User:Berchanhimez/Laundering

Wikipedia has standards which define what is and is not a reliable source for use to satisfy the verifiability pillar. There are many sources which by definition do not meet this standard - examples may include "preprinted" articles which have not yet been reviewed by editorial boards/peer review, sources which are not subjected to fact-checking or which do not have a reputation for correcting errors in reporting, or personal opinions/statements made on social media (used outside WP:PRIMARY).

Sometimes, these "unreliable" sources are the original source for facts which are later reported on, but not verified or fact-checked by reputable sources. Editors may see these facts reported in a reliable source, but the mere reporting of facts does not satisfy the standards of a "reliable source". The use of a citation to an otherwise reliable source to cite unreliable facts is termed "source laundering", and is detrimental to the verifiability pillar of Wikipedia.

What is source laundering?
Source laundering, at its most basic, refers to the use of a typically "reliable" source to cite information that ultimately is unreliable. This is because many organizations, including reputable news organizations, often reprint or republish information from unreliable sources, without applying their normal standards of fact checking or verification to those reprints. Furthermore, as these are commonly considered "third-party contributions", the reliable sources may not entertain retraction requests for that material - as they rightfully consider they are merely the "publisher" and not responsible for the original content itself.

This can be seen when a person "tweets" something and their tweet is then embedded in a news source as an "example" of something an article is written on. When a news source does this, they are by definition "attributing" their source to that tweet - and unless the article gives an indication they have independently fact-checked the information, it should be treated as the original tweet would be treated. The act of embedding a tweet into a website or the article is in fact that website's way of saying "this isn't our content, we can't be held responsible for it's accuracy, etc" - the same as a disclaimer on an "opinion" section which says "these views and information are the contributor's own and have not been verified" or similar. By definition, if the information is not written in the "news's voice", it's not subjected to the fact checking of the organization unless otherwise proven on a case by case basis. What is not source laundering is the use of official tweets from a newsroom - even if those tweets are cited to their news website - because those tweets have come through the editorial process at the news organization.

This is also true when news organizations report on "preprints" of articles - they are not somehow making the information reliable simply by republishing it before it is peer reviewed. This can happen a lot when reviewing systematic reviews, guidelines, and other sources. Because this type of resource is only updated after months, if not years, there is a significant number of these typically medically reliable sources which may include information from preprints, or other non-reliable medical sources. When an article cites a preprint for information, they are not the peer reviewers who are conducting the official "approval" of that article - thus the preprinted article does not suddenly become "peer reviewed" because it was cited in another paper. Many systematic reviews/guidelines/etc actually explicitly call out information which comes from preprints as less reliable or less reviewed than other information - thus that information cannot suddenly become "reliably sourced" by citing it to a source which gives that disclaimer on their citation.

"Breaking-news reports often contain serious inaccuracies. As an electronic publication, Wikipedia can and should be up to date, but Wikipedia is not a newspaper and it does not need to go into all details of a current event in real time. It is better to wait a day or two after an event before adding details to the encyclopedia, than to help spread potentially false rumors. This gives journalists time to collect more information and verify claims, and for investigative authorities to make official announcements. (emphasis added)"

"All breaking-news stories, without exception, are primary sources, and must be treated with caution. (emphasis added)"

Why source laundering hurts Wikipedia
By citing a source such as the New York Times for information that is ultimately sourced to a tweet, preprint, or other unreliable source, it makes it much less likely that editors will be tempted to verify the information themselves. Many editors who see a citation to a reputable source such as the NYT will simply gloss over the source, or click on it simply to verify it exists and says what it claims to. Editors may not realize that the source is simply a "reprint" of a non-reputable source, thus taking it as a reliable source when in fact the information cannot be considered reliable by a logical analysis. It also obfuscates the original source of the information behind multiple links - citing the reprint instead of the original source means editors must "hunt" for the original source to verify the information.

Source laundering also is unnecessary - topics that are truthful and notable will by definition be covered by a news organization through their normal reporting at some point. A reporter tweeting out their "breaking news" is not necessary nor useful to include as a source/information in an article here - not to mention it's discouraged from using "breaking news" in the first place. A news organization which hosts a "live blog of reporters" or even quotes/embeds a tweet from one of their reporters in one of their articles cannot be assumed to have fact checked/verified that source - just as they are not held to that standard for regular "blogs" or "contributed opinions" such as an Op.Ed section.

Further, there is no deadline to have an article "perfect" - but there is a deadline to have an article be verifiable and at a minimum not incorrect, especially with regards to biographies of living persons. If the only reporting/source for it is a preprint/tweet, one must pause and ask themselves Why? - why the information, even if it comes from a reporter at a reliable news source or from a scientist who is normally well known for accuracy, has not yet been "written up" or "reviewed". This is to say, why is it that there isn't an actual news article to cite, and only the tweet - or why the article hasn't been reviewed yet. The obvious answer is time - it takes time to write a news article or peer review an article for a journal - but the slowness of the process does not mean we should bypass it by allowing the original source to become "laundered".

"Some newspapers, magazines, and other news organizations host online columns they call blogs. These may be acceptable sources if the writers are professionals, but use them with caution because blogs may not be subject to the news organization's normal fact-checking process. If a news organization publishes an opinion piece in a blog, attribute the statement to the writer, e.g. "Jane Smith wrote ..." (emphasis added)"

Suggestions

 * Do not try to use a resource which you have not read/looked at in its entirety.
 * Evaluate resources with a critical eye of where the information is coming from - just as we would expect an editor to realize when information is coming from an "interview" (and is an attributed opinion for which in text attribution is required), for example.
 * Be especially cautious for sources of known "preprint status" such as a live blog, contributor's tweet embedded in an otherwise well written article, etc.
 * Do not simply cite the information from wherever you find it - follow links, embeds, etc. back to their original source and cite that. This is especially true for tweets - simply being embedded in another website does not make the tweet more reliable.
 * When in doubt, clearly attribute the source of the information - at a minimum in the citation itself (ex: "An embedded tweet from:) but if possible and warranted in the prose itself (ex: John Smith reported in a tweet that ...)
 * Do not be overly hasty in attempting to add new/breaking information to an article. Wait for the information to be confirmed in fact-checked portions of reliable sources. This is especially true for BLPs.

Examples

 * - an example of a "fact" from a Wikipedia page with dubious citation in the first place being re-reported in "reliable sources" which clearly got the number from Wikipedia - then using those "new" citations as reliable sources to continue reporting the unreliable information.