I don't consult Wikipedia nearly as often as you do, Dan, but I have often found it to be a good source for historical and technical information. However, I've also learned that it's a likely biased and unreliable source for anything political/contentious, which includes "man-made" climate change.
I've contributed a few small corrections myself, the largest one (in 2018) being in an entry on greenhouses. That article erroneously conflated the greenhouse gas theory with the real-world functioning of greenhouses, and I edited the page to clarify that the real greenhouse effect relies on blocking air movement with some kind of film. This page, as I suspect is the case with all connected in any way with a contentious topic, was apparently closely monitored, because there was a quick response completely removing the corrected sentence. I note that the "Climate change" entry is completely gung-ho climate alarmist fare and unworthy of being called an encyclopedia entry. Wikipedia is a mixed bag and most trustworthy only for uncontentious topics. As with all published information, it deserves some skepticism, especially in the age of activists eager to tailor it to support an agenda.
I was going to make the same observation about politicized topics, using the same example. Another good example of Wikipedia spewing conventional biases is the topic of domestic violence.
An interesting topic. I have no Wikipedia moderating experiences, however, in general it strikes me the longer curatorial nature of Wikipedia entries does not lend itself to the short, sharp, emotionally charged, opinionated exclamations which dominate the divisive environment of other social media platforms. In other words, Wikipedia attracts different motives and personalities or aspects of personality. Furthermore, it is curated which means you don't take exclusive ownership of what you write, which is not likely very satisfying for expressing punctuated narcissistic impulses. The centre of gravity may be more the moderated middle rather than extremist positions unlikely to survive a predictable rebuttal.
If people think a problem with Wikipedia is its alleged "left-wing" bias, then perhaps they have some biases of thier own to reflect on. Oh, but they'll comment all the same...
I trust Wikipedia to give me a correct one-pass algorithm for computing variance or to tell me who the Minoans were. So on uncontroversial technical or historical issues, it's great.
I do not trust Wikipedia on any issue that is political or contentious. On such issues, it has a consistent leftist bias, often omits salient facts, and omits, distorts, or smears alternative credible viewpoints. Those who have tried to correct these inaccuracies report that their corrections are quickly removed by a motivated cadre of activist editors. Perhaps in your book with Jimmy Wales, you can discuss possible ways to adjust the Wikipedia model to address this problem.
There is considerable academic research on the subject. What it finds, pretty consistently, is that Wikipedia articles on contentious subjects are MORE accurate than those on less contentious matters -- for the simple reason that such articles attract more attention and discussion. Of course I don't claim that is some universal, invariable principle. But the pattern is clear. And what those claiming bias never consider is that their own perception of bias may itself be biased -- that is, they assume their perceptions of what is correct and fair are objective correct and assume that their failure to make it through the editing process shows others are biased. Which is just what a psychologist would predict. (See my own writing on naive realism.) Again, I'm not saying that's always the case but it is a plausible hypothesis which is seldom/never considered by those claiming bias.
Dan, this scholarly academic research you refer to is produced by a class of people who have been demonstrated to be highly politically biased themselves. Universities these days are chock full of committed progressives - in sociology and psychology it is close to 100% Democrat in the USA, for example. Many of them are nothing more than activists; it is standard operating procedure now to use progressive litmus tests (such as a commitment to DEI) for hiring purposes to perpetuate their own world view. The irony is that if your hypothesis is correct that one's prior commitments tend to bias one's *perception* of bias, then *of course* the overwhelming consensus among academic researchers will be that Wikipedia's very real progressive biases don't exist!
I'm afraid there is no shortcut to assessing bias. You cannot farm the task out to someone else. Read primary sources and form your own judgment.
You can simply assume that anyone with a PhD is hopelessly biased and all those studies are rubbish. Or you can read the studies, judge their methodologies, and decide if, just maybe, they actually have some value. One approach has the advantage of being a lot quicker, but I'll take the other, thanks.
I have a DPhil (Oxon), and an LL.B., so my thoughts are at least as rubbishy as the next man's. You might have missed the deep irony, but you come around to my concluding paragraph in the end, Dan.
Maybe you missed all those studies in the past 10 years showing that modern academic peer review - even in the medical sciences - is a pretty feeble screen.
No, my comment was deeper than that. What I was advising your readers to do is NOT take your word for what the academic studies show, but rather to read the studies and assess them using their own judgment. Because if your hypothesis is correct, chances are the studies contain biases, and your selection of studies may also contain biases. "There is no shortcut to assessing biases."
