« Touchy, those Wikipediasts | Main | We're mad as hell »

A Wikipedia challenge

I should have realized the danger of stepping into the Wikipedia morass, and the comments on today’s earlier post further indicate my folly in doing so. You know, The New York Times gets things wrong, too. As an argument on a sophisticated level, it’s that all texts are constructs reflecting the attitude of the constructor rather than a verifiable external reality; on a less sophisticated level, it’s that all the other kids are smoking pot, too.

I’ve had enough. I’m bringing it down to this challenge.

In a previous post, Grant Barrett, a professional lexicographer and the kind of writer whose expertise should be welcome on Wikipedia, explained why he has given up on it:

Within my experience, every—all, every one, in toto, all inclusive, the whole shebang—Wikipedia article I have checked has had errors in it.

Many of the corrections I have made to Wikipedia in areas in which I have expertise were later erased or effaced, usually by the insertion of provably false information or nuttiness by some self-serving nutjob who doesn't know a dictionary from a dingo. The entry on "slang" comes to mind.

Why should I waste my time in correcting something that I'll just have to correct again? Like John, I don't have the luxury of being able to camp out and defend against ignorance, unlike my colleagues who keep the entry for "jazz (word)" in good order.

In all the justifications and expostulations about Wikipedia sent to this blog to date, no one has ever responded directly to Mr. Barrett’s criticism. So here is the challenge: Give me a response to this criticism that makes a persuasive case that Wikipedia is valuable despite these disabilities, and I will publicly repent of my criticism.

Otherwise, I will consider that Wikipediasts’ have failed to meet the challenge, that the criticisms posted in this blog are valid, and that further discussion is sterile. We will more on to more rewarding subjects.

I will approve all comments that are not outright abusive but will not respond to further comments myself.



Posted by John McIntyre at 2:19 PM | | Comments (39)


This is almost too easy. Wikipedia is chock full of readily available information about all sorts of things, particularly aspects of popular culture. So, for instance, if I miss a detail in a movie I watch while exercising, Wikipedia is the first place I look to fill the gap. Sometimes it's better than conventional media on important things, too. Wikipedia's article on CPSIA may have errors (I haven't checked), but it's more informative than the New York Times, which ignores CPSIA in its news pages and lied about it in its one editorial on the subject. At least you can learn from Wikipedia that there's a controversy.

Anybody who relies on information found in Wikipedia, or in the Times, is a fool; people who do that ought to have guardians. But there are many times when you want either basic information about trivia or leads to real data, and Wikipedia provides both in a handy and inexpensive way. Newspapers and conventional encyclopedias don't. In cases like these, the cost of error is slight. Throughout the comments on the many Wikipedia posts, people have given examples of ways in which they find Wikipedia useful. One of mine involved Wikipedia's list of movies by year, which led me to a DVD of a movie I'd been looking for for years. No other source that I know of could have done that. Nobody is claiming that Wikipedia should replace newspapers and other traditional sources of information. So what is the argument for saying, in effect, that newspapers and other traditional sources should replace Wikipedia? The fact that some people find it useful in some ways justifies its existence.

We don't have to pick the one best source of data. Having many sources available makes us better off than we used to be. Sure, some people are going to misuse Wikipedia. There are also people who make decisions based on their horoscopes (which they find in the papers) or who decide what stocks to buy by listening to someone on TV or reading financial-advice columnists in papers or magazines. If we outlaw everything that can lead people astray, there won't be much left. And newspapers, for all their supposed fact checking, would be one of the first things to go.

Finally, an internet journalist who is neither a half-wit nor a complete twit.

Made my day …


I might be able to shed light on this point. A few students from my undergraduate course in protein science and I analyzed the quality of the scientific articles on the English Wikipedia, and they were (to me) surprisingly good.

You'll find that most articles have an internal ranking on their Talk page (the tab labeled "discussion" at the top). The rankings range from "Stub" and "Start" (the lowest levels) to "Good Article", "A-level" and "Featured Article" (the top levels). In a survey of 800 scientific articles, we found that these top three levels are reliably good. Indeed, some of the reviewers (experts in the field of the reviewed article, and tenured professors all) noted that the Wikipedia article was the best general reference available anywhere on the Internet, as far as they knew.

