Why Wikipedia sucks: Verklebte Nebenhöhlen (was übrigens ein guter Name für eine Deutsch-Punk-Band wäre, aber das gehört jetzt nicht hierher) halten mich seit gestern sowohl von der Arbeit als auch von auch nur halbwegs konstruktivem Denken... [more]

wikipedia sucks: the aardvark speaks:"wikipedia generates noise, not knowledge. previous encyclopedias were well-researched and contained precise information that could be trusted to be correct. wikipedia, on the other hand, contains a large amount of errors, omissi... [more]

Parallel noise: Horst Prillinger: "...Wikipedia generates noise, not knowledge. Previous encyclopedias were well-researched and contained precise information that could be trusted to be correct. Wikipedia, on the other hand, contains a large amount of errors, omissi... [more]

Why Wikipedia sucks. Big time.: Horst Prillinger: Wikipedia generates noise, not knowledge. Previous encyclopedias were well-researched and contained precise information that could be trusted to be correct. Wikipedia, on the other hand, contains a large amount of errors, omissions an... [more]

Why Wikipedia sucks. Big time.: The Aardvark Speaks: Why Wikipedia sucks. Big time.... [more]

Did Wikipedia fail?: Horst is guilty of preventing me from doing university stuff. He wrote a grumpy rant about the shortcomings of Wikipedia. Horst claims: Wikipedia generates noise, not knowledge. Previous encyclopedias were well-researched and contained precise informat... [more]

Scheffredakteur!? We don't need no stinky Scheffredakteur!: Kaum l ä uft ein selbstorganisiertes Projekt im Web einigerma ß en erfolgreich, schreien sie wieder nach einem Scheffredakteur. Horst Prillinger: Why Wikipedia sucks. Big time. Lieber Horst, der » normale « Nutzer der Wikipedia ist sehr wohl mindestewe... [more]

Why Wikipedia sucks for H. Prillinger: Kritik an der Wikipedia geht ungefähr so, glaub ich: Es gibt keinen Redakteur. Da steht doch dann Unsinn im Web wenn einer was falsches schreibt, bis es korrigiert wird. Weblogs hingegen: Es gibt... [more]

Wikipedias in der Kritik: Horst kritisiert Wikipedias, und erinnert mich stark an das “unverkrampfte” Verhältnis vieler Geisteswissenschaftler zum Netz. Ist laut netbib stark diskutiert. Passend der Kommentar von Moe Es gibt keinen Redakteur. Da steht doch dann U... [more]

Recent Changes not enough: [more]

I told you so, didn't I?: I'm well aware that people who say say "I told you so" after something has gone wrong are not exactly universally loved. Still, this time I simply can't resist becoming one of these people myself. Wikipedia administrator Ulrich Fuchs resigns because Th... [more]

Und Wikipedia funktioniert doch: Das sicherlich größte Projekt für Online-Kollaboration, Wikipedia, hat mich schon immer fasziniert. Ich selbst nutze sie viel. Immer ist da aber die nagende Unsicherheit "Stimmt das wirklich?". Wenn jeder mitschreiben kann, wie kann da gewährleiste... [more]

Wicipedia Cymraeg - pwysig iawn ar gyfer yr iaith: O'n i'n siarad â Joi Ito am Wicipedia - y gwyddoniadur arlein - pan penderfynodd o i fy nghyflwyno fi i'i ffrind di, Elian, sy'n gweithio ar <a hre... [more]

Wikipedia Sucks: From http://homepage.univie.ac.at/horst.prillinger/blog/archives/2004/06/000623.html: Wikipedia generates noise, not knowledge. Previous encyclopedias were well-researched and contained precise information that could be trusted to be correct. Wikipe... [more]

Wikipedia revisited: One of my students at the teacher training college had used this Wikipedia entry as the basis for a history lesson on imperialism that he taught in a practice class. His supervisor (not me, by the way) failed him, not for his teaching method, but for g... [more]

Why Wikipedia sucks. Big time.: Horst Prillinger: Wikipedia generates noise, not knowledge. Previous encyclopedias were well-researched and contained precise information that could be trusted to be correct. Wikipedia, on the other hand, contains a large amount of errors, omissions an... [more]

This coming from someone I have learned to know as a fierce defender of open access to knoledge, the state of Wikipedia must be really bad indeed. I must admit that I have *never* visited Wikipedia so far, and reading your blog I am rather unlikely to do so in the immediate future. In general, I fancy the idea of a public space where you can share your knowledge, and what is more, find interesting information. However, if Wikipedia works the way you say, it is really a place of opininion and not of knowledge. The basic principles of scientific work *must* apply for a publication which calls itself "encyclopedia".

