Wikipedia Actively Battling PR Sockpuppets 166
Nerval's Lobster writes "Over the weekend we discussed news that PR firms have been selling their ability to modify Wikipedia entries to help clients clean up their image. Now, the Wikimedia Foundation's executive director has confirmed that Wikipedia editors are actively engaged in a wide-ranging battle against those PR firms. Over the past couple weeks, those editors have isolated several hundred user accounts linked to people 'paid to write articles on Wikipedia promoting organizations or products,' according to Sue Gardner. Those users' accounts violate Wikipedia's guidelines, 'including prohibitions against sockpuppetry and undisclosed conflicts of interest.' Some 250 suspicious user accounts have already been nuked. Correcting biased text is a thankless job for those Wikipedia editors — the literary-world equivalent of killing endless hordes of zombies approaching your protective fence. But that job gets even harder when a PR agency deploys dozens, or even hundreds of writers to systematically adjust clients' Wikipedia pages."
Wikipedia is an MMO (Score:5, Interesting)
http://en.wikipedia.org/wiki/Wikipedia:Wikipedia_is_an_MMORPG [wikipedia.org]
Thankless. Well let me be the first to say (Score:5, Interesting)
Microsoft have been doing this since 2007 (Score:5, Interesting)
To show there is nothing new under the sun...ironically from the wikipedia entry "On January 24, 2007, Rick Jelliffe made claim on his blog that a Microsoft employee offered to pay him to make corrections in Wikipedia articles concerning Office Open XML" http://en.wikipedia.org/wiki/Criticism_of_Microsoft [wikipedia.org]
Vandalism (Score:2, Interesting)
If they are intentionally editing the site to delete factual information or add fake information. Couldn't the constitute vandalism which they operators of Wikipedia could sue them for? Especially since they are doing it "For Profit" so they can't say they did so erroneously or anything?
Watch them get hit with a steep judgement against them PER ATTEMPTED EDIT and you will see them put a stop to it fairly quit I would think.
Free Market (Score:0, Interesting)
In action
3 domains of verifiability (Score:5, Interesting)
To speculate - I've noticed that articles on wikipedia fall into the three broad categories, unsurpisingly the same as those of nouns: 1) people, 2) places, 3) things.
3) "Things" articles are the 'simplest' to disentangle or find the truth of because "things" include works of art, mathematics, science and physical objects. It's easy to tell a lie when an article says that the hit song, "I feel good" is written by Mozart, or when an article claims gravity makes things fall up. These claims are relatively easy to test or refute. The problem with these types of articles is they may require some real expertise (especially philosophical and mathematical articles) to verify - but that is also their virtue
2) "Places" articles (which include 'historical events') become more difficult because often these places do not exist any more, or the events usually have already happened, usually a really long time ago. These articles suffer the classic problems of history multiplied by the power of the internet.
1) "People" articles. These articles are rife with arguments over what actions events in a person's life are significant, and what elements of those actions are significant. The words chosen to describe a person can make all the difference - he was a "Great Leader" or he was a "Good Leader" - which one best describes Hitler*?
So I would venture that this firm has targeted articles in categories 1 and 2, although I guess there maybe product articles in category 3 which could be gainfully modified.
Ah the truth of things. And the relative truth of places and people ;)
*you graciously forgive this overused example.
I'm seeing this more on the biography side (Score:5, Interesting)
Some of the paid PR I've seen recently has been on biographies of living persons, especially rich ones. Lots of happy talk about their charitable work and affiliations gets put in. Stuff about their career failures, lawsuits, and criminal history gets taken out. This is tougher to fight, because Wikipedia has a "biography of living persons" policy which discourages negative comments for anything short of a felony conviction. (Even after a felony conviction, sometimes.)
On the product and business side, though, pushing back against paid editing usually works.
Wikimedia could copy StackOverflow's process (Score:4, Interesting)
We all know how StackOverflow works. You can always ask or answer a question, but other privileges are based on your reputation. Reputation is only gained by creating good questions and answers. It takes work to get a good reputation on StackOverflow.
I actually don't know what Wikimedia has in place, but it could implement a similar reputation based approach as StackOverflow. Of course the algorithm and mechanism would be different since Wikipedia is not a Q&A site.
To make things even harder, they could implement a reputation killer. For example: if user A improved the score of user B, then user B gets nuked because they are one of these PR firms, then user A should suffer a major reputation dive.
While I'm at it... Instead of Wikipedia begging for money once a year, they could implement a wikipedia.com site which has some light advertising. By default all users go to wikipedia.org, but for those who want can manually redirect to wikipedia.com. I would be glad to support such a system.
Re:Wow, I'm going through this now.... (Score:4, Interesting)
Dude, they reverted your posts because what you posted was flat-out wrong, not because they are shills. You stated that Dan Pulcrano owns backpage.com, but he doesn't own it, operate it, or have any direct control over what goes on it. His newspaper does business with it, but that is a far cry from what you actually posted.
Totally different suggestion... (Score:5, Interesting)
On the other hand, there was the case of a German politician where Wikipedia got the name wrong - and wouldn't accept his statement what the correct name is. So it would be really good if that person could add a paragraph saying "these Wikidiots got my name wrong, and here's the correct name..."
