I'm amazed at the general support among comments here so far for the 'tertiary source' argument. Encyclopedias have always been written by experts -- usually o

I'm amazed at the general support among comments here so far for the 'tertiary source' argument. Encyclopedias have always been written by experts -- usually one expert per article. Wikipedia itself has a template, 'This article is in need of input from an expert on the subject'. But this isn't even right, because Wikipedia is a not an encyclopedia. It's like Johnson & Johnson saying Q-tips are for applying makeup or detailing cars and shouldn't be put in the ear. 99% of people who buy Q-tips put them in their ears. 99% of true claims on Wikipedia are contributed by experts and are either unreferenced, cite a source that doesn't really support them, a source at the other end of a broken link that nobody's read, or a source that doesn't meet the guidelines this historian was held to. Career editors spend more time checking for citations than checking citations. And they spend more time checking for citations when they personally don't believe a claim. The policies are applied hypocritically, with the result that opinions of non-experts outweigh the opinions of experts. The only thing limiting the damage has been the relatively small number of career editors. With massive decline in casual participation in recent years, this balance is starting to shift. I just noticed recently that career editors have started to tag mathematics articles. We should be honest that the encyclopedia contrivance is really just a way to avoid flame wars, and is an imperfect one, especially when editors 'merge with their cover story' and blindly enforce it. Wikipedia will be its best when it is recognized for what it is: a truth engine, a first source, and a very important public good. Editing Wikipedia needs to be about more than writing long policy documents and bludgeoning contributors with them. It is an important form of scholarship. The project could benefit greatly by a reputation model more subtle than "barnstars", such as [1]. But a fancy reputation model alone isn't enough. The culture of Wikipedia is in trouble and needs to be revitalized. [1] http://www2007.org/papers/paper692.pdf http://www2007.org/papers/paper692.pdf