Feng Shui and Wikipedia

[grunion]

Though by its very nature it is fraught with misinformation, I have come to look at Wikipedia (http://en.wikipedia.org/wiki/Main_Page) as a fairly objective resource, particularly when it comes to pseudoscience and quackery. It tends to give me faith in the general intelligence of the Internet population when I see sensible, accurate information generally prevail over the agendas of some faction or another. See, for example, their entries on Homeopathy, Alternative Medicine, Dianetics, and Intelligent Design. (My apologies for being unable to provide links as appropriate to these pages as it appears that HTML tags are not functional here.)

Less satisfactory, but still informative, are their entries on Chiropractic and Reflexology. Their entry on Applied Kinesiology looks as if it has been batted around too many times to make much sense anymore. Their entry on Acupuncture is quite bad, written by an apparent True Believer with only the slightest of nods to those who may want to objectively and scientifically consider any of its claims.

But their entry on Feng Shui is packed with outright hokum such as:
"recent discoveries and effective applications have become popular largely because, like many alternative healing practices, they work"
"the rapid modernization of China has led to feng shui becoming a worthy subject for scholarly inquiry at Chinese universities"
"businesses generally use feng shui to increase sales and boost morale"

The article is neither accurate nor objective and is clearly an advertisement to lure the ignorant into throwing money at the scam artists who claim to be tuned into the supernatural and imbued with special powers to determine the best placement of architectural features, furniture, or interior decorating choices.

Unfortunately I neither possess the subject matter expertise nor the writing skills to edit these articles. My attempt to edit the piece would likely consist of my adding the phrase "that is a lie" liberally throughout the article. Is there some sort of skeptical army out there to which we can refer these egregious claims for refutation and correction? I will be glad to participate.

[Unnamed]

The closest you can get to "that is a lie" is to add one of the dispute templates to the article. See http://en.wikipedia.org/wiki/Wikipedia:Template_messages/Disputes

You can put a {{POV}} mark at the beginning and write a note on the Talk page (very similar to what you have here).

[Imrational]

I wonder what wikipedia is going to look like in the distant future. Will it represent knowledge and critical thinking, or will it be drowning in pseudoscience, pop culture, and the like?

Is wikipedia a model for our society? Is it a collection of memes that represent our current culture? If it drowns, does that imply that our culture will also?

[drkitten]

I wonder what wikipedia is going to look like in the distant future. Will it represent knowledge and critical thinking, or will it be drowning in pseudoscience, pop culture, and the like?

Jury is still out on that. My bet is that USENET -- or the Web itself -- provides a good model for what will happen to it.

Back in the day ("when I was your age, we had to walk to school, uphill both ways, through a solid block of ice -- and we liked it"), computers were largely the domain of the technically knowledgeable and highly intelligent, which made it possible to get large amounts of very accurate information on specific subjects very quickly.

Now, of course, it's a vast wasteland of misinformation, as the inmates have gradually taken over the asylum.

[tkingdoll]

Now, of course, it's a vast wasteland of misinformation, as the inmates have gradually taken over the asylum.

That's excellent, I'm going to borrow that analogy if I may. I'm always looking for cute ways of demonstrating why Wikipedia, though useful, should not be regarded as the definitive authority on any subject.

[geni]

That's excellent, I'm going to borrow that analogy if I may. I'm always looking for cute ways of demonstrating why Wikipedia, though useful, should not be regarded as the definitive authority on any subject.

Why? It's a just a reworking of the old "oh noes anyone can edit".

Surely the logical thing to do would be to point out wikipedia's disclaimers.

[geni]

I wonder what wikipedia is going to look like in the distant future. Will it represent knowledge and critical thinking, or will it be drowning in pseudoscience, pop culture, and the like?

No


Is wikipedia a model for our society?

No.


Is it a collection of memes that represent our current culture?


No


If it drowns, does that imply that our culture will also?

No

[geni]

The closest you can get to "that is a lie" is to add one of the dispute templates to the article. See http://en.wikipedia.org/wiki/Wikipedia:Template_messages/Disputes

You can put a {{POV}} mark at the beginning and write a note on the Talk page (very similar to what you have here).

{{verify}} or {{totallydisputed}} would probably be more correct.

[tkingdoll]

Why? It's a just a reworking of the old "oh noes anyone can edit".

Surely the logical thing to do would be to point out wikipedia's disclaimers.