Wikipedia is not opinionated. It's amateurish.

The good thing about it -- giving everybody who visits the opportunity to share their knowledge -- turned into a bad thing because there are just too many people out there who are feeling compelled to share their lack of knowledge.

Open Directory proved how an open structure could work in the right environment. Wikipedia proves how an open structure can go awry in the wrong environment.

ad 1 and 2: Why don't you just remove the incorrect and pointless stuff? Removing content is even easier than c&p and rewriting-content from a different Website?
Apart from the large number of bloat articles there are some really well researched articles that are a lot more informative than the one in my dead-wood encyclopedia. (e.g. http://de.wikipedia.org/wiki/Rattenk%f6nig )
ad 3: In the topics I am interested (CS and math basics) I found that the wikipedia articles on the top ranks of google gave a good overview over the topic. But of course, as a expierienced researcher, you should never trust a single document.
ad 4: If there is a copyright infringement, there is a well defined process to delete content permanently (not only overwrite it, but clearing the history). This is better than in the WWW, where it is difficult to shut down a website on the other side of the globe.
ad parallel WWW: I noticed that the linking inside Wikipedia is a lot better than in the WWW (even better than in Weblogs), so it is very easy to browse from one article to the next one. I think the reason is the well defined naming scheme, which allows linking to articles that don't exist at the time the article was written
ad serious reference: wikipedia is already frequently used in Newspapers and even court decisions
ad scalability: Yes I have to agree, wikipedia already scales bad.

ad wikipedia vs. dmoz. From my personal expierience I found dmoz quite unusable. The categories contained a lot of broken links, and the editors were unresponsive.
This is basically "the Cathedral vs. Bazar" Development Model found in Open Source.
In the beginning of Wikipedia there was peer review, but this model flopped, because not enough people were doing quality peer reviews at that time.
BTW, here a pointer to a similar discussion on the Wikipedia-Mailinglist
http://tomk32.blogger.de/stories/109987/

I don't have the time, nor do I see it as my task to correct incorrect Wikipedia articles, especially as when I did it on a couple of occasions in the past, many were overwritten again with more incorrect data. The incorrect entries should never have been made in the first place.

Are you saying that a model without a review is the solution to the problem that peer reviews didn't work? I hope not.

Frankly, I find what you are saying about the Wikipedia being used by newspapers and in court decisions very frightening indeed. It amounts to inviting some random person from the street who feels compelled to make a statement to give his "expertise" in court. If this is really happening, Wikipedia should be taken offline immediately.

Did you had a problem with another user disagreeing with your content? Usually I was able to resolve the conflict after a dialog with the other author. Or were you editing an article with high conflict potential (like the Iraq War of Fascism?), that is frequently vandalized?
I would really like to understand, why you have such strong feelings against Wikipedia, that you are even thinking about _censorship_. I don't understand why you think that wikipedia is so much worse than other websites.
Although I agree, that Wikipedia is not the best solution, I really like to read Wikipedia articles.

Beside the fact that you have some points here, your critisism would gain much readablity if you shortened the rant and cut out your personal involvement.

BTW, it's not about knowledge but information ;)

Sascha: "your criti[c]ism would gain much readab[i]lity if you shortened the rant".

I know that attention spans seem to be decreasing all the time, but this is ridiculous. Nine short paragraphs of about two or three sentences each is too much for readability?

"and cut out your personal involvement."

Sheesh! What do you expect? This is a personal weblog. I find the personal background quite useful in explaining Horst's motivation -- his "rant" is not just about Wikipedia from a user's, but also from a contributor's perspective.

Having seen some of my own Wikipedia contributions deleted or replaced by half-truths, conjecture disguised as fact, or just plain nonsense, I can relate to his frustration. As a minimum, Wikipedia contributors should be required to cite their sources.

