This report reminds me of some of Jeffrey Lin (Riot Lyte)'s work on League of Legends. He found that 1% of players are toxic, frequently acting badly. But that only accounts for 5% of toxic behavior; 77% of bad behavior comes from people who are just having a bad day. That echoes this report's finding that 34 Wikipedia folks are responsible for 9% of abuse.<p>A related finding is Riot found that toxicity in LoL was contagious; people who played with an abusive player were more likely to be abusive in their next game. The Wikipedia phenomenon of abuse pile-ons seems similar.<p>Some links for Lin's work:
<a href="http://www.nature.com/news/can-a-video-game-company-tame-toxic-behaviour-1.19647?WT.mc_id=SFB_NNEWS_1508_RHBox" rel="nofollow">http://www.nature.com/news/can-a-video-game-company-tame-tox...</a>
<a href="https://www.youtube.com/watch?v=nbYQ0AVVBGU" rel="nofollow">https://www.youtube.com/watch?v=nbYQ0AVVBGU</a>
From my brief experience with (English and Korean) Wikipedia editing many years ago, the problem was not "toxic" editors but more with entrenched ones. They never resorted to personal attacks, but instead used their vast knowledge of WP:this and WP:that to provoke other people until either they left or get banned.<p>E.g., for a long time, the article title in the Korean wikipedia for Japanese Empire was 대일본제국, literally, Great Empire of Japan. Well, this being Korean wikipedia, you could see how some people took offense at it. The sensible thing would be just to drop the "Great" part: everybody knows which Empire we're talking about, nobody thinks the Empire was particularly Great anyway, and then we could move on to actually talking about the article's content.<p>But no, some editors would object, because we have to use Official Names(TM) for everything. (For some reason, nobody bothered to change Bangkok's name to the official "Krungthepmahanakhon Amonrattanakosin Mahintharayutthaya Mahadilokphop Noppharatratchathaniburirom Udomratchaniwetmahasathan Amonphimanawatansathit Sakkathattiyawitsanukamprasit", but if anybody did that I'm sure it would be quickly reverted, citing WP:(You can't make an edit just to prove a point) or something like that.) To this day, I suspect some editors' criterion for "Official Names" and other similar policy was maximal likelihood for alienating potential contributors. Classic power game move.
Not terribly surprising. But looks like the system manages to handle
them pretty well - the paper shows that the probability of getting
blocked is up to ~80% after 5 personal attacks.<p>I've never worried too much about that kind of incivility or 'abuse' -
I've always been far more worried about the 'abuse' which takes the
form of deleting contributions, which, because it can be justified and
rules-lawyered and is not as clearcut as 'you suck', the system does
not handle well and culturally encourages.
An impressive amount of information on Wikipedia is just outright wrong, misleading, clearly biased, or astroturfed. At the rate in which I encounter inaccurate data on Wikipedia that I can identify, it makes me suspicious of nearly everything on the site. It's an interesting experiment to allow the masses to create and edit their own facts and it can provide a broad (albeit often inaccurate) overview of some topics, but it's hardly a substitution for a legitimate source let alone encyclopedia.<p>Nonetheless, I know plenty of people who take it as legitimate and cite it as if it's an authority.
I suspect the sentence "A handful of 'highly toxic' <members> cause <overly large> amounts of abuse." is true for all groups.
"Almost immediately, they found that they could debunk the time-worn idea that anonymity leads to abuse. Although anonymous comments are "six times more likely to be an attack,"<p>Hmm, that actually DOES seem to support the idea the anonymity fosters abuse...
I expend a lot of effort to avoid Wikipedia as a bona dude source of truth, and man is it difficult. The desire for a single easy-to-access repository of knowledge is incredibly strong, but over the years, I've been stung repeatedly by this particular one.<p>As it turns out, the "Neutral Point of View" rule that Wikipedians so proudly tout is quite strongly biased toward a specific world-view: that of the so-called "progressive" mindset.<p>It doesn't matter how much value a system of ideals has; if it transforms a purportedly neutral source of information into a dogmatic one, then the source loses all credibility and usefulness.<p>This seems to be happening on Wikipedia at an alarming rate, which is a real bummer. It's not as ridiculous and blatant as (e.g.) Conservapedia, but... well, frankly, that makes it all the more concerning. Zealotry is way harder to call out when it's cloaked, as Wikipedia tends increasingly to be, in the sly guise of unassailable moral high ground.
