What prevents malicious tampering of Wikipedia?

Mississippienne and I were talking one night about how much she loves Wikipedia. I’ve been there a few times myself, and am truly surprised by how open everything is. If you want to edit an article, go right ahead. Your edits will show immediately.

The English language failed me when I was trying to ask her what keeps Wikipedia protected from malicious attackers? Specifically:

[li]What would they do if someone were to delete or muck up all of their entries with attack messages, hate speech, or other nonsense? I’m pretty sure this could be done fairly easily with a program that can be written by anyone with a moderate knowledge of Visual Basic.[/li][li]What steps does Wikipedia take to prevent such attacks?[/li][li]How can the integrity of individual entries be guaranteed? All it takes is some snotty 12-year-old kid with an “OMGi’msoleet” attitude… how would Wikipedia know that an entry somewhere on their database has been fooled with?[/li][li]What steps are taken to correct an entry? What if every entry on the database were destroyed? [/li][li]Why hasn’t something like this happened yet?[/li][/ul]
I don’t know if I’m making much sense. Wikipedia has almost no defense (from what I can tell) from those who wish to play a practical joke or (even worse) take pleasure in destroying someone else’s hard work. In other words, what keeps Wikipedia from getting slammed?

Thanks for the help.


In a word, nothing. That’s the strength and weakness of Wikipedia. Anybody can edit articles, for better or worse.

However, it’s not quite that bad. Every article has an attached History page detailing each and every change ever made to the page from the time of its creation. In addition, all recent article edits all show up on a global edit history page that many dozens of users go through on a daily basis. They can see all the changes that were made to Wikipedia, and can then choose to Revert the article to a previous version if vandalism is detected.

There is always a danger of you viewing an article right after it was vandalized and before anybody else caught the action. That’s more of a theroetical worry than a practical one, but if you wanted to protect yourself, you should probably check the History page of any Wikipedia article you wanted to use.

Deliberate vandalism aside, Wikipedia tries to maintain a netural point of view in its articles. When two sides argue over things in article, either both sides get presented or the article itself gets temporarily locked and a debate ensues in the discussion pages, the results of which are then transferred back to the main page.

So far, Wikipedia has seemingly relied mostly on the ratio of good users to bad ones. Vandalism can be kept in check because there are more people who want to do Wikipedia good than those who want to vandalize it. But if, say, somebody wanted to make a distributed botnet that randomly vandalized Wikipedia articles on a massive scale… I don’t know what would happen.

And if you want to hear Wikipedia’s own take on this situation, you might want to check out this page:
Wikipedia: Replies to common objections

Reply posted an excellent summary, but I thought I’d address these specific parts.

As Reply pointed out, the complete history of each page is availabe. However, there is no guarantee of the factual accuracy of an article. Wikipedia does not have a formal fact-checking process. While the popular articles are edited by many people and generally approach excellence with respect to time, there are no guarantees. Therefore, don’t rely solely on Wikipedia if you’re researching anything important.

The database is freely downloadable and is mirrored in a number of places, so this is not a likely scenario.

There have been bots written to wreak havoc with Wikipedia articles. This is generally handled by banning IPs (or whole IP ranges) of malicious users. A distributed attack would be quite gruesome. But once a pattern of behaviour could be determined, writing a program to reverse the attack would be trivial.

Also check out this cool article in Wired about Wikipedia and how it works (focusing on its founder, actually)

A lot of Wikipedia editors spend time correcting sabotage. And there is an extensive grievance system run by volunteer editors. However, the system is not foolproof. Take a look at the entries relating to Lyndon LaRouche, for example, and you’ll see a history of a long-running mess. There are LaRouchie editors who do all they can to enter as much LaRouchie propaganda as possible in various entries and there are those who argue with them on every point. The result is an extremely long but ultimately uninformative article on LaRouche, because it fails to give an uninformed reader the basic information in plain English – that LaRouche is a fraud and is irrelevant to any aspect of public or scientific life.

