DrFleischman (talk | contribs) |
Margin1522 (talk | contribs) →Citing republished pieces; What source do we cite?: Copy editors |
||
Line 158: | Line 158: | ||
::As long as its reliable... One thing with wire-service stories is that newspapers typically edit them. So the versions of AP stories that you see in different papers will often be slightly different. So if you get it from a paper you should cite the paper. And set the agency= parameter to indicate that it was a wire-service story. – [[User:Margin1522|Margin1522]] ([[User talk:Margin1522|talk]]) 16:41, 12 March 2015 (UTC) |
::As long as its reliable... One thing with wire-service stories is that newspapers typically edit them. So the versions of AP stories that you see in different papers will often be slightly different. So if you get it from a paper you should cite the paper. And set the agency= parameter to indicate that it was a wire-service story. – [[User:Margin1522|Margin1522]] ([[User talk:Margin1522|talk]]) 16:41, 12 March 2015 (UTC) |
||
::::{{u|Margin1522}}, I'm no journalism expert but that's news to me. I always thought syndicated stories were identical in every newspaper, but I have no evidence to back that up; it was just an assumption. Can you point me to something suggesting otherwise? --[[User:DrFleischman|Dr. Fleischman]] ([[User talk:DrFleischman|talk]]) 17:33, 12 March 2015 (UTC) |
::::{{u|Margin1522}}, I'm no journalism expert but that's news to me. I always thought syndicated stories were identical in every newspaper, but I have no evidence to back that up; it was just an assumption. Can you point me to something suggesting otherwise? --[[User:DrFleischman|Dr. Fleischman]] ([[User talk:DrFleischman|talk]]) 17:33, 12 March 2015 (UTC) |
||
:::::{{reply|DrFleischman}} Mainly experience. The headlines of course are different, and often part of the story gets left out due to space constraints. I used to read a copy editors' mailing list ([http://www.copyediting-l.info/index.shtml Copyediting-L]) and they would talk about how sometimes they would change this or that. Bill Walsh's site ([http://www.theslot.com/ The Slot]) is great for the things that a good copy editor will do with a story. – [[User:Margin1522|Margin1522]] ([[User talk:Margin1522|talk]]) 20:03, 12 March 2015 (UTC) |
|||
:::Concur with [[User:Prototime|Prototime]] & [[User:Margin1522|Margin1522]] & [[User:Ezhiki|Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky)]]. And strongly agree with the comment "''As long as both .. are reliable sources .. the choice .. is guided primarily by practical concerns.''" Frankly, I don't think this concept is explicitly called out in [[Wikipedia:Citing_sources#Say_where_you_read_it]]. Can we add it? Something like; |
:::Concur with [[User:Prototime|Prototime]] & [[User:Margin1522|Margin1522]] & [[User:Ezhiki|Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky)]]. And strongly agree with the comment "''As long as both .. are reliable sources .. the choice .. is guided primarily by practical concerns.''" Frankly, I don't think this concept is explicitly called out in [[Wikipedia:Citing_sources#Say_where_you_read_it]]. Can we add it? Something like; |
||
::::''When referencing republished materials or materials duplicated at two or more different sources (e.g. a news article from an [[News agency|agency]] duplicated by various news outlets, exerts from books reprinted in other media, etc), as long as both the sources are reliable, editors should cite the source they originally got the material from, unless there is good reason to believe that one source is significantly easier to access than another. There is no need to cite the [[best evidence rule|most original source]].'' |
::::''When referencing republished materials or materials duplicated at two or more different sources (e.g. a news article from an [[News agency|agency]] duplicated by various news outlets, exerts from books reprinted in other media, etc), as long as both the sources are reliable, editors should cite the source they originally got the material from, unless there is good reason to believe that one source is significantly easier to access than another. There is no need to cite the [[best evidence rule|most original source]].'' |
Revision as of 20:04, 12 March 2015
Manual of Style | ||||||||||
|
Wikipedia Help Project‑class | |||||||
|
Index 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 |
This page has archives. Sections older than 14 days may be automatically archived by Lowercase sigmabot III when more than 4 sections are present. |
So how to handle "updated" dates?