You can find the list of Featured Articles at

For illustration, you might enjoy the articles on "Niobium" (chemistry), "Enzyme" (biochemistry), "Immune system" (physiology), "Equipartition theorem" (physics), or the "Problem of Apollonius" (mathematics). There are of course hundreds of such articles to choose from. Admittedly, such high-quality articles represent a relatively small fraction of the articles on Wikipedia. But they exist and can be identified reliably by non-experts.

Such technical articles might be beyond your powers to evaluate. But if you doubt our assessments, I suggest that you hire another professional scientist to review them. There are also many Featured Articles in the humanities that you might be able to review yourself.

With all due respect,

Prof. Bill Wedemeyer
Michigan State University

First, I don't think either of your paraphrases applies to my comparison with The New York Times. I'm certainly not questioning the reality of reality, nor am I saying the "contamination" is okay or something to be waved off.

My meaning was actually positive in nature: *both* the NYT and Wikipedia are very useful resources, and their failings are manageable, if employed correctly.

Wikipedia is useful - I'd say irreplaceable - as a *starting point* for serious research on a topic. It costs nothing, is readily accessible, and provides lots of information on a breathtaking array of topics. You need to be more wary with information in Wikipedia than you do with Brittanica or some other sources - I'm not denying that! - but that does not negate its usefulness if you view it as just a starting point.

In other words, you're absolutely correct to instruct people to verify what they read on Wikipedia. But that doesn't negate its value, because there are doubtless many useful things they've included in articles that they wouldn't have uncovered if they hadn't consulted that Wikipedia article.

Finally, I'd like to say that, while I admit that it's not hard to find errors on Wikipedia, my experience with Wikipedia is nothing like Mr. Barrett's. I've read many articles (on subjects I know something about) that are error-free.

I think John's problem is that he wants his information newspaper-style, that is, quotable - "Give me something I can print." Millions of the rest of us don't need or particularly want that. In other words, our information source does not have to be perfect, just usable - by us, not necessarily by newspaperfolk, they can look out for themselves. It helps - helps us non-press types, again - if our information sourice is cheap (no, make that 'free'). And available. And comprehensive. And multilingual (this is vital for me in one of my professional facets. I quite understand if it doesn't matters a darnn to John, what I really don't get is why it is so difficult for him to accept that the lack of reproducibility which offends him so much does not matter a fig to many of us, most of the time, at least). Plus, Wikipedia is often surprising and/or amusing and is likely to lead us into journeys of exploration of byways of knowledge that the press just doesn't manage to do. Et, as they say, cetera. Microsoft has given up the battle and stopped publishing its dreadful Encarta, isn't that something like a hint?

Though I'm reluctant to offer any form of argument regarding Wikipedia's accuracy, I feel it is a necessary step in transforming who controls information and its outlets. As a person who studies Russian literature, many of the minor figures in the canon fail to make it into mainstream media outlets. I often take part in correcting information using the sources I have at hand, and complain to anyone in earshot. However, I am willing to do so in order to keep information about writers like Daniil Kharms and Sasha Chorny in the western media outlets.

Wikipedia is not yet a credible source of information, but it is the start of a necessary movement for public information.

With that said, your blog is always enjoyable reading.


I'd just like to add that, like Thad, I find this blog to be valuable and enjoyable (even irreplaceable, like Wikipedia!). I think John's treatment of Wikipedia has been a bit heavy-handed, but that doesn't detract from the wonderful things he does with this blog.

Acadamics really get upset by wikipedia, but if it's used right, it's a great tool. If you come across something that you have no idea what it is, wikipedia will give you a general-ballpark understanding. That's it. People shouldn't go there for specific word-by-word understanding of a topic.

Wikipedia might be rife with mistakes and hidden agendas, and it might be compiled by "nutjobs," but it is the only place on the Web where you can be assured of finding "something." You must agree that the random topical Web page is no more reliable, on its face.

I use Wikipedia when I come across a topic I don't know much about. But I use it only as a starting point. A minute or two looking over a Wikipedia entry can usefully inform your Google search for something better.

Permit me to repeat what I've said before and elsewhere, as I don't think I can sum it up more concisely than this:

Wikipedia has shown us that a mass medium can be rendered so plastic and so well-leveraged that any part of it can be manipulated by a relatively small number of people, in ways that defy a free society's usual means to guard against it, so long as the special interests in question have a moderate amount of resources and the will to do so. If there are portions of the content that remain untouched, it is for two reasons only: (1) no one has conceived a stake in them yet, (2) virgin forest makes for good cover.


