Wiki knowledge: can you trust the stuff?
posted by Nathan Zeldes on December 04, 2006
Some years ago when my kids were teenagers, they would find stuff they wanted on the ‘net, then get me to order it (After all, I had the plastic). I remember my son’s succinct way of putting it: “I need Mega Street Warrior III for Nintendo, please do your checking thing and order it”. My “checking thing” referred to a rapid but thorough process he’d seen me use whenever I gave my credit card online: checking the seller’s web site for consistency and quality; finding the people behind the site and running Alta Vista (this was pre-Google days) on their names in combination with trouble, problem, scam, etc; searching DejaNews (now Google Groups) for opinions in UseNet forums; uncovering the geographic location of the business and assessing the implications; etc, etc. By the time I was done, I’d have a pretty good idea whether the site offering Mega SW III was trustworthy or not.
What reminded me of this is the never ending discussion, also seen in the comments to this blog, about whether one can trust the content of Wikipedia, or any Wiki for that matter. The answer, to my mind, is a qualified yes – you can trust it all right, after doing the “checking thing” to determine how trustworthy any specific article is. I’ve done a good deal of research using Wikipedia, and I always cross-correlate it with other sources (making sure they didn’t copy from the same place as the Wikipedia author!) and subject its content to my own analysis. I never take content at face value… but by now I can derive a good idea of its worth in fairly short order.
The important thing, though, is to remember that the same thing applies to Real Life. My volumes of the Britannica appear very authoritative, but in addition to the many articles that are dated – nobody can change their hard copy when the world evolves – there are no doubt those that are simply in error. They may have been authored by experts, but experts can and do err too, often with dire results. Just read the newspaper! Con artists, too, existed long before the Internet. The principle of Caveat Emptor always applies, whether one is buying used cars or online knowledge; and arguably, the online world provides much better tools for validation of your seller – if you know how to go about it. Like in the real world, many poor souls lack this knowledge and pay the price.
The interesting question is not which encyclopedia is perfect – none is – but rather which one is more useful at the end of the day, Wikipedia or the Britannica? Personally I believe that for a naïve reader the Britannica would be safer; but for a web-savvy reader, one equipped to examine the information critically, Wikipedia can be far more beneficial, given its considerably greater scope and its superb facilities for self-correction. Doing research in this medium can be outright exhilarating… and - unlike the Britannica - if you find anything dated or erroneous, you can correct it then and there for the benefit of all humanity!
So what does this mean to a company? Obviously, any organization that wants to succeed in this day and age should strive to have its employees be of the kind that relishes Wikipedia, that has the courage and wherewithal to confront the online world and put it to use. Look around you – is this the kind of people you see?
Comments (8) (closed)
tagged: knowledge, wiki, wikipedia
Comments
Dec 04 | steven e. streight aka vaspers the grate said:
Very useful content here, and you do know about the technique of deliberate falsehood in encyclopedias as primitive DRM? They would put a mistake in it, so as to prove in a court of law that plagiarism occured when that was suspected. NYT had an article on this not too long ago.
It’s all about Trust Linking, but few understand the simplicity of it. You start at a site you are very sure of, a high credibility site, then check out its suggested links or blogroll.
Generally, you can’t drill down too deep, though. Funny. Even extremely reputable sites are linked to by bad sites and one form of cyber attack is for a, let’s say, porn site to link to your site/blog.
Then, when someone clicks on, say, a Technorati “Sites That Link to Mine” (or however it’s worded), users will see these horrible sites linking to you. Don’t laugh, it happened to me about a year ago.
There’s no way to get rid of them.
But I usually start with Wikipedia, then navigate to sites they list, or do more Googling.
Dec 05 | david ingram said:
Hi Nathan, I think what you are saying about wiki trust is very relevant to several discussions I have been having. I have posted some excerpts from your article on my blog (http://longstock.blogspot.com). I trust you are happy with that, but of course please let me know if you would prefer i remove it. Good stuff. Dave
Dec 05 | umbrel said:
I think one good fact about wikipedia’s mutability is precisely that it makes you aware that you must check the veracity of the article you are reading, many people forget with “fixed” encyclopedias, that an encyclopedia is not a reference as source article, but like the word says, just a compilation of articles, a second hand reference. The other good thing is that wiki puts the reference for the articles, so it’s easier to start checking the source data.
