Well, personally I don't think the tweet should have been sent. I think it sends the wrong signals and needs to be more nuanced.
But, as a retired academic and active Wikipedia contributor (and reader), I am realistic about both academic publishing and contributing to Wikipedia. In general, I would agree that when writing academic papers, Wikipedia should not be the preferred citation for some fact. Ideally, the researcher would seek to obtain the material cited in the Wikipedia article, read it and cite it directly.
But the world is not ideal. What if the Wikipedia article doesn't contain a citation for that fact? Or there is a citation but the work cited is not accessible (a dead link, a book only in a foreign library, perhaps in another language, etc). A basic principle of citation is "cite it where YOU saw it". So if a Wikipedia article is the best source you consulted in relation to that fact, then you should cite Wikipedia.
OK, some publishers or teachers might not accept a work with a Wikipedia citation. There is nothing we can do about it, they control what it is published or what is acceptable for assessment. But if you are such a publisher or teacher, realise that the risk you run with this rule is that people might then cite the source cited by Wikipedia but without having read it for themselves. ("cite it where you DIDN'T see it"). It might look good to have all sources *appearing* to be reliable but it's dishonest and the reader can be misled. If Wikipedia was the source, then the readers deserves to know that and make their choice about whether they are comfortable accepting the fact on that basis.
However, having said of all that, if the fact involved was central to the argument in an academic paper intended for publication in a reputable place, I would be very uncomfortable relying on a Wikipedia as a citation for that fact. But there are often parts of academic papers where the citation is less crucial to the central point and citing Wikipedia might be acceptable, particularly if weasel words are used. "It has been suggested that velvet cake originated in the Victorian era [cite Wikipedia]".
It comes down to a decision of risk-management. The question we have to ask ourselves about any citation (including Wikipedia) is how much harm could be done as a consequence of that source being wrong? For a student writing a class assignment, nobody (apart from the teacher) is likely to ever read what it is written. Also they have a fairly narrow time window in which to do it, which limits their ability to obtain access to offline or deadlink works cited in the Wikipedia article about the fact. For the purposes of the class assignment, I see little risk in citing Wikipedia in such circumstances. I see great risk in other circumstances "Cancer can be cured by eating velvet cake [cite Wikipedia]" or "It has been suggested that cancer can be cured ..." going into a highly regarded medical journal.
I love Wikipedia, I read it and I write it, but I don't think anyone should make big decisions about their lives or the lives of others solely upon it.
This is why I am concerned about WikiData. It's easily queried, free to use, but we know very little about its coverage and completeness, and that makes it dangerous in the hands of those who don't understand that or who simply don't care. If people start using WikiData as a convenient source of data to draw some conclusion for some important purpose (and lazy political staffers come to mind as they often need to do things quickly and may not be interested in the truth but rather seeking "evidence" that will support a particular political view), I see serious consequences. As a simple example, we could do a query over Wikidata calculating the age at death (for anyone who has a birth and death date in WikiData) and use that to calculate average life expectancy and then use that in some political decision about not paying old age pensions until a later age or reducing health expenditure on children. Would such a Wikidata query really tell us anything about average life expectancy? Of course not, the people listed in Wikidata are not a random sample of the population. They are typically present for having some achievement (or notoriety), which suggests that they probably survived early childhood at least (it's hard to be a professional baseball player or be a serial killer as a toddler). So, all those people who died as babies are unlikely to make it into Wikidata and hence average life expectancy (based on Wikidata queries) will be higher than reality as infant mortality will be over-looked (as well as a lot of other factors that influence live expectancy).
If you are worried about Wikipedia as a source, I suggest you worry a lot more about Wikidata as a source.
Kerry
-----Original Message----- From: Libraries [mailto:libraries-bounces@lists.wikimedia.org] On Behalf Of Federico Leva (Nemo) Sent: Tuesday, 3 December 2019 10:53 PM To: Wikimedia & Libraries libraries@lists.wikimedia.org; Kathleen DeLaurenti kathleendelaurenti@gmail.com Subject: Re: [libraries] Concern about messaging about Wikipedia
Kathleen DeLaurenti, 26/09/19 17:55:
Clearly there are some nefarious potential uses here, but what's more concerning is that the WikiUK twitter account has come forward forcefully saying that Wikipedia shouldn't be cited in the literature. Period. https://twitter.com/wikimediauk/status/1177215917534711808
I see that some of those tweets were mentioned in a recent paper on the usage of Wikipedia in education. https://doi.org/10.11645/13.2.2669
Federico
_______________________________________________ Libraries mailing list Libraries@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/libraries