Slashdot Mirror
How Many Hoaxes Are On Wikipedia? No One Knows
An anonymous reader writes The Washington Post's Caitlin Dewey has written a lengthy feature covering one of Wikipedia's most intractable problems: carefully inserted hoax information that is almost impossible to detect. Dewey's investigation starts with the recent discovery of the nonexistent Australian god "Jar'Edo Wens" (which lasted almost ten years), and discusses a Wikipediocracy post about a recent experiment by critic Greg Kohs, in which 30 articles received cleverly-chosen minor falsehoods. More than half survived for more than two months. Included is also a chart showing that editing participation in Wikipedia has "atrophied" since 2007. It is quite rare to see a feature in a major media outlet as critical as this, of Wikipedia and its little-known internal problems. Especially on the heels of a very favorable CBS 60 Minutes report. As Kohs says, "I think this has proved, beyond a reasonable doubt, that it's not fair to say Wikipedia is 'self-correcting.'"
It would be interesting to see them compare to other sources.
... Included is also a chart showing that editing participation in Wikipedia has "atrophied" since 2007 ...
As Kohs says, "I think this has proved, beyond a reasonable doubt, that it's not fair to say Wikipedia is 'self-correcting.'"...
I could have told you that, and have been telling you that.
.
The big problem with Wikipedia is that in spite of what the publicity says, it is only a small number of people who contribute, and a surprisingly large number of those people have an agenda for what they edit.
imo, with Wikipedia, truth is not the goal. A certain point of view is the goal.
pointing to corrections that haven't been done yet doesn't mean anything. if something is obscure and unimportant it can persist for years, with no impact. and then it's corrected. if it's important, it will probably be corrected in days or minutes
can anyone point to any other media that this isn't true about? (i'm not talking about corrections, that may never be made, simply that all media has a backlog of errors that need correcting)
and questioning wikipedia's veracity, alone, has no value
judge it against other options and their veracity
the traditional encyclopedia is subject to the editorial whims of professionals, and professionals can have agendas and are not automatically superior to a mass of impartial folk. emphasis on "mass." as thousands of editors, even if there's been a drop in participation, is superior to an overworked few with questionable biases
and please note we're talking about brief introductions to topics, not deep dives into esoteric academic specialties. wikipedia is never intended as a replacement for serious texts on topics. and if someone is relying on wikipedia alone for vital topics, that's the reader's fault, not wikipedia
wikipedia's innate superiority is the same reason we have juries instead of professional judges. professional judges can start deciding cases based on having something to prove: "i'm finding this guy guilty because i made the previous guy innocent" or "this guy is clearly innocent, but it's important to send a message, so i'm finding him guilty"
certainly, a million examples of bad juries can be found. we can find problems with the jury system that are truly horrible
as if that means anything. because all other options are worse
this is classic form of propaganda, half-truth, cognitive fallacy: criticism in a vacuum
outside of the context of other choices, anything can be made to look like shit
for example, we can criticize all sort so problems with democracy. there are many problems with democracy and they are real and major. it's just that our other options are clearly worse
likewise with wikipedia: you can list thousands of things wrong with wikipedia, some truly horrendous
but it's still superior to what came before and other current options
intellectual property law is philosophically incoherent. it is your moral duty to ignore it or sabotage it
.
It is easy to use Wikipedia,
It is that ease to use, rather than accuracy, that has made Wikipedia as popular as it is.
The obnoxious cliques of senior editors with god complexes make it virtually impossible to correct anything of substance. And Jimbo cares fuck-all about it as long as enough people click the donation button.
Sure, you can get into revision wars over whether to use the word "which" or "that" in a given context; but fixing a factual error? Good luck!
"Citation needed!"
"But the old, wrong version didn't have a cite either."
"Doesn't matter, it stays, and my minimum wage burger flipping ass has just banned you for daring to challenge me, you pompous PhD-wielding expert in this particular field!"