It's said above that Wikipedia is the only place on the Web where you can be assured of finding "something". I would like to read the Wikipedia article about Carolyn Doran -- where is it? I would like to read the Wikipedia article about the identity of the infamous Jayjg whom I've read about in news sources like the Washington Post and The Register -- but Wikipedia decidedly lacks ANYTHING about the identity of Jayjg. I would like to read about Lori Haigh -- but where is her Wikipedia article?

These three examples above, I can find MUCH more reliable sources elsewhere on the Internet. Wikipedia, practically nothing.

Mr Barrett doesn't provide us with any links to back is assertions or give us the name of the account he used so it's a bit hard to tell what really went on. Still there are a number of problems experts have that he may have run into.

The first is that experts quite reasonably take the view that their opinions are correct. They wouldn't hold them otherwise. The problem is that Wikipedia is more interested in documenting the opinions of the entire field rather than trying to work out which expert is correct. This can result in conflict between the expert who wants to write about their pet theory and Wikipedians who want to write about the views of the field as a whole (this is covered in ).

The second is that experts are used to writing in situations where you are meant to be arguing a position. Wikipedia rather rejects this. A slight exception is textbooks but even then it is common to have an underlying argument that the subject should be of interest to the reader. Wikipedia on the other hand is less about arguing positions but presenting them and documenting both their supporter's and detractor’s evidence and arguments.

A third problem is that experts normally work in environments where original research is actively encouraged. Wikipedia on the other hand takes the view that such research is best published in more appropriate places (peer reviewed journals ideally). This can lead to conflict when the expert wants to write something that is to them obvious but has never actually been previously published.

A related fourth is that what may be obvious to the expert wikipedia may still need a citation for. It’s obvious that the Biber submarine in the Imperial War Museum is Biber No.90 however wikipedia will still want (and in this case has) a citation to that effect.

Fith is the need to work with others. Experts sometimes take the view that they are experts thus don’t need to talk to the average editor who probably couldn’t have much to add in any case. The problem with this is that not only do Wikipedians not really have an effective way of knowing that you are an expert but that even experts need to discuss things with others. At the very least they are likely going to need to discuss stylistic and article scope issues. It’s a very experienced Wikipedia editor indeed who can write a high quality article without input from other people.

Sixth is copyright of all things. Experts sometimes use material they have previously published and since Wikipedians don’t know that the expert who previously published it and the person now editing Wikipedia are the same such material can end up being deleted causing conflict.

Now this isn’t to say that experts can’t edit Wikipedia. they can. It’s just important that they understand that Wikipedia is not their usual stamping ground and a lot of the rules they are used to playing by don’t really apply.

Though I had intended not to comment further myself, I feel obliged to thank Thad and Mike Klein for their kind words about the blog, and John Ross for taking the trouble to try to understand my point of view.

As a copy editor, I hold a responsibility for maintaining the credibility and integrity of a publication and thus need reliable references to consult. When errors of fact get past the copy desk, or the paper is victimized by plagiairsm or fraud, it is a blow to my professionalism and my honor. (I realize that honor sounds increasingly archaic.)

That it is possible, and even likely, that just about anyone can go into an accurate Wikipedia entry and corrupt it with error, error that may or may not be detected and if detected may not be fixed for an extended time, outrages my sense of order and rightness. And that people should excuse that by saying, in effect, that everything is unreliable, stuns me.

I was hopeful in reading the first paragraph from "gallium," that this issue would finally be addressed. But the subsequent paragrpahs are so vague and general about who does what on Wikipedia that I don't consider the challenge met.

Additionally, "gallium" did not supply a real name or an e-mail address, making it difficult to inquire further. One difference between Wikipedia and this blog is that you know who I am, where I am, what I do and what credentials I can offer in support of my views.

Hard data on the probability and persistence of errant editing at Wikipedia can be found in this study that focused on the biographical articles of U.S. Senators:

Wikipedia Vandalism Study

Another case study of error within Wikipedia is found in a blog post which I authored here:

See how an important "fact" (that wasn't a fact) about Abraham Lincoln persisted in Wikipedia for over 600 days. More importantly, come to understand how such nuggets of misinformation are insulated from repair by the use of "assuring text".