Can you provide links to the academic research allegedly proving that "Wikipedia articles on contentious subjects are MORE accurate than those on less contentious matters." As a scientist myself, I have a pretty well honed B.S. filter, and this claim sets it off. In particular, I'd like to know how these researchers measure "accuracy" on a issue that is under dispute.
Google Scholar: Wikipedia, accuracy, contentious. How this happens is no mystery. In particular, contention results in people saying, "OK, if you're going to include that, it should also include this." So the articles get longer and more detailed, which is possible thanks to Wikipedia being digital and effectively having unlimited space.
I don't want to make a career out of this thread, but the first page of results on Google Scholar does not give obvious support to your claim.
From the top link (https://doi.org/10.1017/S1049096511000199): "In this article, I review thousands of Wikipedia articles about candidates, elections, and officeholders to assess both the accuracy and the thoroughness of Wikipedia's coverage. I find that Wikipedia is almost always accurate when a relevant article exists, but errors of omission are extremely frequent. These errors of omission follow a predictable pattern." (Article is paywalled, so I can't see what the pattern is; in any case, articles on political candidates were not what I had in mind as covering controversial issues.)
Other articles cover measures of controversy in Wikipedia articles, which in itself doesn't address the issue of bias.
The only other article on page 1 addressing bias is https://doi.org/10.1177/0894439317715434: "Systematically biased editing, persistently maintained, can occur on Wikipedia while nominally following guidelines. Techniques for biasing an entry include deleting positive material, adding negative material, using a one-sided selection of sources, and exaggerating the significance of particular topics. To maintain bias in an entry in the face of resistance, key techniques are reverting edits, selectively invoking Wikipedia rules, and overruling resistant editors. Options for dealing with sustained biased editing include making complaints, mobilizing counterediting, and exposing the bias. To illustrate these techniques and responses, the rewriting of my own Wikipedia entry serves as a case study. It is worthwhile becoming aware of persistent bias and developing ways to counter it in order for Wikipedia to move closer to its goal of providing accurate and balanced information." (Again paywalled, so that's all I can get.)
Friend, there are hundreds of papers -- maybe thousands -- about Wikipedia and related issues. As usual, quality varies wildly. I don't have time to muck through my files and share my research but I assure you there are plenty of quality papers that concluded what I said. They're the reason I said what I said! As to bias... OF COURSE there's bias! It's a human creation. There's no such thing as an unbiased human creation. What matters is the nature of the bias, its extent relative to other similar creations, and whether the bias increases or decreases under the conditions of interest: Your point spoke to that last point, and the research I've seen concluded the dynamic was the reverse of what you assume. Maybe it's all wrong. But there it is.
IOW, "Take what I said on faith because I don't have the time to give you any actual evidence." Sorry, but I will stick with the evidence available to me.
If writing something can make you money, you'll write it and just not care about whether writing it makes people think badly of you. You're feeding your family, screw them.
Writing in the wikipedia can't make you money; I've never even heard of it getting somebody a job. You have to contribute for your own satisfaction, and the esteem of others.
It tells us that Manu Saadia's "Trekonomics" would probably work - the Star Trek notion that we could actually get along without money, if we didn't need to ration even essential resources like food and shelter.
Just take money out of the system; anything of general, fungible value corrupts all actors to some extent. Root of all evil!
Entrepreneurs are genius at privatizing benefits so they can profit by making and selling them. The number and scope of genuine "public goods" is diminishing all the time. Lighthouses, roads, telecommunications, information are easy to privatize with modern technology. The profit motive is not "toxic," it is what drives down prices and drives up quality. Technically speaking, profits are only earned when someone figures out a way to supply a good at a lower-than-market price - or supply a better quality good for the same price.
Disagree. The profit motive also drives cheating, lying about your product, and "holding down prices" by abusing Ned Ludd and all his class of workers. This is not a new argument so I'll leave it at that, and about 1000 books on the subject.
You realize that cheating and lying are prohibited by the free market system, right? When they aren't rectified, blame it on the legal system, not the market system.
No entrepreneur has the ability to set the price of labour; they are price-takers in labour - except in professional sports where scarcity of extreme talent makes labour the price-setter. Entrepreneurs in fact bid up the price of labour by finding ways to increase labour productivity, mostly through capital expenditures. Unions do not improve labour productivity; they are a tax on labour income - except where they are able to control a sector (such as public education). Unions are cartels in restraint of trade, and should be illegal.
