Wikipedia talk:No original research

Primary source VS secondary source

What if there is a contradiction between primary source and secondary source? Common logic says to prefer primary source but wiki policy seems to have no rule regarding this issue. Should we not have it? There may be small or big inaccuracies or differences between primary and secondary sources. --88.231.135.73 (talk) 18:35, 21 May 2018 (UTC)

I think we need more information. Are you talking about conflict between the actual text of the two sources (for example the primary saying “red” and the secondary saying “blue”)... or something else? Blueboar (talk) 19:02, 21 May 2018 (UTC)

Indeed this may be largely case dependent. In general I would say a literal, proven, record of a primary source (saying "red") is to be preferred over a secondary source claiming that the primary source reported "blue". Yet (to take the example a bit further) a primary colourblind source claiming something IS actually red should not be necessarily be used as superior to a secondary source who concludes based on analysis of all evidence that the object was in fact "blue". Arnoutf (talk) 18:31, 22 May 2018 (UTC)

Exactly; the primary source may be colorblind or even deliberately deceiving, for instance in the case of the primary source being someone / a company / a political party wanting to paint a rosy picture of themselves, and in that case the primary source should not be trusted and secondary sources likely be trusted more (if they are trustworthy, did their analysis of all evidence, etc.). --Jhertel (talk) 20:46, 22 May 2018 (UTC)

So I thought about this for a while, and I think the only time there's a clear-cut answer is when the secondary source is talking about the primary source. If a newspaper writes that Bob Johnson said so and so during his senate testimony, but watching the video of the testimony proves that so and so was never said by anyone, or was said by someone else, we ignore the newspaper. What is put in place of that statement, or whether it is removed entirely, would have to be decided on a case by case basis. And anything further from that type of scenario also, I think, doesn't have a simple rule you could apply. But even then, you have to be careful of whether the secondary source is referring to a different version of the primary source than you're looking at, or maybe is just being figurative. Someguy1221 (talk) 03:02, 31 May 2018 (UTC)

To take the question from a different angle: It's common for secondary sources to weigh what primary sources A, B, and C say, then provide conclusion X that is a synthesis/analysis, and this may conflict with, say, primary source C because C was wrong/outdated/speculative. So, yes, "this may be largely case-dependent". It can also be context-sensitive. Two examples: WP:ABOUTSELF is important: We trust what a person says about their beliefs, self-identity, etc. far more than we trust what a newspaper says that person thinks/feels. Second, when it comes to technical topics, secondary sources way below the level of a systematic review may be farcically incorrect in their understanding of what a journal paper says and means and even what the terminology used in it refers to. Newspaper and magazine journalists get scientific stuff flat wrong very frequently, and are also apt to leap for grossly exaggeratory assumptions about the implications of or claims made in a research paper. Not all secondary sources are equal, and the gulf widens the more technical the material is.  — SMcCandlish ☏ ¢ 😼  23:28, 23 June 2018 (UTC)

Nationality of television series

The current consensus of the TV wikiproject is that the nationality of a tv series is identified from the production companies listed in its credits, establishing the nationality of each from reliable sources, and determining sole or multi-nationality accordingly. There is a proposal to replace this with direct referencing of nationality from reliable sources. You are invited to participate in the discussion at Wikipedia talk:Manual of Style/Television#Proposed MoS change: Nationality MapReader (talk) 07:14, 12 June 2018 (UTC)

I am hoping that among watchers of this page there might be an experienced editor or two who is used to dealing with questions of synthesis, who might be able to comment on whether or not there is an issue here? MapReader (talk) 07:46, 20 June 2018 (UTC)

