More Recent Comments

Thursday, August 18, 2022

The trouble with Wikipedia

I used to think that Wikipedia was a pretty good source of information even for scientific subjects. It wasn't perfect, but most of the articles could be fixed.

I was wrong. It took me more than two months to make the article on Non-coding DNA acceptable and my changes met with considerable resistance. Along the way, I learned that the old article on Junk DNA had been deleted ten years ago because the general scientific consensus was that junk DNA doesn't exist. So I started to work on a new "Junk DNA" article only to discover that it was going to be very difficult to get it approved. The powerful cult of experienced Wikipedia editors were clearly going to withhold approval of a new article on that subject.

I tried editing some other articles in order to correct misinformation but I ran into the same kind of resistance [see Allele, Gene, Human genome, Evolution, Alternative splicing, Intron]. Frequently, strange editors pop out the woodwork to restore (revert) my attempts on the grounds that I was refuting well-sourced information. I even had one editor named tgeorgescu tell me that, "Friend, Wikipedians aren’t interested in what you know. They are interested in what you can cite, i.e. reliable sources."

How can you tell which sources are reliable unless you know something about the subject?

Much of this bad behavior is covered in a Wikipedia article on Why Wikipedia is not so great. Here's the part that concerns me the most.

People revert edits without explaining themselves (Example: an edit on Economics) (a proper explanation usually works better on the talk page than in an edit summary). Then, when somebody reverts, also without an explanation, an edit war often results. There's not enough grounding in Wikiquette to explain that reverts without comments are inconsiderate and almost never justified except for spam and simple vandalism, and even in those cases comments need to be made for tracking purposes.

There's a culture of hostility and conflict rather than of good will and cooperation. Even experienced Wikipedians fail to assume good faith in their collaborators. It seems fighting off perceived intruders and making egotistical reversions are a higher priority than incorporating helpful collaborators into Wikipedia's community. Glaring errors and omissions are completely ignored by veteran Wikiholics (many of whom pose as scientists, for example, but have no verifiable credentials) who have nothing to contribute but egotistical reverts.

In another article on Criticism of Wikipedia the contributors raise a number of issues including the bad behavior of the cult of long-time Wikipedia editors. It also points out that anonymous editors who refuse to reveal their identify and areas of expertise leads to a lack of accountability.

This sort of behavior is frustrating and it has an effect. Well-meaning scientists are quickly discouraged from fixing articles because of all the hassle they have to go through.

I now see that the problem can't be easily fixed and Wikipedia science articles are not reliable.


14 comments :

Anonymous said...

I love wiki/ its great onalmost beverything. I think its one of the great inventions for mankind to quicky get info.
YES everyone complains. i know ID/YEC creationists bitterly complain about editing and subjects brought up here in this thread.
I don't like it said its unreliable. its just normal incompetent as one finds everywhere. In science subjects by those who know the subject, as in this thread, it is discovered on the more complicated things the incompetence.

Joe Felsenstein said...

The obsession with finding citations for everything leads to completely silly problems. On my own page, at one point they indicated that there was some controversy about my birth date. I have never heard anyone worry about that before. I supplied the correct date, but they could not accept that. Instead they were only satisfied when they found it on a web page for an award that I had won. And guess who the award people got that date from?

Joe Felsenstein said...

Wikipedia is going to always reflect dominant-consensus views. If there had been Wikipedia in the 1500s, it would have described the Divine Right of Kings as well-established and noncontroversial. When the dominant-consensus view today is wrongheaded, as in the case of the widespread dismissal of junk DNA by molecular biologists and genomicists, Wikipedia will put its head firmly "where the sun don't shine". At the same time, I have heard that when Wikipedia started, statisticians had an organized project to make it correct, and different statisticians were assigned to particular pages covering statistical concepts. The result was very high quality.

Anonymous said...

Robert Byers wrote this comment. not important but I can't figure out how to put my name instead of anonymous. Not important.

Anonymous said...

That's hilarious.
-César D

Larry Moran said...

I have changed the way my blog handles comments because the old way (embedded) requires the use of third party cookies and some of us don’t want to allow that on our browsers.