Arved: "Apart from the large number of bloat articles there are some really well researched articles that are a lot more informative than the one in my dead-wood encyclopedia. (e.g. http://de.wikipedia.org/wiki/Rattenk%f6nig )"

Sure, there are some very nice articles among the many shoddy ones. But the point of an encyclopedia is trust: when I get my Britannica or Brockhaus from the shelf, I can be fairly confident that whatever article I pick has been written by someone who's competent in the field. The article has probably been peer-reviewed, spell-checked and de-fluffed by a good editor. That's why I paid for it in the first place.

With Wikipedia, I'm never quite sure. I cannot trust it more than a random Google hit.

ad Klaus: What happens if there is a bug in your Brockhaus? There is no easy way, you as a reader and maybe expert in one special field can correct it. You have to wait for the next edition before the readers get the fixed version.
What happens if you read a "random hit" from Google and notice a bug? Often it is not easy to find a contact address of the author, and even if there is a contact information, there is no gurantee this bug will ever be fixed.
There is no easy way to tell the world that this information is wrong.
The Wikipedia contributors believe that n+1 eyes find more bugs than n eyes.
I 100% agree with you, that an encylopedia is about trust. To help your decision about the trust of a Wikipedia article, you can also look at the whole history. If a fact has survived more than 100 edits, i would trust it more than one that has been added only recently. (One feature I am still missing in Wikipedia is a "cvs annotate" like functionality)
And btw you don't pay for wikipedia :-)

Like Horst you complained that your valuable content has been replaced by less-valuable content. I really would like to see an example.
I am contributing to Wikipedia since half a year and have around 40 articles on my Watchlist. Sometimes bad edits happen, but usually they get reverted within a few days.

Heh, this reminds me of the griping about GNU and Linux that was common in the early-to-mid 1990s - "amateurish", "anybody can mess it up", "not state of the art", "not reviewed by professionals", concluding with "it can never be successful" and "nobody would use it for real work".

Don't be too trusting of "professional" encyclopedias BTW, my research for WP articles has turned up all kinds of mistakes and omissions. In fact, some of the mistakes in WP are there because they've been trustingly propagated from poor encyclopedia entries.

"Frankly, I find what you are saying about the Wikipedia being used by newspapers and in court decisions very frightening indeed. ... If this is really happening, Wikipedia should be taken offline immediately."

It most certainly is true, as a modest amount of research will confirm. Please get Wikipedia taken offline immediately.

There is not one Wikipedia, there are many. There is an English wikipedia the biggest and there is a Klingon Wikipedia not the smallest.

When you compare all wikipedias, there is a recognisable maturation process. A wikipedia grows not only in number of articles but also in number of contributors. At start they are not perfect but as they mature, the quality improves.

When you compare Wikipedia to any other encyclopedia, it is young. It is finding its way fast and it changes rapidly. You may think it is not good enough for now. But it does provide increasingly better information. It will grow not always in the way that wou would like, but it will get better regardless.

The one big challenge of wikipedia is: will it be able to cross cultural devides. I am hopefull.
Thanks,
Gebruiker:GerardM

ad GerardM: I disagree, the number of contributors may influence the quality of single articles, but not on the whole Wikipedia, because the new contributors will have to learn first and until they learned the basics they will add countless mediocre articles.
And I think Horsts points are addressed to all Wikis.

My points are addressed to all wikis that claim (explicitly or implicitly) to be authoritative. If a Wikipedia claims to be an "online encyclopedia" and thus postulates that all information in it is correct, it ought to have quality control mechanisms that are adequate for an encyclopedia. I see no such mechanisms in place.

Again, this is not about correcting mistakes that have been made (because this task is simply impossible in such a big project), but avoiding the mistakes in the first place.

Wikipedia is an attempt to harness the power of online collaboration in order to build the largest reference work imaginable. And the best part is that the content is freely licensed and will be available as copyleft forever. Already, the English Wikipedia with its 275,000 articles is larger than any current general encyclopedia - see:

http://en.wikipedia.org/wiki/Wikipedia:Size_comparisons

Being totally open and built by volunteers, Wikipedia is far from perfect. There are inaccuracies - more than in most printed reference works. Copyright violations from webpages are the least of our problems, they are quickly detected, reported and deleted. Copyright violations from books are more difficult to detect, but the polished style is often a dead giveaway (Wikipedia articles start rough and become polished over time).