Primary Sources and COI

Wikipedia's guidelines says that primary sources and self citing hurts the integerty of the encyclopedia and should be limited. Have you ever done a poll to see what kind of encyclopedia people want? I think that instead of focusing on what the editors think (age sex and race distrubution of Wikipedia editors is very different from the general population, according to statistics) Wiki should focus on what rules READERS think would improve the encyclopedia. That would make Wikipedia a better and more reliable encyclopeia. ............ Wikipedia has been losing editors for quite a while now, according to statistics and it seems like many people are blaming it on some of the guidelines. If there ever was a poll asking readers what they think would make Wiki a better encyclopedia that I don't know about, please let me know.SpidersMilk, Drink Spider Milk, it tastes good. (talk) 17:04, 23 June 2018 (UTC)

All I know is that Wikipedia is one of the most popular sites on the Web, and many people I know turn to Wikipedia first when they want to find out about something, so we seem to be doing something right. (Note the warning that Wikipedia should never be taken as gospel, but only as a starting point in research.) If we were to drop those guidelines that you seem to dislike, Wikipedia would become a giant multi-editor blog, with constant arguments over whose version of the description of a topic was correct, and full of hoaxes, attacks, advertisements, articles about real or imagined subjects that no one but the author cares about, etc. I think that would result in Wikipedia withering away very quickly. If someone wants to write about something on the Web without having to adhere to those pesky guidelines, they can write a blog. I occasionally blog, and offer speculation and opinions in blog posts, but I make clear what is supported by reliable sources, and what is my personal observation, speculation and opinion. - Donald Albury 17:50, 23 June 2018 (UTC) (Edited to add to comment.) - Donald Albury 17:53, 23 June 2018 (UTC)

Worse, if we turned it into a content popularity contest and actually gave people what they [think they] want, then we would not have an encyclopedia at all. What people want (respond to) on a short-term basis on the Internet is mostly clickbait: scandals, exaggerated claims, nudity, explosion, cute pets, and highly-PoV socio-political ranting. Part of our "job" is providing a rational oasis in that desert, whether people understand that they need it or not.  — SMcCandlish ☏ ¢ 😼  18:32, 23 June 2018 (UTC)

I just personally think that a lot of information about certain subjects considered "fringe" is being held back, and that most people would have more faith in Wikipedia if more of these subjects were included in articles. For example, due to my intest in physics theories, I am learning a lot of info about Superfluid vacuum theory and Pilot wave theory and how these could possibly explain the EM drive. But maybe that is just me, and most others disagree with me. In any case, there needs to be a way to attract more non-hoaxing and non-vandalizing editors.SpidersMilk, Drink Spider Milk, it tastes good. (talk) 21:02, 23 June 2018 (UTC)

We do have articles on those subjects. We even have an article about the Dean drive, of which I have fond memories from my youth. The issue is how much weight do we give to accounts of theories that have not been covered in reliable sources. - Donald Albury 21:50, 23 June 2018 (UTC)

Yeah, this doesn't seem to be a valid complaint when it's all blue-linked.  — SMcCandlish ☏ ¢ 😼  23:22, 23 June 2018 (UTC)

Frequently misinterpreted sourcing policy

FYI

 – Pointer to relevant page elsewhere.

For your reading pleasure or displeasure: Wikipedia:Frequently misinterpreted sourcing policy.

Originated as a WP:Village pump (policy) post, now developed into an essay. Reception has been uniformly positive so far, though it's a bit of a mix of a list of issues and recommendations of what to do about them. I might split off the latter material to a userspace page at some point, especially if a key boldfaced item gets resolved.  — SMcCandlish ☏ ¢ 😼  03:00, 1 July 2018 (UTC)

Rfc on How we interpret "published sources that are directly related to the topic of the article"

[Please excuse the length ... the topic is complicated] This post concerns the lede sentence:

"To demonstrate that you are not adding OR, you must be able to cite reliable, published sources that are directly related to the topic of the article, and directly support the material being presented."

