4 ms·

> For anything important, the dialectic (i.e., edit-warring) produces something that at least covers everything, even if the resulting article has awful prose.

by glenra 12y ago

> For anything important, the dialectic (i.e., edit-warring) produces something that at least covers everything, even if the resulting article has awful prose. That's not really true unless you include the talk page. On any sufficiently contentious topic where there are two or more sides, what happens quite often is that one faction has more persistent editors successfully adding slant and sources in their direction and keeping out slant and sources in the other direction. You have to look at the talk page to see what facts and points of view are being actively omitted from the main page. A true NPOV is hard to maintain, because people often tend to think their own point of view is "fact" and the other side is "fringe". People also seem to fear that to allow a point of view to stand and be explained legitimizes it. I've noticed this phenomenon most in articles involving climate or medicine.

davidgerard 12y ago
Wikipedia's epistemology is shitty, shallow and brittle and it's trivial to find cases where it breaks. However, it's observably true that it's almost always enough almost all of the time, and certainly enough to have won utterly and not have even many readers complaining.
- glenra 12y ago
  Hmm, I just did a spot-check and medicine (including alternative medicine and supplements) seems to be getting better - the articles I recalled being terrible have improved. Climate is still pretty broken - the article on "climategate" is exhibit A - but maybe that'll come around eventually too. One can hope.
  - davidgerard 12y ago
    WikiProject Medicine has come down hard on anything even tangentially medicine-related, including alternative medicine practices. (Although wikiprojects don't "own" content areas, obviously sensible application of the content rules in a manner anyone not stupid would support is likely to pass.)
    - glenra 12y ago
      Right, that might have been the problem. The thing I had noticed as an issue is that if you walk down the "supplements" aisle at Whole Foods or whatever, you'll find hundreds of pill bottles with cryptic labels (eg, "Tonalin CLA" or "Garcinia Cambogia") with recommended dosage info ("take twice a day before meals!"), and no info whatsoever explaining what the product is for. If it's on the shelf, thousands of people must think this product is good for something but the FDA won't let the manufacturer say what. So you might think you could type the product name into wikipedia and see what's up. When I do that, what I want to know first of all is: (1) What is this product for? Why do people take it, what benefit do they expect to obtain? (2) What evidence supports the claims that advocates make for this product? What's the backstory behind how this product came into the public eye? (3) Last but not least, what does mainstream medicine have to say about it? When you actually look up these products in wikipedia, most of what you get is (3). (2) is nonexistent and (1) is at best given short shrift. I'm sure the editors who want to keep any claims related to what they see as "quack" products out of wikipedia are well-meaning but I think they are misguided and thereby do more harm than good. It'd be better if wikipedia pages reliably told both sides. Then supplement enthusiasts would learn to go to wikipedia first where they might see the medical caveats alongside the info they originally seek. But if wikipedia only tells one side, the nuts/enthusiasts just learn to avoid wikipedia and instead seek out alternative sources that DO include info of type (1) and (2), even at the cost of omitting (3).
      - davidgerard 12y ago
        Sounds like a good idea to me. If you think third-party sources for this stuff are findable, then https://en.wikipedia.org/wiki/Wikipedia_talk:MED https://en.wikipedia.org/wiki/Wikipedia_talk:MED might be a place to suggest it, with negative and positive examples (something that would inspire volunteers to take on your idea).
        glenra 12y ago
        I'm positive third-party sources are findable to answer at a minimum "what is this supplement supposed to do and who is supposed to take it and why do people think it's useful". I'd expect to find that stuff best laid out in books like this: http://www.amazon.com/NutriSearch-Comparative-Nutritional-Supplements-Professional/dp/097325386X/ http://www.amazon.com/NutriSearch-Comparative-Nutritional-Su... Thanks for the link suggestion - I might go ahead and do that.

by glenra 12y ago

davidgerard 12y ago

Wikipedia's epistemology is shitty, shallow and brittle and it's trivial to find cases where it breaks. However, it's observably true that it's almost always enough almost all of the time, and certainly enough to have won utterly and not have even many readers complaining.

glenra 12y ago
Hmm, I just did a spot-check and medicine (including alternative medicine and supplements) seems to be getting better - the articles I recalled being terrible have improved. Climate is still pretty broken - the article on "climategate" is exhibit A - but maybe that'll come around eventually too. One can hope.
- davidgerard 12y ago
  WikiProject Medicine has come down hard on anything even tangentially medicine-related, including alternative medicine practices. (Although wikiprojects don't "own" content areas, obviously sensible application of the content rules in a manner anyone not stupid would support is likely to pass.)
  - glenra 12y ago
    Right, that might have been the problem. The thing I had noticed as an issue is that if you walk down the "supplements" aisle at Whole Foods or whatever, you'll find hundreds of pill bottles with cryptic labels (eg, "Tonalin CLA" or "Garcinia Cambogia") with recommended dosage info ("take twice a day before meals!"), and no info whatsoever explaining what the product is for. If it's on the shelf, thousands of people must think this product is good for something but the FDA won't let the manufacturer say what. So you might think you could type the product name into wikipedia and see what's up. When I do that, what I want to know first of all is: (1) What is this product for? Why do people take it, what benefit do they expect to obtain? (2) What evidence supports the claims that advocates make for this product? What's the backstory behind how this product came into the public eye? (3) Last but not least, what does mainstream medicine have to say about it? When you actually look up these products in wikipedia, most of what you get is (3). (2) is nonexistent and (1) is at best given short shrift. I'm sure the editors who want to keep any claims related to what they see as "quack" products out of wikipedia are well-meaning but I think they are misguided and thereby do more harm than good. It'd be better if wikipedia pages reliably told both sides. Then supplement enthusiasts would learn to go to wikipedia first where they might see the medical caveats alongside the info they originally seek. But if wikipedia only tells one side, the nuts/enthusiasts just learn to avoid wikipedia and instead seek out alternative sources that DO include info of type (1) and (2), even at the cost of omitting (3).
    - davidgerard 12y ago
      Sounds like a good idea to me. If you think third-party sources for this stuff are findable, then https://en.wikipedia.org/wiki/Wikipedia_talk:MED https://en.wikipedia.org/wiki/Wikipedia_talk:MED might be a place to suggest it, with negative and positive examples (something that would inspire volunteers to take on your idea).
      - glenra 12y ago
        I'm positive third-party sources are findable to answer at a minimum "what is this supplement supposed to do and who is supposed to take it and why do people think it's useful". I'd expect to find that stuff best laid out in books like this: http://www.amazon.com/NutriSearch-Comparative-Nutritional-Supplements-Professional/dp/097325386X/ http://www.amazon.com/NutriSearch-Comparative-Nutritional-Su... Thanks for the link suggestion - I might go ahead and do that.