However, to believe that because Wikipedia is not as trustworthy as other sources now, it will always remain so is a deeply anti-intellectual attitude. In the last 3 1/2 years, we have gone from zero to beating Britannica, Encarta and Brockhaus in terms of quantity. We will also beat them in terms of quality. How? By using the same principles of wiki technology to create a peer review process.

Wikipedia is, after all, an outgrowth of the much more rigidly controlled Nupedia project, an open encyclopedia with a multi-stage peer review process by certified experts -- which managed to produce about 20 fairly decent articles. We have learned from this a lesson that such a process must not interfere with the fun people have building an encyclopedia.

You can read a bit about the ideas which have been tossed around since the very beginning of Wikipedia here:
http://en.wikipedia.org/wiki/Wikipedia:Approval_mechanism
http://en.wikipedia.org/wiki/User:Eloquence/Platform#Quality_control_and_print_editions

In short, a peer review process is inevitable. Just like we have managed to build 275,000 articles in 3 1/2 years, we will manage to systematically review them in the same time. Eventually there will be an unstable area of Wikipedia - which you can check if you need the latest information, similar to a preprint server - and a stable one, where all facts have been checked.

If you want to see some of the best examples of our work so far, check out:
http://en.wikipedia.org/wiki/WP:FA

Unfortunately Horst is partially right in his criticism of Wikipedia. But only partially. There is no inherent reason why the wikipedia principle should not work given that all people working on wikipedia would be aware that they are working on an *encyclopaedia*. Unfortunatly many people - even many people which are with Wikipedia from its very beginning - think of Wikipedia as being a *Wiki* in the first place. They do not believe that a strict policy on what is accepted as an encyclopedic subject is needed. That generates noise and, more dangerous, it creates an atmosphere where people think that google is sufficent as a research tool. What Wikipedia would need are more people working as editors, throwing out content that is badly researched, pushing Wikipedia towards quality.

However, Wikipedia is just at it's beginning. Keep in mind that the german Wikipedia is just three years old. Don't expect us to be on the Brockhaus or Britannica level by today. I expect Wikipedia to be a reliable encyclopaedia in ten or fifteen years from now, if we manage to dam up the noise. If we don't manage this, then we end up as a second WWW indeed, not as being the most reliable source of information on the internet. The danger does exist, indeed, but as there are also people being aware of that danger, it's unlikely that worst case scenario will happen.

Uli

Erik: What you wrote here sounds like it's copied from a marketing brochure. As such, it totally fails to convince me. I am also quite unimpressed by the number of articles. Reference is a business in which only quality matters. At least that is my experience as a reference librarian.

Uli: Sounds like you got my point. But tell me: with the sheer number of articles that Erik is quoting, and assuming that the number is steadily rising, how will you be able to impose any kind of quality control over this huge body of data, especially as the data is constantly being altered by more or less random people? I think it's already beyond anybody's control.

I see your point. Right at the moment most articles in Wikipedia are not authoritative information.

But Wikipedia is a selflearning, selfregulating system. Thus it gets better day by day. Good authors will respect more authoritative knowledge, while rubbish will be deleted if every author maintains his articles.

in the following you will see the setup of diverse "epistemic communities" caring for the validity of information! So in my eyes wikipedia is becoming the vastest authoritative encyclopedia ever.

Update: Schockwellenreiter says that institutional quality control is unnecessary because it is imposed by the readers, who are supposedly able to determine the quality of the article they are reading. Put mildly, I'd say that's _very_ optimistic. I should invite him to work with me one of these days.

Question Authority, even if "encyclopedia" is written on its cover!

a few comments on the comments:

- comparing Wikipedia to Linux is rather misleading. actually, it's comparing apples and oranges. Linux and other successful open source software projects are characterised by strong leadership and peer review, things that are missing from Wikipedia. do you know any successful open source project that gives commit rights to anyone? that doesn't have clear responsibilities for modules?