Dear Mr. McIntyre,

I'm surprised that you say your challenge has not been met. Perhaps you should clarify for your readers more precisely what is required to meet your challenge.

I took your challenge as asking your readers to show that Wikipedia had easily identifiable, high-quality articles. In response, I provided you with a list of roughly 10,000 articles (Featured, A-level and Good Articles), of which a representative fraction had been vetted by tenured faculty experts at top US research universities. Therefore, Mr. Barrett's hypothesis that all of Wikipedia ("the whole shebang") is riddled with errors is false. Is the challenge met?

As a professional scientist, I sympathize with your incredulity that Wikipedia can be good, and with your outrage that its entries can be changed by the malicious and ignorant. I do not dispute that many Wikipedia entries are poorly written and may well be inaccurate. But our study clearly shows that Wikipedia can produce high-quality scientific articles, among the best available.

You needn't take our word for it. If you are sincerely interested in the truth of the question of Wikipedia's quality, you can carry out a similar analysis of these three categories of articles. I challenge you, on your honor as a journalist, to do so and to report on your findings.

Prof. Bill Wedemeyer
Michigan State University

My personal response would be, if you don't care about wikipedia, shut up and go do something that is useful.

You don't HAVE to be part of Wikipedia. There are problems everywhere in the world. There is hunger in Africa and even in many houses/tents in the US. All kinds of topics and feuds that have existed for centuries. Complain about that, and prepare to expect a similar amount of controversy, just with less of the "horrids" of the interwebz around it.

In my reading of Mr. Barrett's criticism I took the nub or the rub of it to be contained in the following question:

"Why should I waste my time in correcting something that I'll just have to correct again?"

That query speaks not merely to the transient state of any article but to the systematic process that maintains them all. And that process is radically flawed — in ways that Wikipedysiasts steadfastly refuse, as a matter of cult-like dogma, to fix.

So those of us who have been observing the dynamics of Wikipedia for many years now are not impressed by the ephemera of quality that it exhibits here and there, now and again, but we ask — "What safeguards are there against the day when this or that pseudonymous interest group conceives an interest in deceiving the public about its target domain?"

That is the real challenge to a well-informed public that Wikipedism presents.

Despite Derek-Jan Hartman's kind invitation to me to shut my mouth (which is no doubt echoed in other quarters than Wikipedia), I would love to be able to make use of Wikipedia in my work. But since I would have to confirm every datum myself, I can't spare the time. If Wikipedia took adequate measures to verify entries and prevent tampering, you would find no happier participants than copy editors.

Much as it pains me to disappoint a faculty member from my alma mater, I can't share Professor Wedemeyer's confidence. He assures me that 10,000 technical entries have been vetted. But that is out of more than 2.8 million Wikipedia entries in English. And while I believe it is generally conceded that the scientific and technical articles are the more reliable ones on Wikipedia, how can Professor Wedemeyer be assured that errors and sabotage have not been introduced into some of those 10,000 articles since their vetting?

Mr. Barrett's challenge stands.

One of my favorite experimental edits to Wikipedia was on a scientific article that was observed by at least 150 people that day ( ), because the article had just been mentioned in the scientific trade press as the first of a planned program to require Wikipedia publication before peer-reviewed journal publication.

The article was about an obscure nematode protein, and I deliberately messed with its content, in much the same way a Fox News watching conservative might mess with the article about Al Gore.

For more than a day and a half, nobody detected the deliberate scientific fiction, because I used big words like "introns" and "copurified". The breaching experiment was then uncloaked after I revealed my identity elsewhere online, the account was put through a "CheckUser" test, and the account was blocked.

I must agree with Jon Awbrey, that the only reason scientific articles are able to get so "good" is that they simply haven't been found (yet) to be a desirable target for mischief and deceit.

As an academic who's examined Wikipedia along with other commons-based peer products, I can offer a decent example, though it departs from Wikipedia per se:

If you've ever used the Internet, you've relied on the product of commons-based peer production. It exists in software called Apache, MySQL, Linux and Mozilla. These incredibly complex systems are produced in an open source model that fundamentally resembles the knowledge presentation of Wikipedia (Running on MediaWiki). These pieces of software all have bugs, too, but so does your proprietary software, and often times the open betas are rife with them (Just like the open betas that exist on Wikipedia, the stub and start-ranked articles). You can criticize Wikipedia for having errors, but those errors seem to be as common as those found in traditional encyclopedias (You may have heard of the study done by Nature, though it focused of science articles--we have to remember that there's more living disagreement with the humanities subjects and our desire to have a secure epistemology is really not a problem of Wikipedia but of life).