It arose in Gaza beach explosion (2006) - Talk - section "3. Article Improvement Discussions", with discussions regarding the addressing of events in Gaza prior to beach explosions which all but wiped out an extended Palestinian family and sent shockwaves through the media. As with many of the Israel-Palestine conflicts (perhaps even 'all') the climax event, from which the name of the article is usually derived, hardly ever occurs randomly out of the blue with no prior build-up. Israel does not invade Gaza without militants from the Strip first having fired rockets at the southern Negev. Militants rarely do that out of pure spite, but because Israel has done something to prejudice their lives, like the assassination of a senior official, a crippling blockade, the killing of civilians near the border zone, etc. And those in turn have their antecedents also; the murder of teenagers, etc. This is not a new issue, and one related question goes - "How far back does one go? To the apple in the garden?" A few years ago KingsIndian (talk · contribs), fully aware of this issue, made what was accepted at the time as a very reasonable suggestion. Why not go back to a level playing field when all parties agreed to a cessation of hostilities, or to some informal but significant period of mutual calm - what I think he called a "blank slate". In that way we can avoid having an article "XXX event" and a separate article "Origin of the XXX Event" by including the prequel development of tensions between a cessation of hostilities (e.g. as a result of the ceasefire/truce) and the next bellicose flurry.

The above applies to less than epic "events". The prelude to the Six Day War for instance, would be as long as the rest of the article, which would be cumbersome. A brief flurry of a few days or even weeks might warrant having the prelude included in the main article. [I have no idea how one reasonably chooses how to make that split - a future topic]

So to get to the essence: Many events in the prelude of, say, the Gaza War (2008–09), have sources which never contemplated the future climactic war - Operation Cast Lead (OCL). The source on Arafat's death, that for the details of the understanding of the June 2008 ceasefire terms; the breaking of the ceasefire on 4 November 2008; the subsequent hail of rockets on Israel, many of the authors of the provided sources had no idea that OCL was soon to take place. To apply the WP:OR policy that sources must be directly related to the topic (the 2008/09 war - OCL) therefore usually requires some degree of subjective interpretation ... dangerous!! Yet a sequence of reliable sources may show that one individual event triggered the next. HOWEVER, there are VERY few quality sources that stitch all of these events together, and even more rarely, do so in Wiki-acceptable detail. Alternatively there are almost no sources that deny causal chains. Instead, one may appeal to "the sky is blue" principle that if A led to B, and B led to C, then A and C are part of the same sequence or chain. Sadly this process is usually interrupted by lengthy edit wars.

My proposal is that in the seemingly never-ending series of bellicose events, the preludes - the history of tension building - is encylopedically relevant to the climactic straw, and should be welcomed. But were sources to be disallowed that did not directly link the individual events in the chain to the breaking of the camel's back, (according to the Wiki condition quoted at the start), then this valuable chain of causative information would be excluded from Wiki. I am too much of a n000b to dredge Wiki protocol for a solution, so I look to more experienced peers. Maybe the solution is as simple as changing the article name from "Gaza beach explosion (2006)" to "Gaza beach explosion (2006) and its origin". Erictheenquirer (talk) 15:19, 4 July 2018 (UTC)

Sorry, but that just is OR. There's no way around that. Wikipedia is not Truth, nor here to Right the Great Wrongs. This consequently, necessarily means that on current and even quasi-recent events, we do not have or provide all the answers. We have to wait for multiple, independent reliable sources to provide to us the scholarly consensus. That can take years. Even decades. And that off-site consensus can change, and our coverage changes to go along with it. WP's basic purpose isn't injecting perfect understanding of reality into people's brains, it's providing an overview, and sources that are a good starting point for their own further reading.  — SMcCandlish ☏ ¢ 😼  11:40, 5 July 2018 (UTC)

So, if a RS source like the Journal of Palestine Affairs (UCLA) published the sequence/chain of events, that would NOT be OR and would be RS? The essence is that an editor should not string the events together, ... do I interpret that correctly? Erictheenquirer (talk) 11:53, 5 July 2018 (UTC)