- yes, Wikipedia entries can be corrected by anyone anytime. they can also be messed up by anyone anytime. how am i to tell if an entry is currently in messed-up or good state? do i really have to comb through the version history for each sentence? and then guess that if something survived 100 edits by possibly a bunch of nitwits it is indeed correct? great.

dead-tree enyclopedias are not perfect (and what is?), but they are at least able to maintain a certain quality minimum throughout all entries. actually I think that a comments model would be best, i.e. entries are edited only by designated editors, but anyone can add comments, or suggest new entries perhaps. that's how a growing number of open source software projects maintain their documentation, and it works ok.

i'm not against wikis in general, but wikipedia shows that they are not for everything. thanks to horst for his politically incorrect rant :)

Horst: "how will you be able to impose any kind of quality control over this huge body of data, especially as the data is constantly being altered by more or less random people?"

The Huge body of data can be peer-reviewed by a huge number of people. If an articles is altered it has to be checked again but there is still the old trustworthy version. At the moment Wikipedia is not en encyclopedia but parts of it. You do not have to peer-review all the 300.000 articles but parts of it. If you collect these parts you'll get an encyclopedia.

Unfortunately there is no function to rate articles and edits yet but I hope we will get it soon, so you could see which expert trusts which version of which article.

I suppose that a lot of articles and edits won't get rated high, but as the number of articles and editors keeps growing, the number of highly rated articles also growths and these articles will form the encyclopedia.

How quickly history is forgotten! GNU/Linux has always had strong leadership and peer review, but most people thought it was doomed to early failure anyway, and would tell me about all their fancy reasoning supporting their belief, how foolish it was to work on GNU for a living, etc. I've gotten to enjoy the distinct pleasure of having known some of the doubters long enough that they've had to admit they were wrong. Some still don't understand *why* they were wrong.

Wikipedia is the same way, in that there is no past experience suggesting that it could possibly work as well as it has so far. But the empirical evidence is that it will surpass other sources of information within a few years, and people should start adjusting their theories to fit the facts, not tell us the facts can't be true because the theory doesn't allow for it.

Klaus: *rant* First off all thanks for straightening out my sloppy spelling. I am very sure most readers here couldn't have done this on their own ;) *rant* scnr

As I said earlier, Horst has some valid points. But by mixing up serious criticism with personal rants and anger Horst gives away much of the readers' confidence.

Just my 2 cents. Rip it apart, leave it alone or go get a red pen ;)

Sorry for self-reply, I simply read over arveds reply:

> arved said on June 2, 2004 11:44 PM:
> Question Authority, even if "encyclopedia" is
> written on its cover!

First thing you learn as journalist, information broker, researcher, etc. Always double-check your sources. Even newspapers, magazins, authors, editors make mistakes.

Balasz C: "do you know any successful open source project that gives commit rights to anyone? "
: Software is a single unit that is only working if all pieces work. Wikipedia articles (normally) don't depend on their neighbouring articles.

My opinion is that we don't need regulations on peer review, they would only slow us down and make servers faster.
There is a kind of peer, we've started doing WikiReader which are collections of 50-150 articles. Also there's the Qualitätsoffensive, our second method of improving a certain topic.

Wikipedia is no encyclopedia but a revolution!

Yet another de:Wikipedia Admin

Hi Horst,

it took me some time to write an answer to your original posting. I've placed it here.

Here's a copy of it (no links, no layout, no text formating):

[The copy of Mathias' article was removed from here because it was no less than 2,692 words (about five pages) long. Please follow this link to read it. —Horst]

Sorry, I couldn't correct your point 2 because this isn't a wiki. Please consider reconciling your use of "no editorial selection" and "Some entries". More seriously, you have made some valid points and those points are being addressed as the project evolves.

1. It's very mixed. Often what happens is those who don't know a subject well write on it and those who know it well correct any errors. If you see errors, that's fine - if time allows, please correct them, providing one or more persuasive sources for others to see. If you don't have time, I expect that eventually someone will choose to correct the errors, since the project is steadily rising in prominence. The lack of a system for collecting and archiving source information is one of the current limitations of the project. However, lack of formal structures is also a deliberate objective of wiki originator Ward Cunningham, since it is intended to allow rapid evolution and flexibility. Perhaps the most important thing to remember about a wiki of this sort is that the end result is not intended to be the version you see today. It's supposed to improve over time.