There are millions of books written in economics. Your selection seems to be out of date. I'll leave it at that.
While accepting that unions certainly can be a net negative, I find the certainty and sweep of your conclusions quite astonishing. Are you familiar with labour history in the late-19th and early 20th centuries?
The first thing I thought of when reading this was Bruce Schneier's book "Liars and Outliers" which is about trust, albeit in a largely different context. I've also read several books on the topic of Leadership that too arrive at the conclusion that trust is paramount in a successful organization (however that's measured). Leaders have to trust those they lead, those being led have to trust their leaders. Within that context, it is reasonable to suggest that at least some of the current societal angst arises from lack of trust, especially in many of our "leaders" in business and politics.
When I mentioned your latest project to a pal of mine he immediately thought of the similarities to open source software. That is both have a high degree of trust that has steadily increased and both have similar methods to maintain their reliability.
I taught biology students so I never put restrictions on the use of Wikipedia for researching class assignments. But many of my colleagues in the social sciences either banned or restricted use of Wikipedia. The main defect with Wikipedia today remains socio-political biases. For example, you are more likely to find pejorative terms applied to right-leaning persons than left-leaning ones. But in the domain of biology and related sciences Wikipedia does a respectable job.
Since you're in touch with Mr. Wales, you could pass along the recommendation that article pages like "Greenhouse effect" include the note shown on the Talk page:
"This page is related to climate change, which has been designated as a contentious topic."
That would go a long way to helping readers distinguish such articles from uncontentious, and arguably more trustworthy, articles and improve the overall trustworthiness of Wikipedia itself.
Except that, contrary to what you seem to be assuming, there is research that shows it is the uncontentious articles that are of less quality. Why? Because they're uncontentious they don't draw lots of eyeballs and comments and debate. And it is precisely eyeballs and comments and debate that improve article quality.
That sort of decision would have to be made by the Wikipedia community, rather than by Jimmy Wales himself. You can raise it at the Village Pump. That said, I wouldn't expect it to go anywhere: Wikipedians know that Wikipedia has limitations (hence the standard warning-box on, for example, articles relating to developing events), but they expect those limitations to be localized and temporary. They wouldn't accept a blanket declaration (or even implication) that all articles on an important subject are untrustworthy; the expectation is that if articles are unreliable then they should be fixed via discussion, consensus, and applying standard Wikipedia principles and policies. (And in my experience that expectation is usually borne out, though I have to admit that my experience is mostly out-of-date: it's been at least fifteen years since I really spent a lot of time editing Wikipedia.)
And that's why Wikipedia can't be as trustworthy as we might hope -- or as Gardner might believe. A lot has changed in fifteen years, and I've seen a rise in determined bias promoting specific views.
David, you may be right. But there's an alternative hypothesis that may also be right: You are treating your own perceptions of reality as correct and deeming deviations from those perceptions to be bias on the part of others. Again, that may be true. Or it may be that your views are themselves biased and the deviations may, in fact, be less biased and closer to reality.
The best way to illustrate this is the phenomenon of "hostile media bias" -- a well-researched psychological feature in which highly committed partisans perceive far more bias in news reports than others do. And the bias they perceive is always "hostile," ie contrary to their views. Hence, left-wing partisans see a news article as having a right-wing bias while right-wing partisans see it as having a left-wing bias. The same article.
The problem with your hypothesis, Dan, is that articles on contentious political subjects - climate change, domestic violence - do not tend to treat the subject AS contentious. They give short shrift to one side or the other. This observation can be made by anyone, regardless of their own pre-commitments.
What I'm referring to is the phenomenon of strong control of article content that eliminates valid uncertainty or mixes inappropriate opinion with facts. Climate change is a glaring example, but articles on politicians or other notable people or topics can suffer from the same weakness. I've learned to not treat Wikipedia as an unbiased source for such information but as just the quickest entry point into such topics. Biased page editors are the issue, but I doubt that can be eliminated.
Good luck with this book re: Wikipedia. I have no stories to share but will surely amplify it on my socials when it is published.
I don't consult Wikipedia nearly as often as you do, Dan, but I have often found it to be a good source for historical and technical information. However, I've also learned that it's a likely biased and unreliable source for anything political/contentious, which includes "man-made" climate change.