In many of the more relaxed civilizations on the Outer Eastern Rim of the Galaxy, the Wikipedia has already supplanted the great Encyclopedia Galactica as the standard repository of all knowledge and wisdom, for though it has many omissions and contains much that is apocryphal, or at least wildly inaccurate, it scores over the older, more pedestrian work in two important respects.
First, it is slightly cheaper; and secondly it has the words Don't Panic inscribed in large friendly letters on its cover.
All those moments will be lost in time, like tears in rain.
If people want to monkey with Wikipedia, have at it. We're told over and over again that Wikipedia is not a suitable reference; however the references on the page can sometime be useful.
And then there's http://www.dailydot.com/lol/am...
The person in the story inserted a little fake factoid into an otherwise proper article. This little factoid ended up very quickly
- cited in a lesson plan by a Taiwanese English professor
- cited in a book about Jews and Jesus
- cited in innumerable blog posts and book reports, as well as a piece by blogger Hanny Hernandez, who speculated that Amelia Bedelia’s tendency toward malapropisms was inspired by Parish’s experiences in Cameroon, as “several messages can be
misinterpreted between a Cameroonian maid who is serving an American family.” One blogger even speculated that Amelia Bedelia wasn’t a maid, but a slave.
- cited in the Amelia Bedelia entry on the website TV Tropes and Idioms, and Peggy Parish’s Find-A-Grave page
- cited by Mr. Amelia Bedelia himself: Herman Parish, Peggy’s nephew and author of the books after his aunt passed away in 1988, who apparently told a reporter from the Waterloo-Cedar Falls Courier that his aunt based “the lead character on a French colonial
maid in Cameroon.”
Once again, Wikipedia can be a useful overview of a subject and a launch-pad for further research. But after all these years of Wikipedia hoaxes (and Wikipedia maintains a list of hoaxes; http://en.wikipedia.org/wiki/W...), the mantra must be "trust but verify".
Because, in Wikipedia's own words:
Misinformation on Wikipedia misleads readers, causing them to make errors with real consequences, including hurt feelings, public embarrassment, reprints of books, lost points on school assignments, and other costs. With some articles, like medical topics, they could lead to injury or death.
Having worked on this problem for a while, ive found exactly 5 total hoaxes on wikipedia (no more.) Please remove the following articles:
1. Edward Snowden: is not actually a person, this is an old wives tale. E. Snowden is a hybrid cultivar of the genus Aechmea in the Bromeliad family.
2. 9/11: Although commonly thought of as a terrorist attack that claimed the lives of several thousand americans, this too is just a silly rumour. 9/11 is a town and municipality located in the province and autonomous community of Navarre, northern Spain.
3. Barack Obama: This fools several laymen and scholars alike! Obama isnt a president, but was a motor race set to Formula One rules, held on 30 July 1950. The race was won by Argentinean driver Juan Manuel Fangio after a distance of 68 laps.
4. Christmas: Again, not a holiday at all. He was actually a Polish Air Force Captain and Allied double agent during World War II, using the codename Brutus. After having been offered safety by the Germans, he was sent to England as an agent. However, he made himself known to the British authorities. He was de-briefed by the British (MI6) and Polish authorities about the security lapses of his organization in France. And thats why we have Christmas trees today!
5. Computers: could NEVER have been real, and most of us know this one to be true. The computer is actually a Ukrainian professional football coach and a former player. As of 2009, he works as an assistant coach with FC Dnipro Dnipropetrovsk.
Good people go to bed earlier.
I have to agree. The experiment cited modified 30 articles with minor and cleverly-chosen falsehoods, and more than half were fixed within two months.
From that, Kohs then claims, "I think this has proved, beyond a reasonable doubt, that it's not fair to say Wikipedia is 'self-correcting.'"
Um, WTF? That statement proves he's not very good at making accurate statements. If he added a time period to that, and maybe some disclaimer about the popularity of articles being modified, then it wouldn't be much of a point, but it'd be closer to true.