2. There's continued interest in quality or review rating systems, since there is clear need for some sort of quality guide. What is known is that prior peer review has proved ineffective for the goals of the project - the preceding Nupedia project essentially failed using that model. Proposals include such things as simply counting views since an edit, counting views by administrators or some more formal review system. It seems likely that initial versions will be some sort of view counting, as a way to assure basic quality - the elimination of deliberate vandalism. The proposals often include showing anonymous users only the "known quality" version of an article.

3. That Google ranking rise is increasing the number of well qualified individuals who will find the work on their specialist topic well ranked. That increases the probability of corrections and the likely overall quality of the project. It will take time for esoteric topics, or topics of low general interest, to evolve.

4. The ability of copyright holders to do something about it are limited, in part by copyright law, which protects expressions, not ideas. The ability to do something about it is enhanced by the easy editability and because those involved in the project know that redistributable content is an objective. For that reason, when text is reported as a possible copyright infringement it's usually replaced, in part because it's relatively easy to rewrite and ensure a lack of infringement. Close paraphrasings do get reported as possible copyright infringements and eliminated or rewritten. If you notice one, please feel free to report or correct it. Or to ignore it if you don't want to spend the time. Someone else will get to it eventually, or it'll be enhanced until it's no longer sufficiently close to the original source for that to be anything other than a de minimis and ignorable part of the work for copyright purposes.

The growth of an article due to the participation of enthusiasts as opposed to the smaller size of article or more filtered content which you appear to consider appropriate is an interesting problem for the project, since there is some merit in both approaches. The depth to which an article should cover a subject is not currently well defined. It seems fairly likely that as the number of experts rises the depth will also rise but that is speculation.

Please see http://www.artima.com/intv/wiki.html for an interview with Ward Cunningham which may explain some of the basic premises of a wiki.

Please see the paper at http://journalism.utexas.edu/onlinejournalism/wikipedia.pdf for a review of the Wikipedia as a source. You may find the positive effect of press use of the Wikipedia as a source on the quality of an article of interest, since that supports one of the fundamental principles of the project: that an increase in the number of viewers will lead to an increase in quality.

This post is hereby licensed under either version of the BSD license. Any license on my user page (at the supplied URL) may also be used. Other licenses are available on request.

wow, talking about wasp nests.

speechless, literally

Dear Horst, you misinterpret Wikipedia. This is not a place to gain "accurate information". That is what real encyclopedias are meant for. Wikipedia is supposed to be faster than Google when you just want to find the explanation for some word.

And it does that task splendidly for me. For which I am most grateful. If it is important to get it right, I look up a researched encyclopedia.

And you were complaining about the Atom feed discussion...

Mathias: Sorry for deleting your comment, but I felt that 2,692 words was more than an excessive length for a comment. If you have such a lot to say, please say it on your weblog and send me a Trackback ping, because that's exactly what Trackbacks are for.

Horst: It's okay the way you stripped my comment :)

Wikipedia is a true disaster. 'Common opinion' and popular belief dominate, and expert views that do not correspond with those, are banned. Many articles are badly written, and spelling mistakes are a rule rather than an exception.

I have participated in several language versions of the project for a couple of months, and have lost all faith in the project ever turning into something useful. I have found it impossible to work in the Wikipedia framework. Calls for previous research are called 'elitist', 'academic', 'unpractical' by the moderators. Vague claims such as 'some say that' or 'There are people who believe that' are defended by sword and fire. Research is considered impossible, because 'more specific information cannot be found on the Internet', research would be 'too strenuous'.

Wikipedia does not even want to be on an academic level, it wants to reflect popular belief. I sincerely hope this poor excuse of an encyclopedia is not ever going to be consulted as a serious source.

Apparently, it IS used as a source of serious information. I just received an invitation to a conference in linguistics, that is accompanied by a description of the region where the conference is to be held. It contains a very familiar text in German, that I helped write for the German Wikipedia some time ago. :-(