My hobby is an obscure branch of thermodynamics that has meant I seem to have spent an unusual amount of time reading Wikipedia articles.
What I have found is that Wikipedia is amazingly accurate (contrary to what some would claim) but very badly written (something that isn’t often talked about as a principal flaw).
The accuracy of Wikipedia is usually challenged for topics which have a subjective aspect that people refuse to acknowledge (such as historical events, religion, social sciences), the stuff I have been looking at tends to be less subjective, since it is backed up by far more evidence and has a method for reaching consensus. Here the unusually high level of accuracy of Wikipedia is apparent.
To say that something is badly written is a more difficult thing to prove, however rather than try to do so, I’ll suggest an explanation as to why this would be a plausible outcome. The 1911 Encyclopedia Britannica is regarded as something of a classic, having more articles written by experts in their field than any other issue. In many ways it is the antithesis of Wikipedia.
Reading the 1911 Britannica, for equivalent entries about science and the results are obvious. Wikipedia is more verbose and less clear, and this is what I think is happening:
Wikipedia will always attract a certain percentage of people that the vanity reward of being published would not normally be offered to.
Less intelligent people in technical circles seem to like to make things appear more difficult and use a proprietary language or jargon (computer programmers are often terribly guilty of this) because the language of the craft is relatively hard to master and those that will never go beyond the craft to the art will protect their mastery of the craft.
Less obvious than the above is the fact that people who really know what they are talking about can invent new and better ways of describing things, whereas people who don’t will have to regurgitate accepted wisdom for fear of being wrong on the bits they are re-interpreting .
This last point is the principal reason that, for hard to understand scientific concepts, Wikipedia ironically really falls down, more than for supposed inaccuracy in contentious subjects.
Our writing is bloody awful, it’s true. Unfortunately, good writers are a damn sight rarer than good-enough researchers and hobbyists … but that said, writing is second to accuracy and content, I think. Though not third.