A lot of sources have two dates, a "published" date and an "updated" date. The citation templates have a place only for the publication date, and the rules given here mention only the publication date. I suppose that "update" covers everything from a small correction to a major rewrite. In the case of the latter (and the former too, in some cases maybe) isn't this misleading? A person searching for the article as of that date is going to find something different from the source I used. Isn't this a problem? Has this been addressed? Herostratus (talk) 18:58, 20 February 2015 (UTC)
- To some extent it may depend on the nature of the source. E.g., for some web pages there may be no definite "original" publication date, and the date of the latest revision ("revised publication"?) of the specific material is effectively the date of publication. In other cases, such as the IPCC reports, there is definite (and well known) date of original publication, which needs to be retained. If that is done as "orig-date" then any subsequent updates/corrections could be done as the "date" of (latest) publication. But don't forget that additional information can always be appended to the template, so you do have the option to craft whatever works best. ~ J. Johnson (JJ) (talk) 23:04, 20 February 2015 (UTC)
Supported and unsupported facts in one sentence
Recently (in this diff) there was a major addition about some common errors relating to supported and unsupported facts in one sentence. I see the point – if you add a cite to support one fact, you don't want to imply that other facts are also supported. But I am wondering if we should demand that editors insert {{citation needed}} for the other facts.
The new text is also really hard to understand: "Inserting new text with an inline citation into text that is not supported with an inline citation, requires that the cited text is clearly defined with a {{citation needed}} template placed before the new text (unless the new text is placed at the start of a paragraph)." IMO this is almost incomprehensible until you look at the example.
The new text could probably be rewritten to be easier to understand. But I'm wondering, are editors actually going to do that? Adding a source improves the verifiability of the sentence by supporting a fact. But we are demanding that editors add one or more {{citation needed}} templates, which makes it look less verifiable. That strikes me as a bit obsessive, and I doubt that anyone is going to it.
Would it not be possible to make the same point without demanding that editors add new {{citation needed}} templates? For example, suggest that they do something (e.g. rewrite the sentence) to clarify which fact they are supporting. – Margin1522 (talk) 05:52, 1 March 2015 (UTC)
- The addition is wrong. It says that if a supported claim is added to the end of a pre-existing sentence that is not supported with a citation, a {{citation needed}} template must be added before the supported additon. But it is not required that all statements be supported by an inline cite; "the sun is bigger than the earth" is a well-known fact that does not require a citation. This whole new passage strives for a level of specificity in citations that is just not feasible in English and is not expected in quality writing. Some of the burden falls on the reader: if the reader cares which cite supports which fact, it is up to the reader to obtain the sources, read the specified pages, and figure out for herself which part of the sentence is supported by the source. I favor reverting User:PBS's edit. Jc3s5h (talk) 14:02, 1 March 2015 (UTC)
You write "But it is not required that all statements be supported by an inline cite" yes that is true. Blue sky citations are not needed. The Sun and the moon examples are used throughout this text so please put to one side the idea that it is obvious, otherwise we have to change all the example. Here is problem expressed using facts that would not be obvious:
- "The girls came from Liverpool, England. They both had dark hair.source a"
Let us suppose a new phrase is added
- "The girls came from Liverpool, in north-west England.source b" They both had dark hair.source a"
Let us suppose that source b is a book about Liverpool not one about the girls. We have now lost text–source integrity. This can be fixed so:
- "The girls came from Liverpool,source a in north-west England.source b" They both had dark hair.source a"
Of course the other solution is as the sentence are now a summary of two sources to place the sources in other positions. Like this
- "The girls came from Liverpool, in north-west England.source a;source b They both had dark hair.source a"
or
- "The girls came from Liverpool, north-west England. They both had dark hair.source a;source b"
It depends on exactly what is being inserted where, and what the editor considers the best solution, but which ever solution is chosen it can not be
- "The girls came from Liverpool, in north-west England.source b" They both had dark hair.source a
as that looses text–source integrity. In a similar way if the same paragraph exists in a paragraph without a source:
- "The girls came from Liverpool, England. They both had dark hair."
If source b is added then there also needs to be a citation needed so that it is clear that source be does not cover the girls:
- "The girls came from Liverpool,[citation needed] in north-west England source b They both had dark hair."
Jc3s5h, now that the example does not use a well know blue sky facts it is probably clearer to you. The inclusion of "source b", implies that it supports all the facts in the sentence. Without a {{citation needed}} one has lost Text—source integrity, but equally no {{citation needed}} needs to be place at the end of the second sentence due the inclusion of "source b" because its inclusion does not affect the Text—source integrity of that second sentence.
