Please ban Wikipedia links

Question

"I actually do think Wikipedia is an amazing thing. It is the first place I go when I'm looking for knowledge. Or when I want to create some." -- Stephen Colbert.

The History Stack Exchange is supposed to be a high-brow place for people with an academic interest in history. As such, it's nonsense to allow answers based on low-quality non-peer reviewed sources like Wikipedia. The problems with Wikipedia are many and well-documented:

• It contains numerous inconsistencies, lies by omission, and downright falsehoods.
• Nothing on it is attributable to real authors. Writing "according to Joe Smith 'bla bla'" is materially different from writing "according to Wikipedia 'bla bla'" because Joe Smith is a real person and you can check their credentials.
• Most significantly: Wikipedia articles change. One article may have been a good source at one point in time, but many years later, after the article was rewritten, it may now be terrible.

If you as an answerer on History Stack Exchange consider linking to Wikipedia to get your links-to-text ratio up, pretty please don't do it. Instead, check what Wikipedia references, read that reference, and reference that. If you don't have access to that reference or this is too much work, well then, tough luck, maybe order that reference or don't write an answer. As a reader of History Stack Exchange, I very much prefer no references at all over links to Wikipedia because then I know the author is "winging it".

If an outright ban is not possible, then perhaps warn answer writers: "You are linking to Wikipedia which is not a reliable source. Here is why Wikipedia is not reliable: ..."

Sources:

• Wikipedia Is Good for You!?, James P. Purdy, 2010.
• Can History Be Open Source? ­ Wikipedia and the Future of the Past, Roy Rosenzweig, 2007.
• Why You Can't Cite Wikipedia in My Class, Neil L. Waters, 2007.

Answer 1

Normally I wait 24 hours for community response before I answer, but not in this case. I disagree with the premise of the question.

First, this question argues for a change without establishing that there is any evidence of a problem. The question would be stronger if supported by quantified evidence. What percentage of Wikipedia articles are inaccurate? What fraction are merely incomplete? What percentage are misinformation rather than legitimate disputes? Before I endorse any significant change to the culture of H:SE I'm going to want to understand that.

Second, the question is based on a false premise. Wikipedia has several quality control features that the question ignores. Well done Wikipedia pages have references. Poor quality Wikipedia pages have annotations that invite revisions. Revisions are subject to a feedback loop. I'm not sure how effective Wikipedia quality control is, but I'm comfortable that it is stronger than this question assumes. Ultimately though Wikipedia is a source and history is about assembling a narrative based on existing sources. The practice of history requires assembling a narrative that accounts for the most evidence and culls the least evidence. History is about understanding the flaws in the evidence. All sources (including Wikipedia) need to be examined for bias, for flaws, for omissions, and for context. If history were simply assembling sources like Legos, it wouldn't be very interesting, and it definitely wouldn't be about humanity. Different sources must be treated differently - Oral history is read differently than epigraphy, which is read differently than unreliable sources, which are read differently than references to unavailable primary sources. Banning a category of sources simply isn't the way I understand history to be practiced.

• Yes, Wikipedia contains many inconsistencies, omissions and falsehoods. Like any other source. The number of people participating in many pre-modern battles exceeds the population of the participating countries. IIRC, a similar statistic was generated for the US civil war based on monument inscriptions. We have records that state that Edward II of England died on several different dates and that he lived long after those dates in exile. We have multiple records with multiple conflicting explanations for most historical phenomenon (the Great Depression, for example).

• Many sources cannot be attributed to real authors. The books of the Christian Bible were probably not authored by the listed authors. The Federalist papers were not authored by a person named Publius.

• Yes, Wikipedia articles change; many modern, internet based sources change (that is a feature, not a bug; science is a process, not a product) a well cited article will indicate the date of the citation.

If the real problem is not Wikipedia, but upvotes on substandard answers, then banning Wikipedia won't solve the problem. The problem is voting. voting is a very rough indicator of quality. I don't think many will defend voting, since it was discovered that Fastest Gun In the Internet was a relatively sure-fire way to build rep. Voting is flawed. The flaw has nothing to do with Wikipedia, it has to do with human psychology, and I don't propose to change that.

Just because we're working in text, which can fail to convey nuance, I want to hasten to point out that I'm glad you asked the question - I disagree with the conclusions, but I appreciate the fact that you raised the question, and I hope that others will weigh in to support, or to provide new, alternative perspectives.