Why not? If I like something, I like it. Why do you care?

You may not realise this, but in the real world, some people respond better to analogies than than the quoting of legal disclaimers. When educating individuals about the dangers of relying on Wikipedia, it's useful to be able to engage them rather than send them to sleep.

But, whatever.

[Jeff Corey]

Ask Askolnik about his experiences at that useless "reference" site.

[geni]

Why not? If I like something, I like it. Why do you care?

You may not realise this, but in the real world, some people respond better to analogies than than the quoting of legal disclaimers.

Of course. Why would people bother comitting the appeal to emotion logical fallacy otherwise.


When educating individuals about the dangers of relying on Wikipedia, it's useful to be able to engage them rather than send them to sleep.

You can rely on wikipedia. The reason is that once you reed the disclaimers you relise that wikipedia isn't claiming anything. There are also other very good reasons to read through this disclaimers. It is so pages such as this don't come as a shock:

http://en.wikipedia.org/wiki/Jyllands-Posten_Muhammad_cartoons_controversy

[geni]

Ask Askolnik about his experiences at that useless "reference" site.

We could or we could take the view that sometimes it is better to look at scientific studies rather than relying on anicdotal evidence.

http://www.nature.com/nature/journal/v438/n7070/full/438900a.html

Intestingly one of the errors was traced back to one of the review's books.

[Complexity]

I've never used Wikipedia. I'm not looking to use it in the near future.

I don't think I'd find it very trustworthy.

[tsg]

Wikipedia is a useful tool providing you excercise a little, dare I say it, critical thinking and common sense.

It is not an authority on any subject, merely a collection of articles written by many different people with varying levels of accuracy. Saying Wikipedia is or is not a reliable resource is like saying the Internet is or is not a reliable resource. Each article is only as good as its references and if any article, be it it Wikipedia or the Encyclopedia Brittanica, makes claims without references or citations, it should not be believed without corroborating evidence.

In short, don't believe everything you read no matter where you read it.

[John de Combe]

One of the worst examples of BS I've found on Wikipedia is:


http://en.wikipedia.org/wiki/Doris_Stokes (http://en.wikipedia.org/wiki/Doris_Stokes)

[Mojo]

You can rely on wikipedia. The reason is that once you reed the disclaimers you relise that wikipedia isn't claiming anything.You can rely on Wikipedia for, er, nothing then? :D

[geni]

You can rely on Wikipedia for, er, nothing then? :D

You can rely on it for disclaimers and it can be hard to argue with articles that have 87 references:

http://en.wikipedia.org/wiki/Katie_Holmes

[Mojo]

You can rely on it for disclaimers :dl:

[Darat]

One of the worst examples of BS I've found on Wikipedia is:


http://en.wikipedia.org/wiki/Doris_Stokes (http://en.wikipedia.org/wiki/Doris_Stokes)

NPOV! :D

[Mojo]

... it can be hard to argue with articles that have 87 references:

http://en.wikipedia.org/wiki/Katie_HolmesThat depends on the quality of the references and whether they actually support what the article says. Here (http://www.bailii.org/ew/cases/EWHC/Fam/2003/1376.html), for example, is a case in which a report with over 120 references was successfully argued against (scroll down to the section about Dr. Donegan's evidence at paragraph 37). From the judgement in the appeal (http://www.bailii.org/cgi-bin/markup.cgi?doc=/ew/cases/EWCA/Civ/2003/1148.html):
The judge concluded that the medical evidence relied on by the two mothers to show that vaccination is dangerous and unnecessary was untenable. Dr Donegan's report was based on no independent research, and most of the published papers cited by her in support of her views turned out either to support the contrary position or at least to give no support to her own. Not to mince words, the court below was presented with junk science.

[geni]

That depends on the quality of the references and whether they actually support what the article says.


As a general rule anthing with over say 10 references is the result of two or more sides dissagreeing with the result that people tend to go over each others references with a fine tooth comb looking for any excuse to reject them.

[Darat]

As a general rule anthing with over say 10 references is the result of two or more sides dissagreeing with the result that people tend to go over each others references with a fine tooth comb looking for any excuse to reject them.


Is that in one of the reliable disclaimers? ;)

[geni]

Is that in one of the reliable disclaimers? ;)

No. If it was they would not be called disclaimers.

[Darat]

No. If it was they would not be called disclaimers.

Why ever not?