I've contributed a few small corrections myself, the largest one (in 2018) being in an entry on greenhouses. That article erroneously conflated the greenhouse gas theory with the real-world functioning of greenhouses, and I edited the page to clarify that the real greenhouse effect relies on blocking air movement with some kind of film. This page, as I suspect is the case with all connected in any way with a contentious topic, was apparently closely monitored, because there was a quick response completely removing the corrected sentence. I note that the "Climate change" entry is completely gung-ho climate alarmist fare and unworthy of being called an encyclopedia entry. Wikipedia is a mixed bag and most trustworthy only for uncontentious topics. As with all published information, it deserves some skepticism, especially in the age of activists eager to tailor it to support an agenda.
I was going to make the same observation about politicized topics, using the same example. Another good example of Wikipedia spewing conventional biases is the topic of domestic violence.
An interesting topic. I have no Wikipedia moderating experiences, however, in general it strikes me the longer curatorial nature of Wikipedia entries does not lend itself to the short, sharp, emotionally charged, opinionated exclamations which dominate the divisive environment of other social media platforms. In other words, Wikipedia attracts different motives and personalities or aspects of personality. Furthermore, it is curated which means you don't take exclusive ownership of what you write, which is not likely very satisfying for expressing punctuated narcissistic impulses. The centre of gravity may be more the moderated middle rather than extremist positions unlikely to survive a predictable rebuttal.
If people think a problem with Wikipedia is its alleged "left-wing" bias, then perhaps they have some biases of thier own to reflect on. Oh, but they'll comment all the same...
I trust Wikipedia to give me a correct one-pass algorithm for computing variance or to tell me who the Minoans were. So on uncontroversial technical or historical issues, it's great.
I do not trust Wikipedia on any issue that is political or contentious. On such issues, it has a consistent leftist bias, often omits salient facts, and omits, distorts, or smears alternative credible viewpoints. Those who have tried to correct these inaccuracies report that their corrections are quickly removed by a motivated cadre of activist editors. Perhaps in your book with Jimmy Wales, you can discuss possible ways to adjust the Wikipedia model to address this problem.
There is considerable academic research on the subject. What it finds, pretty consistently, is that Wikipedia articles on contentious subjects are MORE accurate than those on less contentious matters -- for the simple reason that such articles attract more attention and discussion. Of course I don't claim that is some universal, invariable principle. But the pattern is clear. And what those claiming bias never consider is that their own perception of bias may itself be biased -- that is, they assume their perceptions of what is correct and fair are objective correct and assume that their failure to make it through the editing process shows others are biased. Which is just what a psychologist would predict. (See my own writing on naive realism.) Again, I'm not saying that's always the case but it is a plausible hypothesis which is seldom/never considered by those claiming bias.
Dan, this scholarly academic research you refer to is produced by a class of people who have been demonstrated to be highly politically biased themselves. Universities these days are chock full of committed progressives - in sociology and psychology it is close to 100% Democrat in the USA, for example. Many of them are nothing more than activists; it is standard operating procedure now to use progressive litmus tests (such as a commitment to DEI) for hiring purposes to perpetuate their own world view. The irony is that if your hypothesis is correct that one's prior commitments tend to bias one's *perception* of bias, then *of course* the overwhelming consensus among academic researchers will be that Wikipedia's very real progressive biases don't exist!
I'm afraid there is no shortcut to assessing bias. You cannot farm the task out to someone else. Read primary sources and form your own judgment.
You can simply assume that anyone with a PhD is hopelessly biased and all those studies are rubbish. Or you can read the studies, judge their methodologies, and decide if, just maybe, they actually have some value. One approach has the advantage of being a lot quicker, but I'll take the other, thanks.
I have a DPhil (Oxon), and an LL.B., so my thoughts are at least as rubbishy as the next man's. You might have missed the deep irony, but you come around to my concluding paragraph in the end, Dan.
Maybe you missed all those studies in the past 10 years showing that modern academic peer review - even in the medical sciences - is a pretty feeble screen.
Wait. So all you meant was we should read studies we cite before citing them? That's rather less dramatic than you made it sound.
No, my comment was deeper than that. What I was advising your readers to do is NOT take your word for what the academic studies show, but rather to read the studies and assess them using their own judgment. Because if your hypothesis is correct, chances are the studies contain biases, and your selection of studies may also contain biases. "There is no shortcut to assessing biases."
