Is Wikipedia the most comprehensive collection of knowledge in human history?

[First_Amongst_Daves]

Your nitpick is well-founded. My apologies - turns out Nature are arguing the point:

Battle of Britannica

AT FIRST, it was an innocuous test of accuracy. Last year, journalists at Nature wondered how scientific entries in Wikipedia—a free web encyclopedia that anyone can edit—compared with those in “Encyclopædia Britannica”. They compiled a list of subjects, downloaded relevant entries from each website, and sent the results to experts.

The findings were published in December and Britannica won—a blow had been struck by the gold-standard encyclopedia compiled by experts over the collective knowledge of a bunch of hobbyists and amateurs. Except that the results held a surprise. Britannica contained a lot of errors, and it was only 30% more accurate than the free encyclopedia.

This was all too much for Britannica. So, five weeks ago, it launched a 30-strong team of rebuttal editors and experts to pick apart the Nature study. The team’s 20-page report concluded that “almost everything about the journal’s investigation was wrong and misleading”. Oh, and it demanded that Nature retract its study. Nature hit back with its own rebuttal and refused to retract anything. On March 27th Britannica fired another salvo, with advertisements in the Times and the New York Times. Nature returned fire with a rebuttal editorial.

Some of Britannica’s criticisms are quibbles. It didn’t like Nature’s headline, and Ted Pappas, Britannica’s executive editor, says Nature didn’t allow Britannica to see any of the evidence on which the study was based until a week after it was published. That was a bit unsporting. In the media binge after the study’s release, Nature’s journalists talked widely about their findings, while Britannica’s editors felt they could not defend themselves.

Besides these quibbles, there are a couple of more serious issues. One is the overall accuracy of Britannica, the other is its relative accuracy compared with Wikipedia. On the first, Nature identified 123 errors in 42 Britannica articles. These comprise factual errors, misleading statements and critical omissions.

However, many of these “errors” are really the opinion of the reviewers. For example, a Nature reviewer says of the Britannica entry on the Cambrian period that the “evolution of hard parts at the beginning of the Cambrian involved much more than development of calcium carbonate.” Britannica replies: “the article does not claim or imply that calcium carbonate explains all there is to know about hard parts.” In total, Britannica contests 58 “errors”—all points on which a legitimate debate might be had. This is important. If these are not mistakes, then Britannica has 1.5 errors in each article, not three.

Nature’s response has been to argue (correctly) that the point of the study was to compare encyclopedias. Because reviewers were not told which article was from which encyclopedia, there would be no systematic bias one way or another. They would be just as likely to have differences of opinion with Wikipedia.

Nevertheless, there are some concerns about the study. For a start, one has to wonder how reviewers could fairly compare the misleading statements in Britannica’s largely well written and edited articles with those in Wikipedia’s sometimes poorly written and edited articles. Comments like, “this is a badly written article” were not counted.

Ignoring this unknown, there are two potential sources of bias. One is in errors of omission. Of Britannica’s 58 debatable errors, about 20 are omissions. But Wikipedia articles are longer than Britannica’s and have a tendency to accumulate material and facts—gathering all that is known on a subject. Britannica, by contrast, edits its subjects for style and readability. This means leaving things out for clarity. Any side-by-side comparison is therefore likely to notice more omissions in Britannica, because fewer facts have been included. That does not necessarily make Britannica less accurate. And if all the 20 omissions in Britannica were unfair, Wikipedia would actually be 50% less accurate not 30%.

The other concern is in the way the data were compiled. Nature’s journalists were not always able to find identical subject entries in both encyclopedias. When that happened, they did an online search on the subject at both websites and, if necessary, bolted together bits from several articles. (Britannica complains that in some cases they used paragraphs from a children’s encyclopedia.)

Yet this compilation was not being done blind. So when Nature’s journalists compiled the material, they could have introduced an unconscious but systematic bias. Also, because Wikipedia has 1m entries and Britannica has only 124,000, it is possible that reviewers were more often sent complete Wikipedia articles to compare with cobbled-together Britannica articles than the other way round.

It is difficult to avoid concluding that the Nature study was comparing apples with oranges. In any case, most people don’t need an expert to tell them that, while Britannica is readable and reliable, Wikipedia is a fantastically useful source of rough and ready information. And, on top of that, it’s free.

[scotandrsn]

@Lama Pacos

[It seems obvious to me that Wikipedia is an entity whereas the LoC is a collection.

You’re not making sense here. Did you read your own thread title? “Is Wikipedia the most comprehensive collection of knowledge in human history?”, I believe it says. And then you dismiss the LoC because it’s a collection?

From the Library of Congress website:

The Library of Congress is the nation’s oldest federal cultural institution and serves as the research arm of Congress. It is also the largest library in the world, with more than 130 million items on approximately 530 miles of bookshelves. The collections include more than 29 million books and other printed materials, 2.7 million recordings, 12 million photographs, 4.8 million maps, and 58 million manuscripts.

The Library’s mission is to make its resources available and useful to the Congress and the American people and to sustain and preserve a universal collection of knowledge and creativity for future generations.

