Wikipedia to begin color coding untrustworthy text

Hot on the heels of Wikipedia's decision to begin using editors for certain pages, they've announced another method of attempting to keep things accurate on their website. According to Wired, the organization will begin color coding the text of their entries in orange, to indicate untrustworthiness, based on a couple of factors.

The feature will be aptly named "WikiTrust", and will color the text after analyzing the reliability of the author in question, as well as how long the text has been on the entry. Virgil Griffith, a Wikipedia software developer and neuroscientist at the California Institute of Technology, said, "They've hit on the fundamentally Darwinian nature of Wikipedia. Everyone's injecting random crap into Wikipedia, and what people agree with more often sticks around. Crap that people don't like goes away."

It should be noted that this algorithm has actually been available since November, last year, in the form of a MediaWiki plug-in; it's only now that the WikiMedia Foundation has decided to roll it out to the full site, with changes starting to happen this fall. A computer scientist named Luca de Alfaro was the one who originally had the idea, and thus he lead the WikiMedia Project; he said, "Online collaboration is becoming more and more central to the way in which knowledge is created and assembled worldwide. There are more and more services that simply cannot exist without some notion of user reputation and trust in the content."

Work is progressing nicely on the project, with algorithm speeds increasing at a huge rate, so it'll be exciting to see how it fares when it's rolled out to the millions of current users.

Report a problem with article
Previous Story

Adobe announces no CS3 support on Snow Leopard

Next Story

Stolen iPhone recovered using "Find my iPhone"

26 Comments

View more comments

lee26 said,
Why?

Surely it's better to just delete stuff that's wrong and correct it rather than keep and highlight it.


A good question, but I think they're targeting 'untrustworthy' text; after all, you could have a guy with a PhD sign up and edit an article in his field of study, but people don't know that he has a PhD, so this is more of a case of guilty until proven innocent. It could be right, but it's just... unconfirmed that it is

Sam Symons said,
A good question, but I think they're targeting 'untrustworthy' text; after all, you could have a guy with a PhD sign up and edit an article in his field of study, but people don't know that he has a PhD, so this is more of a case of guilty until proven innocent. It could be right, but it's just... unconfirmed that it is :)

imho, i think that part of having a PhD should be the ability to convince others that you are right... not just throwing out facts which you know, but properly presenting it like you do in a proper journal, with citations and references and all that...
if the text is orange but its content is really good, i'd say that'd better stay as a rare occurence... it does beat manually checking the edit histories of the articles, tho...

lee26 said,
Why?

Surely it's better to just delete stuff that's wrong and correct it rather than keep and highlight it.

And to remove a big part of the wikipedia?. I don't think they will take such posture. Wikipedia puts quantity over quality.

How are they seriously trying to improve?.

Switching form tags to color is not that i call a real improvement.

Take note that many editors are kids in disguise.

ermax said,
I thought the standard rule of thumb was to not trust anything on Wikipedia? :)

Not really.

It's "use wikipedia with care" or something along that line.

ermax said,
I thought the standard rule of thumb was to not trust anything on Wikipedia? :)

Please, tell me why Wikipedia is any different to any other random site on the internet that people would get their information from by searching with Google?

Funny how in schools we can reference any website whether or not it's regarded as trustworthy, but if it's Wikipedia, it's a no-no. (Australia)

Mikeyx11 said,
Please, tell me why Wikipedia is any different to any other random site on the internet that people would get their information from by searching with Google?

Funny how in schools we can reference any website whether or not it's regarded as trustworthy, but if it's Wikipedia, it's a no-no. (Australia)

Because most of the sources are peer-reviewed, academically revered or just generally accepted as true (i.e. press releases)?

booboo said,
UK as well. They will go ape **** if you refer to a Wikipedia page.

That is why you use the references for the wikipedia page as your references Job done!

The Teej said,

Because most of the sources are peer-reviewed, academically revered or just generally accepted as true (i.e. press releases)?


Obviously you haven't seen much of the internet.

Mikeyx11 said,

Obviously you haven't seen much of the internet.