[CFLarsen]

No. If it was they would not be called disclaimers.

The disclaimer at the end of John Edward's "Crossing Over" is not a disclaimer?

[geni]

Why ever not?
Because it would mean that wikipedia was accepting responcibilty for something in some way shape or form.

[Darat]

Because it would mean that wikipedia was accepting responcibilty for something in some way shape or form.

How?

[geni]

How?

Well in this case it would mean that wikipedia took responcibilty for the claim. There are of course exceptions. Algerian Civil War racks up 11 references but hasn't really be edit wared over.

http://en.wikipedia.org/wiki/Algerian_Civil_War

[Darat]

Well in this case it would mean that wikipedia took responcibilty for the claim. There are of course exceptions. Algerian Civil War racks up 11 references but hasn't really be edit wared over.

http://en.wikipedia.org/wiki/Algerian_Civil_War

In what way?

[drkitten]

Because it would mean that wikipedia was accepting responcibilty for something in some way shape or form.

You know, one might argue that this is part of the problem. Part of the reason that USENET and the Web went to hell as quickly as they did is because no one was accepting responsibility for it, while the OED and Brittanica are still accepted as reliable sources after a century....

[tsg]

One of the worst examples of BS I've found on Wikipedia is:


http://en.wikipedia.org/wiki/Doris_Stokes (http://en.wikipedia.org/wiki/Doris_Stokes)

Zero references, one external link to a very obviously woo "spiritualist union" website, three books in the Bibliography, two of which Doris Stokes wrote and the other a "tribute". It's pretty easy to dismiss this article out of hand and I have never heard of Doris Stokes.

I also see that it's been flagged as disputed since I looked at it the first time 30 minutes ago.

[tsg]

You can rely on it for disclaimers and it can be hard to argue with articles that have 87 references:

http://en.wikipedia.org/wiki/Katie_Holmes

Is there something inaccurate in the article that you have a problem with?

[CFLarsen]

I also see that it's been flagged as disputed since I looked at it the first time 30 minutes ago.

Does anyone know how many entries in Wikipedia that has been flagged as disputed? Out of how many in all?

It seems to me a group of people could destroy Wikipedia this way, simply by disputing everything.

[tsg]

Does anyone know how many entries in Wikipedia that has been flagged as disputed? Out of how many in all?

The best I can figure, out of 973,456 articles (http://en.wikipedia.org/wiki/Special:Statistics), over 2000 articles (http://en.wikipedia.org/wiki/Category:NPOV_disputes) have been tagged as NPOV (neutral point of view) Dispute, and 197 (http://en.wikipedia.org/wiki/Category:Accuracy_disputes) as Accuracy Dispute. It isn't clear if a TotallyDisputed article counts in either or both lists.

It seems to me a group of people could destroy Wikipedia this way, simply by disputing everything.

Dispute tags can be removed just as easily as they are added. Unless you list some specific reasons on the talk page why the accuracy or neutrality of the article is in dispute, someone will just remove the tag.

[grunion]

I see someone added a "unverified" tag to the Feng Shui entry and began editing it with an eye toward accuracy. I added a number of additional edits to one section, yet a great deal of work still needs to be done to fix this scam job.

[geni]

It seems to me a group of people could destroy Wikipedia this way, simply by disputing everything.

People have tried. I can revert your edits at a rate of greater than once per second. How fast can you edit?

[geni]

The best I can figure, out of 973,456 articles (http://en.wikipedia.org/wiki/Special:Statistics), over 2000 articles (http://en.wikipedia.org/wiki/Category:NPOV_disputes) have been tagged as NPOV (neutral point of view) Dispute, and 197 (http://en.wikipedia.org/wiki/Category:Accuracy_disputes) as Accuracy Dispute. It isn't clear if a TotallyDisputed article counts in either or both lists.

TotallyDisputed should be both.


Dispute tags can be removed just as easily as they are added. Unless you list some specific reasons on the talk page why the accuracy or neutrality of the article is in dispute, someone will just remove the tag.

Of course some people then chose to edit war over wether an article is dissputed.

[geni]

Is there something inaccurate in the article that you have a problem with?

No idea. It just has the second highest number of references of any article I know of.

[geni]

You know, one might argue that this is part of the problem. Part of the reason that USENET and the Web went to hell as quickly as they did is because no one was accepting responsibility for it, while the OED and Brittanica are still accepted as reliable sources after a century....

