A trustwiki world

by Francois Joseph de Kermadec

On the Internet, there is no such thing as a point of reference, as they exist in the offline world. Indeed, we have all come to consider certain books, magazines or newspapers as “serious” and pointing people towards a select group of publications is considered acceptable, even sometimes a sign of political and economical awareness. In France, for example, if you wish to recommend a “neutral” and “good” newspaper, you would direct people towards Le Monde while recommending any of that paper’s competitors would come across as a political message.

Truth is, there is no such thing as a “neutral” and “good” publication — and not even Le Monde can qualify for that status. Every single publication out there has its faults and a real expert in a field can almost always dismantle and prove wrong any article that would make it to a general publication. Since there is however a need for such reference points to exist, at least in the abstract, we have come to a consensus around them.

The Internet, being so culturally diverse and “fluid”, in that information circulates a lot more quickly than through traditional channels, has not yet allowed for the establishment of such references. No matter what you cite, someone can accuse you of being partial, politically biased or to engage in some shady propaganda practices. Why? For a slew of reasons ranging from the lack of a good time-tested reference point to the ease with which WHOIS databases can be polled to reveal links between sites and organizations one wouldn’t have linked otherwise.

Enter Wikipedia, the online encyclopedia editable and reviewable by everyone. Instinctively, I believe we all know this is not a guarantee of accuracy or seriousness — nor does it prevent them, of course. After all, the chances experts on nuclear fission are roaming on the web, willing to share their knowledge for free and to take the time to post on Wikipedia are few. And even if these good people were to post there, should their opinion on Keynes be trusted? When it comes to Wikipedia, what matters is its editable quality: since there is no single editor and since we have no control over what the site will display when our readers will go consult it, we can always claim “it has changed” (most people aren’t familiar with Wikipedia’s revision histories) or “it’s not my opinion”.

By mixing everybody’s voice, Wikipedia does provide something no other encyclopedia can provide: anonymity and distance. Recommending Wikipedia is not recommending one text, it is recommending a range of possibles, opening a door to some more information but denying any stance on it.

Is Wikipedia good? So far, it has served me well and I wish the project all the success it deserves. But even if it were to become the worst, most inaccurate encyclopedia around, it would still have its place on the web.


2006-01-03 03:27:32
In the Wikimedia foundation there is currently a big push to include the sources for the information that can be read in an article. This will improve quality on the one hand and it will probably make the articles more difficult for some of the Wikipedia public on the other hand.

Quality will also be improved when we can include relational data in Mediawiki. When these databases are localized as well, they can be used in the many Wikipedias. The result will be that this data needs to be maintained only once.

Wikidata will be part of the Mediawiki software. The first implementation can be found this (http://epov.org/wd-gemet/index.php/Main_Page) This is a read only proof of concept implementation of a project called Ultimate Wiktionary.


2006-01-03 09:11:17
What Place Is That?
>>...if it were to become the worst, most inaccurate encyclopedia around, it would still have its place on the web.

I'm not sure what place that might be? The place students go if they want to receive a failing grade on a paper? The place researchers go if they want to lose the contract?

Most defenses of Wikipedia that I've seen confuse technology with human behavior. Le Monde is no more or no less neutrral when read online versus read on paper. It's how the news is gathered and produced that counts, not the technology used to distribute it. Wikipedia's use of the Interent as a distribution vehicle is essentially irrelevant. The words would be the same if they were published on tree bark. It's Wikipedia's work processes, its human behavior, that produces content that seems to be something less than credible.

I don't expect an anonymous mob of people to faithfully generate accurate content, if they're gathered on the streets of a city or at keyboards around the net. Defending fundamental faults in its way of doing business by pointing fingers at the imperfections in traditional content production is very much like an errrant child wailing that the neighbor's kid does the same thing.

2006-01-03 09:35:24
The nature of stuff.
The thing to remember (and I think most wikipedia visitors know or will learn) is that it is not a collection of indisputable academic facts - it is a collective interpretation of knowledge by our culture. It's a social experiment, not an academic one. Factual or not, it's a pretty good indicator of what people think is important, entertaining and relevant to their lives (especially the mindbogglingly trivial stuff). I've long believed that if a thing can't be expressed using nothing but math, it's just an opinion. Wikipedia shows that human 'knowledge' is defined by opinion, and it remains fluid, despite our best efforts to carve it in stone. It is often misshaped, deformed and even obliterated by social phenomenon. Don't worship the mountain of data, climb it and get your bearings once you're standing on top of it. I'll take perspective over 'knowledge' any day of the week. It's more useful. Especially when you find that you're standing on the wrong mountain.
2006-01-03 10:19:06
A fellow of mine once produced this summary of Wikipedia:

What do you get when you stir a spoon of shit in a gallon of ice-cream? A gallon of shit.

2006-01-03 16:55:29
This mountain
Either you know a mountain and its surroundings OR you have to climb the mountain to know.

When Wikipedia is properly reviewed, like recently by Nature, it does suprisingly well. Given that it is clearly stated that Wikipedia should, like any other resource, not be the only resource that you rely on. I think you probably lost the way to this top.