More people should be able to comment now but it looks really ugly.

Larry Moran said...

@Joe

It’s true that Wikipedia tends to favor the dominant popular view but it also allows an abundance of irrelevant and incorrect viewpoints long as they are supported by the magical “sources.” It’s extremely difficult to remove something with a “reliable source” because those entries are jealously protected by the Wikipolice.

I bet that if a prominent journal published an incorrect birthdate for you then you would not be able to correct it!

Peter Turney said...

I agree that Wikipedia policies are somewhat problematic. Perhaps you could suggest a new set of policies that would address the problems you have encountered? If your new policies are clearly an improvement on the old policies, it might be possible to persuade Wikipedia to update its policies.

It seems that the core problem is how to resolve disagreements. What exactly is a reliable source? There is no easy answer to this question. Perhaps some formula based on the recency and citation counts of relevant publications could be used to resolve disputes? There will be some arbitrariness in any formula, but it would be less arbitrary than the current policy.

Larry Moran said...

@Peter Turney

The policies are not unreasonable. It's the enforcement of those policies by intransigent Wikipedia editors that causes most of the problems. They need to understand that determining which sources are reliable and which ones are relevant requires a substantial amount of knowledge about the subject.

I understand that they can't just trust every user who claims to be an expert but there are ways of determining whether a give user is responsible and trustworthy. You can do this by looking at their qualifications and at the arguments they make on the Talk page to justify their edits. If they get support from other qualified experts then that's a good clue.

Unqualified editors who know nothing about the subject should be treated skeptically when they make accusations. They don't deserve the benefit of the doubt when they are trying to ban someone who is knowledgeable about the subject.

One policy change that I would encourage is that all accusations of misbehavior should be judged by editors who have established credentials in the subject matter. I would also make it a rule that all editors who are granted extra authority have to reveal their identify and their areas of expertise. We can't have secret kangaroo courts where you get accused and judged within minutes by people who hide behind secrecy and anonymity.

Peter Turney said...

"One policy change that I would encourage is that all accusations of misbehavior should be judged by editors who have established credentials in the subject matter. I would also make it a rule that all editors who are granted extra authority have to reveal their identify and their areas of expertise. We can't have secret kangaroo courts where you get accused and judged within minutes by people who hide behind secrecy and anonymity."

Your proposed policy changes make sense to me for academic topics, although they may not work for non-academic topics. Why don't you propose the above policy change for academic topics? Or perhaps specifically for the topics that interest you the most.

Here is a webpage that describes how policies are developed:

https://en.wikipedia.org/wiki/Wikipedia:Policies_and_guidelines

It seems to me that you will have a good chance of having your new policies approved, if you can gather some support from the community. For example, you might write a couple of paragraphs outlining your proposed policy changes and then contact your academic colleagues and ask them to put their names on your document as a show of support for your new policies.

John Harshman said...

Nothing to say here. Just seeing how the new comment system deals with things. Is it now entirely linear?

Athel Cornish-Bowden said...

I have modified the non-coding DNA article (reinstating the ridiculous Scientific American article) in a way that I hope will satisfy the bosses while at the same time sticking with the truth. Maybe it'll end up pleasing no one.

Jonathan Badger said...

The problem with Wikipedia is related to Carl Sagan's comment about cranks: "Yes, they laughed at Galileo, but they also laughed at Bozo the Clown". Basically, people with non-orthodox, yet correct viewpoints are indistinguishable from cranks. If you don't want to give every crank a soapbox on Wikipedia (which is the problem in most of the rest of the Internet), the admins have to be overly skeptical of people claiming that the consensus opinion is wrong. Sometimes this impedes progress, but it is infinitely better than not being this skeptical.

drs. Tudor Georgescu said...

"How can you tell which sources are reliable unless you know something about the subject?" Well, that has to be shown objectively. There are objective criteria for what counts as a reliable source. Hint: Wikipedia prefers systematic reviews published in reputable scientific journals.

The irony is that I had already quoted as RS your article at your faculty website.