Brittanica didn't do much better than wikipedia in the nature study (and isn't selling too well although encarta probably has a lot to do with that). EB 1911 is public domain. It has it's fans but most of them would admit it has point of view problems.

[Darat]

Well you could say it did around 30% worse then Britannica. :) (http://manojar.blogspot.com/2005/12/wiki-or-britannia.html)


ETA:


Which?|Count%|better
Equal|10|23.81%
Britannica|22|52.38%
Wikipedia|10|23.81%

[drkitten]

Brittanica didn't do much better than wikipedia in the nature study

... if you consider a 30% lower error rate "not too much better," then, yes.

Furthermore, the Nature study specifically studied one of the strengths of Wikipedia -- the technical articles. I would be very surprised to see if the historical, and especially political, articles, could achieve the same degree of factual accuracy in Wikipedia. There aren't that many nutcases with axes to grind about IEEE 754 floating point specifications.

[geni]

... if you consider a 30% lower error rate "not too much better," then, yes.

Same order of magnitude and in many cases they have had over 100 years to get it right.


Furthermore, the Nature study specifically studied one of the strengths of Wikipedia -- the technical articles. I would be very surprised to see if the historical, and especially political, articles, could achieve the same degree of factual accuracy in Wikipedia.

Dito Britannic. Factual accuracy isn't much difffernt on contenious politial articles. They tend to be factual for much the same reason that anything the survives intact in a blast furnace tends to have a high melting point.


There aren't that many nutcases with axes to grind about IEEE 754 floating point specifications.

So? Nutcases don't really have much of an effect on factal accury.

[tsg]

No idea. It just has the second highest number of references of any article I know of.

It struck me as one of the better articles in Wikipedia with citations for just about every fact. I didn't check the resources to see if they were accurate, but I also don't care about Katie Holmes enough to bother.

I was just wondering why you brought it up.

[geni]

Well you could say it did around 30% worse then Britannica. :)


You can play with the data in all sorts of ways (for example you could say that since more wikipedia articles were found to be without error then wikipedia has more good articles). I think I'm going to stick to what the journal did.

[geni]

It struck me as one of the better articles in Wikipedia with citations for just about every fact. I didn't check the resources to see if they were accurate, but I also don't care about Katie Holmes enough to bother.

I was just wondering why you brought it up.

Just pointing out what references can do.

[drkitten]

Dito Britannic. Factual accuracy isn't much difffernt on contenious politial articles.

Evidence?

[geni]

Evidence?

Various smaller media reports. For example the lastest would be the independent article:

http://news.independent.co.uk/media/article345103.ece
We do ok on politics but less well on history in this case.

[drkitten]

Various smaller media reports. For example the lastest would be the independent article:

http://news.independent.co.uk/media/article345103.ece
We do ok on politics but less well on history in this case.

Thanks for the citation. I'm not entirely sure how to read it, though; the article reads like a summary of a critical review of various Wikipedia entries (many of them of dubious quality), but it doesn't do a comparison of factual accuracy vs. Brittanica. Did I miss something?

[CFLarsen]

The best I can figure, out of 973,456 articles (http://en.wikipedia.org/wiki/Special:Statistics), over 2000 articles (http://en.wikipedia.org/wiki/Category:NPOV_disputes) have been tagged as NPOV (neutral point of view) Dispute, and 197 (http://en.wikipedia.org/wiki/Category:Accuracy_disputes) as Accuracy Dispute. It isn't clear if a TotallyDisputed article counts in either or both lists.

Thanks!

Dispute tags can be removed just as easily as they are added. Unless you list some specific reasons on the talk page why the accuracy or neutrality of the article is in dispute, someone will just remove the tag.

This strengthens my concern.

People have tried. I can revert your edits at a rate of greater than once per second. How fast can you edit?

I can get 20 people who, combined, can effectively remove anything you edit. If you get 40, I get 100. And so on. Now what?

[tsg]

This strengthens my concern.

Which is what, exactly?

[geni]

Thanks!
This strengthens my concern.


Not really. Watchlists mean we will slightly notice if someone removes the tag


I can get 20 people who, combined, can effectively remove anything you edit.


Been tried. I won.


If you get 40, I get 100. And so on. Now what?

I don't need 40. I need at most one other person

[geni]

Thanks for the citation. I'm not entirely sure how to read it, though; the article reads like a summary of a critical review of various Wikipedia entries (many of them of dubious quality), but it doesn't do a comparison of factual accuracy vs. Brittanica. Did I miss something?