In general, the system of commons-based peer production has proven successful, and Wikipedia isn't so special in its product or its design, but only in that the information it produces is readily accessible to the lay public, and so these errors seem more egregious than the occasional problem with Microsoft Word or Windows. In many cases, knowledge presented by Wikipedia isn't "provably wrong", it simply goes against the literary, historical or lexicographical theory of the academic reading it. This happens in the peer review environment all the time, and we don't decide that the academy must, therefore, be discarded, any more than we would discard Apache, Linux or MySQL.

Prof. Bill Wedemeyer, I think the claim you are making needs to be less general. Give us a list of specific articles that you have found to be without error.

There's no way McIntyre can check all 10,000 articles you claim must be good on the basis of their rankings, even if he did have the expertise.

If you think The Baltimore Sun has the resources to hire a staff of scientists to check these articles...oh, wait. Wikipedia doesn't have that kind of money, either.

What Wikipedia has provided to a group of dedicated scientists is good. But it's not the same as a peer-reviewed journal, for the reason that the peers who are reviewing the articles on Wikipedia are sometimes malicious, often ignorant, and essentially untraceable.

John, while I completely share your frustration about Wikipedia and its problems and apostles, you are right not to carry on this conversation any further, because it is an apples and oranges conversation in which the argument comes from opposing poles.

McIntyre: "I hold a responsibility for maintaining the credibility and integrity of a publication and thus need reliable references to consult. When errors of fact get past the copy desk, or the paper is victimized by plagiarism or fraud, it is a blow to my professionalism and my honor. ..."

John Ross: "Our information source does not have to be perfect, just usable... It helps if our information sourice is cheap (no, make that 'free'). And available. And comprehensive. .. The lack of reproducibility which offends him so much does not matter a fig to many of us, most of the time, at least.... Plus, Wikipedia is often surprising and/or amusing and is likely to lead us into journeys of exploration of byways of knowledge "

In other words, we of the copy desk are looking for something to be absolutely correct, because that is what we do, and if a reference says, "Well, it may be correct, or it may not," it is useless to us. Fans of Wikipedia are looking for something to be, well, probably mostly correct, but also interesting, stimulating, and if it turns out to be partly made up, well, that's OK too -- it's part of the game, part of the fun.

To them it's not just a reference book like the World Almanac, used to state, this is when this happened, you can take it to the bank -- which is useful, but probably only copy editors grew up actually reading the World Almanac and saying, this is the final reference, if it says it in the Sun it's so. You can find that stuff, like who won the gold medal in 1908 in archery, in other places, like, oh, the World Almanac, so if you don't trust Wikipedia on it, it's OK.

And this creates a fallout problem for newspapers, which are in the business of saying, "Here is what happened..." -- to an audience that increasingly says, "Nothing happened, all is happening." Maybe some people believe that the Civil War did not begin with the shots at Fort Sumter. Who knows, maybe it didn't? Or maybe the fact that people believe it didn't is as important as the fact that it did? Or maybe it was actually caused by Lincoln's trying to hide a gay affair? And what about the books that could be published speculating on whether Lincoln's gay affair led to the sacking of McClellan? Does that not then become part of the historical record, because people have talked about it and published it even if it is totally wrong? You still might want to know the name of the nutcase who wrote the book, and what he said, and the fact that he was the great-great-grandson of Bulwer-Lytton, or maybe he wasn't but some people said he was.

So maybe that should all be on Wikipedia even though it's total nonsense? Once there is no truth, there is no truth, only myriad truths.

So if you believe that what should be presented to people as fact should be entirely factual, you just have to pursue that, and not argue the point with those who don't. Because you're not arguing the same point.

Several people seem to have misapprehended what I wrote. Let me clarify.