Can you provide links to the academic research allegedly proving that "Wikipedia articles on contentious subjects are MORE accurate than those on less contentious matters." As a scientist myself, I have a pretty well honed B.S. filter, and this claim sets it off. In particular, I'd like to know how these researchers measure "accuracy" on a issue that is under dispute.
Google Scholar: Wikipedia, accuracy, contentious. How this happens is no mystery. In particular, contention results in people saying, "OK, if you're going to include that, it should also include this." So the articles get longer and more detailed, which is possible thanks to Wikipedia being digital and effectively having unlimited space.
I don't want to make a career out of this thread, but the first page of results on Google Scholar does not give obvious support to your claim.
From the top link (https://doi.org/10.1017/S1049096511000199): "In this article, I review thousands of Wikipedia articles about candidates, elections, and officeholders to assess both the accuracy and the thoroughness of Wikipedia's coverage. I find that Wikipedia is almost always accurate when a relevant article exists, but errors of omission are extremely frequent. These errors of omission follow a predictable pattern." (Article is paywalled, so I can't see what the pattern is; in any case, articles on political candidates were not what I had in mind as covering controversial issues.)
Other articles cover measures of controversy in Wikipedia articles, which in itself doesn't address the issue of bias.
The only other article on page 1 addressing bias is https://doi.org/10.1177/0894439317715434: "Systematically biased editing, persistently maintained, can occur on Wikipedia while nominally following guidelines. Techniques for biasing an entry include deleting positive material, adding negative material, using a one-sided selection of sources, and exaggerating the significance of particular topics. To maintain bias in an entry in the face of resistance, key techniques are reverting edits, selectively invoking Wikipedia rules, and overruling resistant editors. Options for dealing with sustained biased editing include making complaints, mobilizing counterediting, and exposing the bias. To illustrate these techniques and responses, the rewriting of my own Wikipedia entry serves as a case study. It is worthwhile becoming aware of persistent bias and developing ways to counter it in order for Wikipedia to move closer to its goal of providing accurate and balanced information." (Again paywalled, so that's all I can get.)
Friend, there are hundreds of papers -- maybe thousands -- about Wikipedia and related issues. As usual, quality varies wildly. I don't have time to muck through my files and share my research but I assure you there are plenty of quality papers that concluded what I said. They're the reason I said what I said! As to bias... OF COURSE there's bias! It's a human creation. There's no such thing as an unbiased human creation. What matters is the nature of the bias, its extent relative to other similar creations, and whether the bias increases or decreases under the conditions of interest: Your point spoke to that last point, and the research I've seen concluded the dynamic was the reverse of what you assume. Maybe it's all wrong. But there it is.
IOW, "Take what I said on faith because I don't have the time to give you any actual evidence." Sorry, but I will stick with the evidence available to me.
If writing something can make you money, you'll write it and just not care about whether writing it makes people think badly of you. You're feeding your family, screw them.
Writing in the wikipedia can't make you money; I've never even heard of it getting somebody a job. You have to contribute for your own satisfaction, and the esteem of others.
It tells us that Manu Saadia's "Trekonomics" would probably work - the Star Trek notion that we could actually get along without money, if we didn't need to ration even essential resources like food and shelter.
Just take money out of the system; anything of general, fungible value corrupts all actors to some extent. Root of all evil!
Money is necessary; it is not the root of all evil. The saying actually goes, "The love of money is the root of all evil."
Quite agree; I eye-roll at the term "late-stage capitalism" because we've had capitalism and money since pretty shells were hoarded.
Let me ditch the ol' cliche', and say "profit motive is toxic to public goods".
Entrepreneurs are genius at privatizing benefits so they can profit by making and selling them. The number and scope of genuine "public goods" is diminishing all the time. Lighthouses, roads, telecommunications, information are easy to privatize with modern technology. The profit motive is not "toxic," it is what drives down prices and drives up quality. Technically speaking, profits are only earned when someone figures out a way to supply a good at a lower-than-market price - or supply a better quality good for the same price.
Disagree. The profit motive also drives cheating, lying about your product, and "holding down prices" by abusing Ned Ludd and all his class of workers. This is not a new argument so I'll leave it at that, and about 1000 books on the subject.
You realize that cheating and lying are prohibited by the free market system, right? When they aren't rectified, blame it on the legal system, not the market system.
No entrepreneur has the ability to set the price of labour; they are price-takers in labour - except in professional sports where scarcity of extreme talent makes labour the price-setter. Entrepreneurs in fact bid up the price of labour by finding ways to increase labour productivity, mostly through capital expenditures. Unions do not improve labour productivity; they are a tax on labour income - except where they are able to control a sector (such as public education). Unions are cartels in restraint of trade, and should be illegal.