Re:3 domains of verifiability (Score:5, Interesting)
If you read some press stories about the sock puppet companies they are mostly targeting products or corporations, attempting to manage commercial reputation.
In truth, this is not always unwarranted. When someone writes about the rash of Brakes failures on Toyota vehicles, the company ought to be able to have a clearly labeled Official Response position, rather than having them feel forced to resort to sockpuppetry to get some actual facts, or corporate nattering as the case may be, across.
Having policies in place that forbid official statements just begs for sockpuppet tactics. There have been cases where authors of books had their remarks removed because they were not considered a credible source [arstechnica.com] for information about their own books.
Furthermore, it would seem that as long as the sockpuppet was indicated, in a foot note, as a paid source of a product or company, that fact alone should make the posting more official and credible rather than less so. Who is more authoritative on the capabilities of a product than the manufacturer? Some random user perhaps?
Re:Wikimedia could copy StackOverflow's process (Score:4, Interesting)
We all know how StackOverflow works. You can always ask or answer a question, but other privileges are based on your reputation.
I'd hardly call that a model, it rapidly becomes as annoying as expert sex change.
Half the questions are stuffed with meta bickering about the "rules", who should get reputation, who shouldn't, disputes over whether the question is too much like some other question, a bunch of asshats duplicating and expanding on the same answer while trying to out-answer it to game the system for reputation.
I think my least favorite though is "not constructive". Yes, yes, it doesn't' have a definitive answer, but the ensuing discussion is generally pretty enlightening, and should be encouraged. And most of the real "hard" programming stuff falls into that category. (How to name things well for example (be it classes, database tables, interfaces, etc...)
No, actually that's wrong, my least favorite thing about stackoverflow is shitty hacks being up-voted. The C stuff about strcpy strncpy, or snprintf vs _snprintf vs sprintf_s etc is full of just really piss poor advice as a for instance. There's good info mixed in, but the bad never goes away and some of its rated really high.
If wikipedia followed stackoverflows model, every article would be 50 pages long of competing articles, voted up and down by
Re:Wikimedia could copy StackOverflow's process (Score:3, Interesting)
First sock-puppets and then the OCD admins (Score:5, Interesting)
Then you get admins who "own" an article. So the new census will come out and someone will update the population of a city to reflect the new census and even cite the new census; but 5 minutes later the old numbers are back and the citation gone.
Then you get excellent articles filled with excellent information but some OCD twerp is threatening to delete it due to formatting issues.
Lastly you get the show off types. This is at its worst in mathematics. There is almost zero educational value in many of the mathematics related articles. The mathematics are perfect but instead of making things clear they use the most esoteric terminology/symbology. So 2+2=4 will become something like:
Using a Yungra transform you can compute that the Dirac set of {1, 1} when concatenated with the set Ramublablajan set {1,1} (each of which represents the empty set {0}) will have a resultant set of the Miller-Shiefler Series {1, 1, {0}, 1}.
An example of a complicated thing being made simple would be the article on RSA. The example math they use could be done using a pen and paper. Most Wiki Admins and editors would seem to despise this sort of simplicity and instead would probably rewrite the entire (excellent) article as a single formula that concisely sums up RSA. I personally prefer the sing-along version that is there now.
Re:Wikipedia editors? (Score:5, Interesting)
I take it that you've never maintained any sort of wiki for more than a few minutes. I dare you to start your own wiki on any topic of your own choosing and try as you might to keep the cruft off of it. Use your own definition of cruft as well, but penis enlargment pill links thrown into the middle of a discussion of Princess Leia might easily fall into most typical definitions I can imagine. Going to a place like wikia.com is one way to start your own wiki, although other wiki hosting services are plentiful.
After awhile, you will find that random nonsense and pure junk is the rule rather than the exception, and for high profile wikis like Wikipedia really get some absurd junk. For those poor souls who make a hobby of diving into the sewer of recent changes and the new page patrol, it is just freaking amazing that they don't simply say "screw it" and reject everything.
There are reasons why the Wikipedia social structure is the way that it is. Admittedly it could certainly be improved and there are some real asshats and jerks that make putting edits onto Wikipedia sort of a pain in the behind. None the less, it is possible for anybody to edit Wikipedia and to get the edits to stick as generally permanent content. It just takes more than a few seconds to get that to happen.
Re:3 domains of verifiability (Score:5, Interesting)
> Having policies in place that forbid official statements just begs for sock puppet tactics
And if such a thing existed, you might have an argument worth considering.
But as no such thing exists, and anyone can insert any statement from any source *as long as it is independent* then I don't see what you're caterwauling about.
Using the example in the previous post, if Toyota wants to refute claims of sudden acceleration in the Wiki, all they have to do is publish an article on the topic in a major source - say the Wall Street Journal or IEEE Spectrum. Such an event will quickly result in the information being included in the article, as well as also informing many people that wouldn't have seen it otherwise.
A lot of these complaints boil down to "well I don't think you should have to have a 3rd party reference". That's just laziness. If the topic in question isn't interesting enough to warrant such publication, then there's probably nothing worth complaining about in the first place. YMMV, because if it's about *you* then it's hyper-focused. But just because you find something important doesn't mean anyone else will, and that's the *whole point* of the wiki's policy,