Mr. McIntyre posed the challenge: "make a persuasive case that Wikipedia is valuable." I believe I have done so, by pointing out a set of identifiable articles that appear to be reliably good for educational purposes. I cited the evidence of reviews of randomly selected articles by impartial experts. Our study also found that the quality of these articles did not degrade significantly over two years. I invited Mr. McIntyre to repeat our experiment himself, by randomly selecting Featured Articles and subjecting them to review by one or more experts. He needn't review 10,000, nor did I ask him to. Why not try a random set of 20-30 articles from disparate categories? That should be a fair test and within the means of the Baltimore Sun and Mr. McIntyre.

I cited five technical Featured Articles above. I have not read them all, but I am sufficiently confident in our analysis that examination of these articles by experts will bear out our conclusions. In addition to these five, here are 23 other articles, most of which I haven't read but which caught my eye while scanning the list:

Las Meninas (art)
Order of the Thistle (awards)
Action potential (physiology)
Evolution (biology)
Killer whale (taxonomy)
Panic of 1907 (business)
Rosetta@home (computing)
Emmeline Pankhurst (feminism)
Georgetown University (education)
Welding (technology)
Detroit (geography)
Hurricane Katrina (weather)
Influenza (medicine)
Mohandas Karamchand Gandhi (biography)
Turkish language (linguistics)
Edgar Allan Poe (literature)
Maggie Gyllenhall (theatre)
Music of Nigeria (music)
Franklin D. Roosevelt (politics)
James I of England (royalty)
History of America football (sports)
Panama Canal (transport)
William Tecumseh Sherman (warfare)

Since my students and I have studied only technical articles, we can't claim to have shown that non-technical Featured Articles are reliably good. I conjecture that they are, based on my experience. Thus, we will all learn something if you commission expert reviews of a few such articles.

A word about standards. Mr. McIntyre and others seem to require proof that Wikipedia articles are flawless and guaranteed to remain so. If so, then the challenge is impossible, not only for Wikipedia but for any educational work. In my nearly 30 years in academia, I don't think I have ever read a textbook, even the 7th edition of a math textbook, that was perfectly without error. As every copy editor knows, a few errors (typographic, stylistic, factual and of omission) are inevitable; no written work is a Platonic ideal of Truth. A more practical goal is to have infrequent errors, to have minor errors, and to build in redundancy so that a typical reader can correct the errors themselves. I agree wholeheartedly with Mr. McIntyre that we should strive for these goals, on our honors as scholars and journalists. Wikipedia could undoubtedly do a better job at error detection and correction, and I do not wish to defend its myriad faults. But it is also false and unjust to say that Wikipedia has nothing of educational value, based on the evidence I have presented. I ask Mr. McIntyre to concede that this is so, on his honor as a journalist.

Prof. Bill Wedemeyer
Michigan State University

Well, let's throw at least one straw man on the bonfire. I did not ever say or suggest that Wikipedia articles should be flawless and guaranteed to remain so and would have been a prize fool if I had. The point I have repeatedly made is that I could have confidence in Wikipedia if it (a) had an adequate system for vetting entries, preferably before publication, and (b) a mechanism to prevent ignorant or malicious tampering with published entries.

Since it has neither, My checking a set of random articles would have only an emphemeral value, since I would have no assurance that the articles I looked at would be free of subsequently introduced errors. That was the point Mr. Barrett made, and that was the point of this challenge.

I though the challenge was *not* to disprove the very real perils you and Mr. Barrett have discussed, but to argue that Wikipedia was "valuable despite all these disabilities." And I think several people on this thread have made pretty good arguments to that effect.

Seems as though articles on Evolution or Edgar Allan Poe would be prime targets for tampering by certain parties.

Different readers appear to have read the challenge in different ways.

In my experience, Wikipedia is a social software system that manifests the qualities of an addictive game and even, in many cases, a narcotic drug, and so I understood the criterion to be whether Wikipedia, on balance, is both safe and effective, and whether its benefits outweigh its deleterious side-effects.

All things considered, I would have to answer with a resounding "No" — but it's possible that I know a lot more about the side-effects than the average reader here, so I may try to list a number of those when I get time for further remarks.

It is amusing to me that Professor Wedemeyer would enter 23 articles into this debate, without addressing the potential effects of the survey participants (us) on the subject (the articles). I could go and vandalize every one of those 23 articles now.

If a proper "study" is to be conducted on those 23 articles, I suggest that the "research team" chose a date and time sufficiently in the past (for example, April 4, 2009, at 00:01 UTC), and examine the article AS IT APPEARED at that particular time. That way, we can judge the articles without fear that this contest has adversely impacted them.