There are some sites, mainly university websites where the content is pretty much golden.

If you use untrustworthy references - that's up to you, they hate wikipedia because its the most commonly abused one, so they tell you not to use... they could go on for ever with a list, but what would be the point? If they see you are using cr*p websites they are probably lowering you mark for it anyway, just they can't pre guess website "xzy" before you start writing. If you want to see the source of the wikipedia information then find the original reference and question its legibility.

There are many trustworthy websites out their for finding information...

I have only ever used a handful of online sources - two were digital version of reputable specialist journals related to my course. These are all subjective but all written by people recognised in their field.

A couple more were government websites reading up on laws and policies to my field, and then a websites belonging to a statutory regulator and professional body. None of these are opinion, instead all written requirements.

The reason books can generally be trusted more is because they are not published on mass unless trustworthy - and you university or bookshop is unlikely to stock a book again unless it is reputable. (of course some subjects are a little more questionable than others). Whereas on-line - anybody could have created the website, ranging from a kid hating his subject at school, to a retired person spreading wartime propaganda.

Mikeyx11 said,
"on mass"... LOL

oops typo, I should proof read - who knows someone could take my comment as a reputable source in an essay.

mmck said,
oops typo, I should proof read - who knows someone could take my comment as a reputable source in an essay.

If they are talking about people's bad grasp of English, then using your comment as an example would be fine

I don't get why people are so concerned over accuracy on Wikipedia. I've been using Wikipedia exclusively for all my schoolwork since 6th grade, and it's never failed me yet!

DanielZ said,
I don't get why people are so concerned over accuracy on Wikipedia. I've been using Wikipedia exclusively for all my schoolwork since 6th grade, and it's never failed me yet!

It's because wikipedia attack the ideology that one should make money out of everything even informations. An ideology with a lot of advocate here on Neowin.

LaP said,
It's because wikipedia attack the ideology that one should make money out of everything even informations. An ideology with a lot of advocate here on Neowin.

Hey, this sound like a tin-foil hat conspiracy. AFAIK there are a lot of libraries for free around the globe, and i don't find any (evil scheme) trying to close it.

Magallanes said,
Hey, this sound like a tin-foil hat conspiracy. AFAIK there are a lot of libraries for free around the globe, and i don't find any (evil scheme) trying to close it.

The tin-foil hat is overrated. Few people see conspiracy everywhere.

It was mainly a joke. But you have to wonder what is the motive of someone who post things like this "I thought the standard rule of thumb was to not trust anything on Wikipedia? " (which was probably a joke too).

I too used Wikipedia for a lot of home works and for the most part the informations there was very accurate.

A "real" Encyclopedia might have a little less errors but might feature outdated informations. I doubt if you go to a library, open a book and search for Roger Federer that you will read he his to men with the most glam slam victory of modern tennis.

LaP said,
I doubt if you go to a library, open a book and search for Roger Federer that you will read he his to men with the most glam slam victory of modern tennis.

But who on earth is writing dissertations on that? lol

Sorry to be sarcastic I do get your point, but generally "news" can be found in specialist journals or individual publications. which at University and beyond you should be reading anyway.

A book Encyclopaedia is a poor comparison, as unless a specialised one or a complete set which some are huge, they are generally too vague to use for essay writing. I have an encyclopaedia about one persons work, it is bigger than A3 in size and about 3 inches thick - still not even that is comprehensive. Anyone who uses "Wikipedia exclusively" should probably still be in 6th grade.

I don't think it's a big deal. They already tag sections that are under review. This will make it easier to see that the information may not be accurate.

Wikipedia is a great resource, especially for nerdy things. But it does suffer from Wiki-fascism, where people seem to think they own the page. I often find the discussion page more interesting as the article itself. There you get real opinions, rather than stiff attributable stuff, which constantly takes you back to the traditional "gatekeepers of wisdom", the ones that have built the hypocritical mess we now live in. George Orwell was right!

"In a time of universal deceit, telling the truth is a revolutionary act." - George Orwell.

Commenting is disabled on this article.