From Wikipedia:

There are 48,000 active contributors working on over 3,800,000 articles in more than 100 languages. As of today, there are 1,254,184 articles in English; every day hundreds of thousands of visitors from around the world make tens of thousands of edits and create thousands of new articles to enhance the amount of knowledge held by the Wikipedia encyclopedia. Visitors do not need any special qualifications to contribute, and people of all ages help to write Wikipedia articles.

and from here

Wikipedia is an online free-content encyclopedia that anyone can edit. Wikipedia co-founder Jimmy Wales has described Wikipedia as “an effort to create and distribute a multilingual free encyclopedia of the highest possible quality to every single person on the planet in their own language.” Wikipedia exists to bring knowledge to everyone.

So the LoC is far larger, which gives it the better claim of being comprehensive (which I shall interpret to mean, “possessing the greatest amount possible of the most in-depth information possible on the widest variety of subjects possible”).

In addition, since an inordinate number of Wiki’s 1.2 million English-language articles deal with the last five to ten years’ worth of movies, TV shows, celebrities, pop music acts, comic books, and hentai porn, its bid for comprehensiveness falls a bit short, IMO.

While there are certainly distinctions between Wiki and LoC, I (and other posters) fail to see how any of these distinctions draw a clear line that says it is valid to consider Wiki as a comprehensive collection of human knowledge but not the LoC.

Don’t get me wrong, I go to Wikipedia first if I want one-stop-shopping for information on a recent phenomena, and their article on how a CD works (a favorite research topic of my students), for example, is one of the only correct explanations in casual reference sources on the web*, but to call it the most comprehensive collection of human knowledge, especially when you have to redefine the term, is, shall we say, something of a reach.

*For the record, HowStuffWorks is ambiguous about the precise correlation of pits and lands on the CD to binary ones and zeroes, and About.com simply gets it WRONG.

[jimmmy]

I think we can’t answer OP in GQ because we will never agree on terms. It really comes down to how comfortable you are accepting information on the internet. I for instance, do not trust wiki the way I do Britannica and I think it goes to the issue we are all aware of and struggle with : that is the “branding” of information in the information age is so important and an ongoing issue for everyone on earth with access to the internet.

I don’t trust wiki the way I do Britannica. But I would be willing to bet that 20 random wiki articles on 20 subjects will tend to contain more and deeper information than I would ever find on the subject in Britannica. The question really becomes how much of the wiki information contains some crazy error. THAT is why I would trust Britannica more - but I would almost always peek at wiki on a big research project as well.

So to the OP I don’t think we can ever agree on what constitutes a “comprehensive collection of knowledge” in GQ terms… because society is still struggling with the same issue.

TMI
It is a similar conundrum to this when a Blog or obvious crony website talks about (insert political event/person) we are wise to take it with a higher grain of salt than something that appeared in the Washington Post or Wall Street Journal where presumably they have corporate journalistic guidelines they try to adhere to, as well as will retract and correct information and fire people when those guidelines are violated. That happens every day. However, sometimes the bloggers get it right, so to ignore all information until it appears in a branded format - say until then it is not “true”… leaves us at a real news disadvantage I think.

[scotandrsn]

@jimmmy

I think we can’t answer OP in GQ because we will never agree on terms. It really comes down to how comfortable you are accepting information on the internet. I for instance, do not trust wiki the way I do Britannica and I think it goes to the issue we are all aware of and struggle with : that is the “branding” of information in the information age is so important and an ongoing issue for everyone on earth with access to the internet.

I don’t trust wiki the way I do Britannica. But I would be willing to bet that 20 random wiki articles on 20 subjects will tend to contain more and deeper information than I would ever find on the subject in Britannica. The question really becomes how much of the wiki information contains some crazy error. THAT is why I would trust Britannica more - but I would almost always peek at wiki on a big research project as well.

So to the OP I don’t think we can ever agree on what constitutes a “comprehensive collection of knowledge” in GQ terms… because society is still struggling with the same issue.

TMI
It is a similar conundrum to this when a Blog or obvious crony website talks about (insert political event/person) we are wise to take it with a higher grain of salt than something that appeared in the Washington Post or Wall Street Journal where presumably they have corporate journalistic guidelines they try to adhere to, as well as will retract and correct information and fire people when those guidelines are violated. That happens every day. However, sometimes the bloggers get it right, so to ignore all information until it appears in a branded format - say until then it is not “true”… leaves us at a real news disadvantage I think.

The question is not simply one of “branding”, but one of credibility.

As I teach my students, credibility is a separate issue from truth, because when you are researching something, you often don’t know what the truth is. The credibility of Britannica is backed up, not by their popularity or reputation, but by the fact that they publish the names and credentials of the people providing and vetting the information.

Even though the Nature article finds what they deem to be similar error rates for the two sources, Britannica is more credible because they can say “the articles in this area of knowledge was produced by these people, who have the following credentials and experience in that field…”

Wikipedia can make no such claim at this time. It relies on random anonymous people to write it, and random anonymous people to edit it. It provides nearly no verified information on who is doing this work or their credentials.

In the absence of your own knowledge of the truth about a topic you had to produce material about yourself, if Britannica and Wikipedia differed on a particular point, and you had to place your own credibility at the mercy of one or the other, you’d have to go with Britannica every time.