(This is why my team of researchers selected the fourth quarter of 2007 to evaluate the 100 articles about the U.S. senators, which proved to be wrong 6.8% of the time -- a study that the media seems to have ignored.)

Of the 23 articles that Professor Wedemeyer laid out for inspection, the one I probably know most about is "Detroit".

I went back to how the article appeared on April 4, as directed above, so as not to select a spoiled sample:

The very first line of the article states that the word "Detroit" is "French: Détroit, meaning 'strait'".

I am of the understanding that the French word for strait is "étroit", and that it would be far more accurate to say that D'étroit means "at the strait" or "of the strait".

If I am correct, then within the first article I happened to proof, within the FIRST LINE of the article, there is an error.

Oh, and by the way, when I clicked the ".OGG" audio file to hear the French pronunciation of D'étroit, it crashed my browser.

Thanks, Wikipedia.

Grant Barrett and John McIntyre were wise enough to phrase their challenge in a way that is resistant to the fallacies of good intentions and partially good results justifying any means whatever — but all pretenders to the prize so far have failed to see the strength of the "despite" clause that closes out all such paths of easy virtue.

In other words, we cannot escape a thorough analysis of risks and benefits before we prescribe Wikipediot ways to society in general.

The side-effects of Wikipedia use are many — there are effects on the individual, both passive user and active participant, and effects on the society that gets in the habit of using the Wikipedia brand of software system.

These days I worry more about the global side-effects of Wikipedia abuse, so let me break here by mentioning the 3 that concern me the most:

1. The erosion of a free press.

2. The mis-education of learners.

3. The manipulation of scholarship.

"Give me a response to this criticism that makes a persuasive case that Wikipedia is valuable despite these disabilities, and I will publicly repent of my criticism."

Several people have pointed out that Wikipedia gives users a quick introduction to a great many subjects and links to original sources, making it a good starting point for some kinds of research. If this has not shown that Wikipedia "is valuable despite these disabilities," why not? Mr. McIntyre keeps saying that the challenge has not been met because Wikipedia articles contain errors or may have errors added even if they are accurate at the moment. But no one disputes this; the challenge was to show that it's useful "despite these errors."

In the very first comment, Alan Gunn wrote:

"The fact that some people find it useful in some ways justifies its existence."

I honestly do not know how to reason with a person who has such a notion of justification.


"The fact that some people find it useful in some ways justifies its existence."

I honestly do not know how to reason with a person who has such a notion of justification.


Hmm. I, and others, say we find Wikipedia useful. You not only don't consider that a reason for thinking that Wikipedia is useful to us, you consider it unreasonable to even make this argument. Is that because you think we're lying? Or do you have some sort of notion that if you don't approve of the things we like, we shouldn't like them?

I have no very good idea of what a sextant does (though I suppose I could look it up in Wikipedia). But I can infer from the fact that they exist, that there are uses for them. Why is this puzzling? It's called "freedom of choice."

I'll just add that I gave a couple of examples of Wikipedia's usefulness to me as well. I guess those don't count either, if they aren't uses you would make of it.

I think a lot of the disagreement here is more or less political Some people think that "authorities" should decide what's good for everyone. I'd prefer to decide for myself, thank you.

Oh, I think we can do better in this discussion than set up false dichotomies between "authorities deciding for us" and "deciding for ourselves."

I have 11 books on usages within reach at my desk, and more on shelves nearby. They are written by authorities, but I don't blindly subscribe to any of them.

I do think, however, that specialized knowledge and expertise should be respected. I suspect, Mr.Gunn, that if you feel ill, you consult a doctor rather than five people chosen at random on the street. Is that "political"?

"I do think, however, that specialized knowledge and expertise should be respected. I suspect, Mr.Gunn, that if you feel ill, you consult a doctor rather than five people chosen at random on the street. Is that "political"?"

Well, yes, I do consult doctors, though some people see chiropractors and some deal with outright quacks. Then, after consulting a doctor, or several, I decide what course to take. Several years ago I decided to follow advice from USA Today rather than my doctor's advice in dealing with lower-back pain; it worked, and he has since come around to the USA Today position. Expert advice can be useful. There are political aspects to medicine, though, along the lines we see being drawn here. For instance, there's an economic study concluding that the requirement that many drugs be available only by prescription does not improve health in any measurable way; libertarians would therefore scrap the requirement, Doctors probably wouldn't like that change. When I was young, you couldn't buy reading glasses in a drug store, though this has changed in most states; I consider the change an improvement.