You conveniently leave out that Kohs himself (that's me) reverted 6 of the false edits himself, when it became clear after more than 45 days that nobody else was likely to do it anytime soon.
You're also being rather unfair to say that Kohs' statement -- a sound-bite that the Washington Post journalist selected from a lengthy telephone interview -- which you then rip away from the context of the article, "proves" that I'm not good at making accurate statements. Why don't you peruse the actual spreadsheet of the 30 articles, which speaks for itself on my ability to convey accurate information?
The Coati (a small member of the raccoon family native to Brazil) is also known as the Brazillian aardvark. The reason that it's known as the Brazillian aadvark is that someone made the phrase up and added it to Wikipedia - but the coinage gained traction, because journalists copied it, and this led to a citation for that name being added to the article. Now wikipedia is in a quandary... there are, thanks to lazy journalists, people who know the coati as the Brazillian aardvark, because they read that in a newspaper... so is the hoax now true?
Does it become true if the dord of references to that name reaches a certain level?
Does it become false even though people do use the term, just because the etymology of the word was a hoax?
A pizza of radius z and thickness a has a volume of pi z z a
Oddly enough, people question the Wikipedia when it gives more information about the providence of the writing and content than virtually any other source, yet people insist upon making blanket statements about how unreliable it is. All that really says is that people want an authoritative source rather than a verifiable source. They want someone to tell them what is "true" rather than giving them the tools to assess what they are reading.
I get tired of reading apologetics for Wikipedia that present false dichotomies.
"Oh... but things used to be so much worse with other sources which were so bad and evil and... so be grateful for the crap that is Wikipedia today!!"
Nonsense.
Look, just because other things may have been bad or worse in some ways doesn't mean we should accept the stupidity that is Wikipedia.
For example -- there are already vetting processes in place (somewhat) for getting articles approved on Wikipedia as having "good" status, etc. These review processes generally involve doing such things as checking to see sources actually correspond to what's in the article, etc.
Here's the obvious question -- why not actually create "stable" version of "good" articles, and have that be the default page for that article? Make it so such articles can no longer be edited directly -- instead, anonymous editors and random users can edit an "unstable" or "testing" or whatever version of a page that can be easily accessed through a tab (like the talk and history pages, etc. can be now). Periodically an established editor can clean up such proposed edits and migrate the good ones to the "stable" page.
Ideally, approved edits to the stable page should be approved by a consensus of editors for that particular page, some of whom may actually be experts in a subject. I know a LOT of academics who spent a little time here or there and tried to edit Wikipedia, because they actually would like to see it made better, but they get driven away by the politics and bureaucratic nonsense. What if we could actually get them involved? What if they could actually be verified and help to determine what makes it into the "stable" versions of articles that they actually know something about... you know, like old-style encyclopedias used to.
But there could always be checks and balances -- edits need to be approved by 2 or 3 editors with an appropriate level of "clearance" for that page, for example. There are many ways of working out the exact details, but something like this could raise the quality level of central articles significantly -- those stubs on the fringes can still operate as the "wild west" where anybody can edit the live page until someone creates a stable good version.
You don't need to give away authority completely to experts -- have mixtures of experts and other editorial staff able to approve edits. You don't need to lose the tracking information. In fact, you get even MORE tracking information, and you get a "sandbox" for stable pages for better versions to be worked out and incorporated into existing articles.
Or whatever. I don't claim to have all the answers. But I do know that there are serious and legitimate criticisms of Wikipedia's model, and some aspects are just going to get worse (and some have been for some time). Anonymous "wild west style" editing for just about anything was a great way to crowdsouce and build a resource... but it's time to hone this into something better, and that requires people with real skills: subject matter experts, experts in editing, etc.
We should never shut the contributions from crowdsourcing out -- but we can still improve the Wikipedia model while not falling back onto old crappy models either.