Dec 07 | Thomas Conrad said:
Very pertinent content. As someone who’s been dabbling with getting a subject specific technical wiki up and rolling, I can say that I’ve been asked that exact question very pointedly on repeated occasions. I would even broaden the scope of questioning the source of your knowledge from the hard print mediums (i.e. supposedly already verified and “trusted”) to any knowledge source—digital, print, visual, personal conversation. In my experience, a healthy, vibrant wiki (Wikipedia, et al) has been about as trustworthy as any other source of information.
Dec 11 | Jon Whipple said:
Hi Nathan,
I read this with interest as working in a library exposes one to many challenges including those that the librarians face professionally.
Caveat Emptor does indeed apply to any information harvesting. And it adds time. A great deal of time in many cases; and time, of course, is in short supply. It’s omething many of us are willing to sacrifice in the face of a deadline. We choose instead to rely on the source’s reputation, fallible as the source may be.
Collaborative efforts like wikis are an interesting case because the content in the source is so malleable. Again we are confronted with the choice of time and expense in cross checking or relying on the reputation of the source.
When we are handling any large amount of information in the course of our work or research, I suspect we’re all going to be choosing to rely on reputation. Especially in the face of tight deadlines. As critical timelines become more compressed, and comprehensive and detailed data is sought, it makes practical an economic sense that this is the case.
The newspapers realized it a couple of hundred years ago, and stationed reporters in the theatre of action for accuracy and increased reader trust.
Wikis (and indeed all malleable electronic publications) can do the same by adopting policies and procedures that allow a reader to determine the frequency of update and have some idea about the individuals involved in content creation.
Choosing the assistance of a librarian or information scientist can address both the time factor and the caveat emptor/corroboration factor by allowing you to focus on other work and having a skilled professional conduct reliability and reputation screening on your behalf in a specified period of time.
Doing your “checking thing” is good practice, and even econiomical from a defensive investment point of view, but I wonder how practical it is all the time?
Jan 02 | Kim Mason said:
Great article about source checking. I generally agree that verifiable articles (such as science articles) in Wikipedia are usually pretty good.
However, articles about new or contentious items can suffer from the ‘self-perpetuating inaccuracy’ or ‘self-perpetuating bias’ problem. Once an inaccuracy or bias becomes established in a number of internet sources, it can be difficult to fix an inaccurate Wikipedia article.
If one fixes such an article, and then ‘Wikipedians’ (as they like to call themselves) do a web search, declare you wrong, and revert it, what are you to do? Argue with them in the discussion page? They’ll simply cite a larger number of poor quality incorrect sources, which will trump your small number of high quality correct sources.
When I combine the idea of propagation of inaccuracy with my personal belief that some ‘Wikipedians’ (as they like to call themselves) are self-righteous nerds with too much time on their hands and an axe to grind, it causes me to completely distrust any article more contentious than Pythagoras’ theorem.
People also often miss THE major motivator for contributors to conventional encyclopedias to get it right, that being their reputation. They are generally respected in their field. Writing an inaccurate encyclopedia article isn’t going to look good to peers, and will certainly become known. Through anonymity, Wikipedia allows people to write anything that they want, without the chance of building a reputation, or fear of ruining their reputation.
Jan 04 | Micheil said:
Well said, Kim. Anonymity, motive, expertise — these are all things that make Wikis questionable. I much more trust a publication by an expert than I do content that could be edited by anyone. If an article in the journal Nature contradicts an entry in a Wiki, which is more likely to be accurate? Now, such a event would cause both to be examined I’m sure. Caveat Emptor. However, it should be no surprise that your gut tells you the Wiki is likely wrong.
Jan 05 | Nathan Zeldes said:
Kim has hit the point when he distinguishes “verifiable articles” from the others. However, I would argue that any article needs to be verifiable, and those in print media are no exception. The notion that the sun goes around a stationary earth was fully supported by the corpus of scientific literature in the middle ages.
The “caveat” always applies; therefore I consider all sources “questionable” - not in a negative sense but as a statement of fact: you gotta question them and decide for yourself what is right. Once you realize that, the more sources you can cross-reference, the better - Wikis included.