As Professor Pullum's review of Strunk & White pointed out, White was an amateur grammarian, and got many things wrong. Fowler, whose work I like a lot better than White's, was an amateur, too, and he was honest enough to refrain from calling his personal preferences "rules." I use books on usage in the same way you do. I know some people who take every bit of grammar advice they find in print as gospel. I think they're fools.

Copy editors, if I'm not mistaken, are people who are good at language, though many of them seem to have a strange objection to "split verbs." They are not always experts in science, law, economics, or any other substantive field. So why should I take their screening of the material that goes into their papers any more seriously than I take Wikipedia's screening by its own readers? Especially when I see, day after day, once-reputable papers like the New York Times refusing to cover important matters because the coverage would reflect poorly on politicians the Times supports. (Same for Fox News, at least in the evening.)

If having authorities decide things for us and deciding for ourselves is a false dichotomy, why do you complain so strongly about Wikipedia's lack of authoritative editors? In fact, I don't see the matter of experts and individual choice as a dichotomy at all; it's more like a spectrum. For some matters, like medical care or legal technicalities, expert advice is vital. For others, it doesn't matter much at all. I don't need experts to tell me what the plot line of "The Prestige" is, so I use Wikipedia for things like that. Wikipedia's list of movies that came out in 1952 may not be complete, but it's the only one I know of, and it has helped me. I don't use Wikipedia for medical advice. Different people will make different judgments about the weight to be given expert opinion on various matters, and if I weren't a Methodist I'd bet a lot of money that you and Mr. Awbry are a lot more inclined than I am to count expert judgment heavily. And this division figures in many political debates. For instance, the opposition to vouchers for private schools comes largely, though not entirely, from people who think parents are too dumb to choose schools for their children. The objection to "privatizing" part of Social Security comes largely from people who think most workers would make foolish investment decisions.

Several of us have said this before, but it seems to need repeating. Nobody is trying to convince you to use Wikipedia for your work. You have made a convincing case that it wouldn't help. That doesn't mean that it is worthless for other people, doing other kinds of things. Your reluctance to admit this, in the face of examples from many people of ways in which they find Wikipedia useful, seems odd.

The Mess Is Ended.

Go In Peace.

Oops! I meant that for the Nunc Dementis thread …


I'm with Mr. Awbrey; enough is enough. And I should have paid more attention to Mr. Sullivan. We are talking past each other.

It troubles me that I cannot rely on the accuracy of anything I see on Wikipedia. Mr. Gunn is not troubled by his inability to trust the accuracy of anything he sees there, and Professor Wedemeyer apparently doesn't perceive anything amiss at all.

Well, we are all entitled to our opinions, and this thread of comments is closed. I will not be aproving further comments on this subject.

Originally posted by Gregory Kohs:
" ... and I deliberately messed with its content ... "

There can be no more destructive mayhem done to your credibility ... than you have just committed yourself.


Post a comment

(If you haven't left a comment here before, you may need to be approved by the site owner before your comment will appear. Until then, it won't appear on the entry. Thanks for waiting.)

Verification (needed to reduce spam):

About John McIntyre
John McIntyre, mild-mannered editor for a great metropolitan newspaper, has fussed over writers’ work, to sporadic expressions of gratitude, for thirty years. He is The Sun’s night content production manager and former head of its copy desk. He also teaches editing at Loyola University Maryland. A former president of the American Copy Editors Society, a native of Kentucky, a graduate of Michigan State and Syracuse, and a moderate prescriptivist, he writes about language, journalism, and arbitrarily chosen topics. If you are inspired by a spirit of contradiction, comment on the posts or write to him at
Baltimore Sun Facebook page

Most Recent Comments
Sign up for FREE local news alerts
Get free Sun alerts sent to your mobile phone.*
Get free Baltimore Sun mobile alerts
Sign up for local news text alerts

Returning user? Update preferences.
Sign up for more Sun text alerts
*Standard message and data rates apply. Click here for Frequently Asked Questions.
Stay connected