"Depressingly, the study also found that very few personal attacks are moderated. Only 17.9% of personal attacks lead to a warning or ban." I'm not sure, but it seems like 17.9% is probably close to the right rate for moderation, not "depressing" or "abysmal". Especially if there are calm comments from experienced users disapproving of the abusive behavior in many of the other cases.
This is very interesting! So if anonymitiy isn't correlated with abuse, then what is? I believe the answer is community standing. Anonymous and low-ranking users know they will suffer the consequences for foul language therefore they moderate themselves. Users with thousands of edits to their name, knows they will get away with it.<p>About ten years ago when I edited Wikipedia, there was one particularily nasty user who accused everyone who didn't agree with exactly everything he wrote. I belive this person was mentally ill, but he spent a lot of time on Wikipedia and wrote a lot of content.<p>Several people brought complaints about him and his toxic behaviour, but he always had the backing of Jimmy Wales (the leader of the project), who would make excuses for him. If the community intervened against this person by for example temporarily banning him, Wales would intervene and undo the ban.<p>So this person kept being nasty because there were no repercussion. Eventually I believe he got tired of the Wikipedia project and left on his own violiton.
Abuse and harassment on Wikipedia does not belong to a handful of editors, it's endemic, because the only way to succeed on Wikipedia <i>is to be an asshole</i>. But it's important to note that bad behaviour on Wikipedia is a symptom of poor software design. The software that Wikipedia is built encourages conflict, and uses hundreds of rules combined with poorly-trained admins to try to patch things up. The boundaries between conflict and abuse are wide, arbitrarily defined, and patrolled by people who have neither training, nor consistency. A better designed system would install systems to minimize conflict, and would give clearer rules and standards to it's moderators.
So - how do you rate partisan editors in geopolitical conflicts? The Irish Troubles, Russian Federation and especially Israel-Palestine come to mind. Especially the Israeli side was very successful in casting a very favourable light on its position, and it took great effort to rein them in a little. Would the pro-Israel faction be called abusive under the present metrics?
Why do threads like these talking about a specific aspect of something always devolve into hatejerks about the subject? I've noticed the same thing on threads about Apple as well. It's talking specifically about personal attacks yet people are somehow relating that to the reliability of the site. It's not wrong to criticize Wikipedia, but then you should focus on the topic of the toxic atmosphere instead like how that might compromise neutrality by keeping out editors who are more conflict-averse. (or actually back up how you think the site is unreliable, considering a study in <i>Nature</i> found Wikipedia to be just as reliable as <i>Brittanica</i>)
Ah..Wikipedia. Nice to know my HN Username is relevant for this thread.<p>Jokes aside, Wikipedia has a ton of flaws and the abuse is definitely one of them regardless of what percentage of people do it, but there are just no other realistic substitutes that are actually large enough to matter. The largest and most respectable competitors are Everipedia (only one that doesn't reuse mediawiki software), RationalWiki, InfoGalactic, and some other rolled mediawiki crap. And none of those alternatives are even Alexa top 1,000 let alone Alexa top 100 traffic.
One has to wonder about the logic of unemployed lute players (people like most armchair anchor "journalists" without a shred of topical understanding) given power to referee edits on highly-scientic topics or judging vital concepts/details "irrelevant" because they don't understand historical context or subject matter.
This is an example of what Jo Freeman famously called "The Tyranny of Structurelessness": <a href="http://www.jofreeman.com/joreen/tyranny.htm" rel="nofollow">http://www.jofreeman.com/joreen/tyranny.htm</a>
Somewhat tangential, for those who think Wikipedia's ideologically-driven revert wars are anything new, a 19th century instance I ran across some months back involving Chamber's Encyclopaedia:<p><a href="https://www.reddit.com/r/dredmorbius/comments/4xe2k1/chambers_encyclopaedia_editorial_statement/" rel="nofollow">https://www.reddit.com/r/dredmorbius/comments/4xe2k1/chamber...</a>
Good article, but I wish she'd reached out to Wikipedia or the community of editors to solicit some answer on what they intend to do about it. Once one is aware of such an issue, failure to wield the banhammer is treated as an implicit endorsement.
If you want to see for yourself the seedy underbelly of Wikipedia, check out <a href="http://wikipediocracy.com/" rel="nofollow">http://wikipediocracy.com/</a>