) 15:21, 1 February 2015 (UTC)

#NotYourShield Reversion Explanation

I've reverted an edit, seen here [15], to the section concerning #NotYourShield and the IRC logs posted by Quinn. I think it's an edit that needs more discussion before it's added to the article body (if it is), and I also think it's poorly worded (and possibly sourced) as is. They also seem to be an unnecessary skewing of the article's wording.

For one thing, as far as I can tell (not speaking French), the sources only claim that "4Chan has released what they claim to be the full chat logs from when the IRC channel that was created on August 17th." I'm also not certain that this edit (and arguably some of the rest of the paragraph) doesn't give too much weight to the chat logs, in what is ostensibly a section about #NotYourShield as a whole.

It's also worth noting that Shadowrunner56 made these edits, without explanation or discussion in the summary or on the talk page, after a reversion of previous similar (but less extensive) edits by Peter Isotalo. AtomsOrSystems (talk) 04:12, 1 February 2015 (UTC)

The French article wrote only about logs released by Quinn. So I removed one of footnotes. -M.Altenmann >t
re:What quinn posted: actually, she posted not logs, but excerpts from logs, which are "logs of discussion threads", and IMO may be safely called simply "logs". As for screenshots or not, it is a nitpicking. We write "This is a photo of John R", not "this is a png image of the photo of Jonh R." -M.Altenmann >t
re:what 4Chan posted: I agree this is rather irrelevant, since there is no discussion what exactly was posted and how it is relevant to the section. -M.Altenmann >t 05:07, 1 February 2015 (UTC)
In summary, I support the revert of the unnecessary increase of verbosity. -M.Altenmann >t 05:07, 1 February 2015 (UTC)
I didn't think my small edits really needed to be discussed as I was only adding info from sources that were already in the section. But I'm alright with my edits being discussed further.
Also, to answer some of your concerns, the Escapist says the same thing about Quinn's spotify collection of screenshots, "Quinn claims the chat logs she released proves", not just with the 4chan member released logs (and think other sources in the section state "Quinn claims"). I also agree with the notyourshield section needs work and refinement in general, not just with the chatlog bit, but I thought any more edits from me would've needed a talk page section from me explaining myself. Shadowrunner(stuff) 05:31, 1 February 2015 (UTC)
We already include the sentence Members of 4chan have said that some information has been taken out of context or misrepresented. Why do you think we need to extend the sentence.
Also, we currently have Quinn's release worded as ...which she said showed..., as you note. If we were to add more about 4chans release of logs (I don't believe we need to), we would need to word it similarly, in line with the sources. AtomsOrSystems (talk) 18:05, 1 February 2015 (UTC)
Well the full chat logs were released as a direct response to the screenshots of the IRC she released herself. I thought that would be some good information to put in, and it is supported by the Escapist article. I'm fine with wording it similarly, that's reasonable. Shadowrunner(stuff) 20:36, 1 February 2015 (UTC)