There are millions of books written in economics. Your selection seems to be out of date. I'll leave it at that.
While accepting that unions certainly can be a net negative, I find the certainty and sweep of your conclusions quite astonishing. Are you familiar with labour history in the late-19th and early 20th centuries?
The first thing I thought of when reading this was Bruce Schneier's book "Liars and Outliers" which is about trust, albeit in a largely different context. I've also read several books on the topic of Leadership that too arrive at the conclusion that trust is paramount in a successful organization (however that's measured). Leaders have to trust those they lead, those being led have to trust their leaders. Within that context, it is reasonable to suggest that at least some of the current societal angst arises from lack of trust, especially in many of our "leaders" in business and politics.
I look forward to reading your take on that.
When I mentioned your latest project to a pal of mine he immediately thought of the similarities to open source software. That is both have a high degree of trust that has steadily increased and both have similar methods to maintain their reliability.
Yes, very much. That will be a (small) part of the story. Jimmy actually worked in open source software before starting Wikipedia.
I taught biology students so I never put restrictions on the use of Wikipedia for researching class assignments. But many of my colleagues in the social sciences either banned or restricted use of Wikipedia. The main defect with Wikipedia today remains socio-political biases. For example, you are more likely to find pejorative terms applied to right-leaning persons than left-leaning ones. But in the domain of biology and related sciences Wikipedia does a respectable job.
It is very contentious these days to say that mammals have two sexes and individuals cannot change sex.
Since you're in touch with Mr. Wales, you could pass along the recommendation that article pages like "Greenhouse effect" include the note shown on the Talk page:
"This page is related to climate change, which has been designated as a contentious topic."
That would go a long way to helping readers distinguish such articles from uncontentious, and arguably more trustworthy, articles and improve the overall trustworthiness of Wikipedia itself.
Except that, contrary to what you seem to be assuming, there is research that shows it is the uncontentious articles that are of less quality. Why? Because they're uncontentious they don't draw lots of eyeballs and comments and debate. And it is precisely eyeballs and comments and debate that improve article quality.
That sort of decision would have to be made by the Wikipedia community, rather than by Jimmy Wales himself. You can raise it at the Village Pump. That said, I wouldn't expect it to go anywhere: Wikipedians know that Wikipedia has limitations (hence the standard warning-box on, for example, articles relating to developing events), but they expect those limitations to be localized and temporary. They wouldn't accept a blanket declaration (or even implication) that all articles on an important subject are untrustworthy; the expectation is that if articles are unreliable then they should be fixed via discussion, consensus, and applying standard Wikipedia principles and policies. (And in my experience that expectation is usually borne out, though I have to admit that my experience is mostly out-of-date: it's been at least fifteen years since I really spent a lot of time editing Wikipedia.)
And that's why Wikipedia can't be as trustworthy as we might hope -- or as Gardner might believe. A lot has changed in fifteen years, and I've seen a rise in determined bias promoting specific views.
David, you may be right. But there's an alternative hypothesis that may also be right: You are treating your own perceptions of reality as correct and deeming deviations from those perceptions to be bias on the part of others. Again, that may be true. Or it may be that your views are themselves biased and the deviations may, in fact, be less biased and closer to reality.
The best way to illustrate this is the phenomenon of "hostile media bias" -- a well-researched psychological feature in which highly committed partisans perceive far more bias in news reports than others do. And the bias they perceive is always "hostile," ie contrary to their views. Hence, left-wing partisans see a news article as having a right-wing bias while right-wing partisans see it as having a left-wing bias. The same article.
The problem with your hypothesis, Dan, is that articles on contentious political subjects - climate change, domestic violence - do not tend to treat the subject AS contentious. They give short shrift to one side or the other. This observation can be made by anyone, regardless of their own pre-commitments.
What I'm referring to is the phenomenon of strong control of article content that eliminates valid uncertainty or mixes inappropriate opinion with facts. Climate change is a glaring example, but articles on politicians or other notable people or topics can suffer from the same weakness. I've learned to not treat Wikipedia as an unbiased source for such information but as just the quickest entry point into such topics. Biased page editors are the issue, but I doubt that can be eliminated.
An article whose content relies on discussion and consensus will tend to be full of conventional rubbish.