      • Essentially, nothing.
  • Because of its open nature, Wikipedia is not considered by academics to be an authoritative resource. It’s got lots of participants and it’s fun because “everybody gets to be an expert” but with no accountability or required editorial fact-checking, as a reference it is just about worthless. The main problem is “subject Nazis” who wish to be seen as the absolute authority on a subject, and regularly check “their” entries and revert or re-write them if anyone else has changed them.



Part of the fun of screwing with regular websites is simply hacking into them. Look at me, I managed to hack into their servers! Ha! With Wikipedia, this is SUPPOSED to be easy, so no one’s impressed. Plus, it’ll only be up for a few minutes anyway, so why bother?

A few things that the Wired article pointed out:

Some people put entries on their “watch” list. If any changes are made, they get an alert and can check right away. This speeds the time that someone with a serious interest in the topic can change a prankster’s edits.

They have been known to freeze entries that were especially attractive targets for hijinks. For example, during the election, they locked entries on the major political candidates–they could be viewed but not changed.

Neither of these things are guarantees of accuracy.

As one of the articles pointed out, those with entries on a watch list are often self-appointed subject dictators over a subject and they do not allow changes to stand that contradict their own views.

A locked article more often than not stays locked complete with inaccuracies and the last stage of partisan posturing that existed before locking.

An even bigger problem with Wikipedia is that it is populated by a large contingent of fanboy types who record in the minutest detail “facts” gleaned from fictional works, such as Tolkien and Star Wars, but rigidly what they determine to be minutiae in factual articles.

How do other encyclopedias work? How do they verify the trustworthiness of their editors, and how do the editors themselves get their facts checked? It’s not like they can refer to an encyclopedia…

I’ve been amazed at the relative civility in Wikipedia myself. I would think political entries would be slanted if not attacked but they don’t seem to be.

Nothing. Next question?

Not trying to hijack your thread or anything :smiley: , But I just recently visited this gasoline price site which can be contributed to by guests. Seems someone has been submitting fake prices and just writing stupid comments to their buddies.

I read the Wired article but I am wondering what is this Wiki software? Is it freeware?

      • Ecyclopedia companies pay employees to do the fact-checking. If the employees are found to not be doing their jobs properly, the employees get fired and the articles in question get reviewed by someone else. If you heave BS on wikipedia you do not generally risk losing your source of income.

Gangster Octopus, hopefully, you will not ask them for protection money.

The software that runs Wikipedia is called MediaWiki, although there are many other ‘Wiki’ programs out there, pretty much for every web language imaginable.


Yeah, but how do they actually check the facts? Who do they go to? The average joe, when confronted with uncertainty, turns to a dictionary, encyclopedia, or other “authoritive” source. So who do the editors, in turn, go to?

Suppose an encylopedia says that Martin Luther King’s birthday is January 22, 1975 (it’s not really). Do the paid employees go to a records office and look it up, or what? What about other types of information? Like… for an article on a certain type of fish, do they consult marine biologists? What about politics? Certainly not the politicians involved, I hope. And for less important things, like software programs, TV shows, movies, etc.?

I guess what I’m asking about is not the motivation for fact-checking, but rather the process. Could Wikipedia, if it wanted to, employ a team of fact-checkers that go through the same process the big encyclopedia editors do?

On a semi-related note, Encarta seems to have adopted a Wikipedia-esque process that allows users to add content. The added content is then checked by a team of editors before being added to the encyclopedia. The best of both worlds, perhaps?

Vandalism isn’t a problem. It’s usually cleaned up within a few minutes, and anyone looking for facts will know right away that the page should not be trusted. The real problem is small errors that never get fixed. For example, Belarus’ population is given as 11,196,394. That’s nearly one million too high, and it has been that way for eight months. To find the edit, you need to review nearly 250 history pages. And even then, the average reader would not know to reject the population number, as the first conclusion one is likely to reach after finding the edit is that the new number was a factual correction to the old number.

Solution? If you’re just checking on something for your own interest, Wikipedia is great. But before you use something from the Wiki for any serious purpose, consult other sources.