I'm not aware of any other comparisions with britanica. The german wikipedia was compared to Brockhaus and the german version of encarta and came out on top.

Britancia has exactly the same real issues as wikipedia when it comes to politics and history (the "people will dissagree over it" is a destraction).

[luchog]

We could or we could take the view that sometimes it is better to look at scientific studies rather than relying on anicdotal evidence.

http://www.nature.com/nature/journal/v438/n7070/full/438900a.html

Intestingly one of the errors was traced back to one of the review's books.
What you're not telling, however, is that the study was too flawed to use as a general survey of quality.

1) It focussed only one one very limited field, scientific information, long acknowledged even by detractors as Wiki's strong point. It's much easier to keep scientific information accurate. Other, less precise fields, such as history and sociology, were not tested, and are far more likely to have significant errors due to author agenda and POV, as well as vandalism.

2) If you read the study, you'll notice that Wiki still had a statistically significant higher number of profound errors, not total errors, than EB. Minor errors were comperable. In EB's case, the number of profound errors are due in large part to recent advances that have invalidated some of the older information. Wiki has not kept up with recent advances significantly better.

3) Due to vandalism and agenda-pushing; articles on Wiki are not static, but can change daily, even hourly. There is no guarantee that the information you're viewing at any particular moment is accurate, and not the result of changes made shortly prior to viewing. There is also no guarantee that information that may be accurate at the moment will not be altered subsequently.

Oh, remember that experiment I announced a few months back, about those three significant factual errors I have been observing? Guess how many of them have been corrected? (Hint: if guess>0 then guess again.) I also noticed another major error in an article, and corrected it; only to have the article reverted back to it's previous erroneous state. Guess that "infinite monkeys" approach doesn't work as well as the wikivangelists would have us beleive.

[luchog]

Britancia has exactly the same real issues as wikipedia when it comes to politics and history (the "people will dissagree over it" is a destraction).
Provide evidence.

[drkitten]

Britancia has exactly the same real issues as wikipedia when it comes to politics and history.

With respect, I think this is the same claim for which I asked you to provide evidence a half-dozen posts ago. The article you referred me to didn't really address this claim )the comparative accuracy of Brittanica in history/politics) , so I think this still qualifies as "assertion without evidence."

[geni]

What you're not telling, however, is that the study was too flawed to use as a general survey of quality.

1) It focussed only one one very limited field, scientific information, long acknowledged even by detractors as Wiki's strong point.

Evidence?

It's much easier to keep scientific information accurate.

Not it isn't since the stuff keeps changeing.


Other, less precise fields, such as history and sociology, were not tested, and are far more likely to have significant errors due to author agenda and POV, as well as vandalism.


Nah vandalim hits scientific articles as well (particularly dolly the sheep for some reason).

Author agenda is less of a problem than you would think. Something to do with the issue of multiple authors


2) If you read the study, you'll notice that Wiki still had a statistically significant higher number of profound errors, not total errors, than EB. Minor errors were comperable. In EB's case, the number of profound errors are due in large part to recent advances that have invalidated some of the older information. Wiki has not kept up with recent advances significantly better.

Wikipedia is 5 years old. Britanica is over a centry old.


3) Due to vandalism and agenda-pushing; articles on Wiki are not static, but can change daily, even hourly. There is no guarantee that the information you're viewing at any particular moment is accurate, and not the result of changes made shortly prior to viewing. There is also no guarantee that information that may be accurate at the moment will not be altered subsequently.

Fourth tab along is the history tab. It exists for a reason


Oh, remember that experiment I announced a few months back, about those three significant factual errors I have been observing? Guess how many of them have been corrected? (Hint: if guess>0 then guess again.) I also noticed another major error in an article, and corrected it; only to have the article reverted back to it's previous erroneous state.

Details?


Guess that "infinite monkeys" approach doesn't work as well as the wikivangelists would have us beleive.

Dunno we will find out if anyone ever tries the "infinite monkeys" approach.

[geni]

Provide evidence.

Britanica emplooys people who have a reasonable level of knowage of the field. It is to be expected that they are likely to have at least some POV.

[geni]

With respect, I think this is the same claim for which I asked you to provide evidence a half-dozen posts ago.

No. This claim is about the source of problems.