@user:Margin1522 you write "The new text could probably be rewritten to be easier to understand." My explanation is a first cut and no doubt it can be improved in the usual way (and I look forward to seeing your improvements). As to your second point "For example, suggest that they do something (e.g. rewrite the sentence) to clarify which fact they are supporting." In reality that is not what happens. If we followed your suggestion we could also remove the older:
- "Do not add other facts or assertions into a fully cited paragraph or sentence ...without including a source to support the new information."
These are really common errors that degrade Text–source integrity in lots of articles, and while rewriting might be desirable as often as not that is not what editors do. Instead the incrementally adjust a paragraph by making incremental changes. -- PBS (talk) 15:23, 2 March 2015 (UTC)
- PBS's post does nothing to change my opposition. Text-source integrity is impossible and this addition is trying to hold the ocean back. Jc3s5h (talk) 15:31, 2 March 2015 (UTC)
I agree in opposition with Jc3s5h. The project page addition is confusingly inadequate in this regard, as detailed somewhat in the examples above. It articulates requirements which do not solve all problems and which, if strictly complied with, create new problems by isolating previously supported content from their supporting sources by inserting intervening cite(s) not providing support to the newly isolated assertion. I think that the insertion does not provide a comprehensive solution to the problem it attempts to address, that the addition is both overly specific and overly complicated as it stands, and that making it more comprehensive would only worsen the overcomplication problem. Perhaps an essay or a help page addressing this issue more completely which could be mentioned in and wikilinked from this project page would be a better approach. Wtmitchell (talk) (earlier Boracay Bill) 23:39, 2 March 2015 (UTC)
- I think deleting the entire section would be better than leaving it as it was just before PBS's change. I think text-source integrity is good, but only if it can be achieved without turning a passage into a tangled confusing mess. It's just one factor to consider in good writing. My cardinal principle is that placing a citation does not verify a claim; readers verify claims by obtaining the cited source and comparing it to the Wikipedia passage; at that time, the reader will know which parts of the claim are supported and which parts are not. Citations are merely citation aids. Jc3s5h (talk) 14:17, 3 March 2015 (UTC)
- text-source integrity was introduced into this guideline by SlimVirgin with Revision as of 22:11, 29 September 2010. She introduced a section describing the conecpt later the same year (8 November 2010). As the concept has been present in this guideline for four and half years. It helps new editors, who are not familiar with the concept of inline citations, and the particular problems that a collective work such as this has, meet the policy requirements of WP:CHALLENGE in a way that most more experienced editors consider acceptable. -- PBS (talk) 17:28, 5 March 2015 (UTC)
- I think deleting the entire section would be better than leaving it as it was just before PBS's change. I think text-source integrity is good, but only if it can be achieved without turning a passage into a tangled confusing mess. It's just one factor to consider in good writing. My cardinal principle is that placing a citation does not verify a claim; readers verify claims by obtaining the cited source and comparing it to the Wikipedia passage; at that time, the reader will know which parts of the claim are supported and which parts are not. Citations are merely citation aids. Jc3s5h (talk) 14:17, 3 March 2015 (UTC)
- @Wtmitchell you write "create new problems by isolating previously supported content from their supporting sources by inserting intervening cite(s) not providing support to the newly isolated assertion." I am not advocating anything I am trying to highlight some of the problems that already exist in editing and maintaining Text-source integrity. What I am describing is what happens at the moment. Typically there will be a paragraph supported by a citation at the end of the paragraph. A new fact in a new sentence will be inserted into the paragraph with a citation. If a copy of the paragraph's citation is not added before the new sentence Text-source integrity is lost. -- PBS (talk) 12:29, 3 March 2015 (UTC)
- I removed that recent extension of text-source integrity, which made it hard to understand. Also, that section needs to be higher in the guideline, near footnotes. Sarah (SV) (talk) 17:48, 5 March 2015 (UTC)
- SV moving the secion up has been discussed before and rejected (simply because "Text–source integrity", "Bundling citations" and "In-text attribution" do not fit into the how to create a citation which is what most of this guideline is about. If you wanted to move it up where would you want to place it? The two examples that I added any you have removed document problems that frequently occur when inexperienced editors add text into a fully cited article, you said in the history "too long and confusing" (aside from deleting them completely!) how would you suggest shorting them and making them less confusing? -- PBS (talk) 19:52, 5 March 2015 (UTC)
- I don't recall it being rejected; I recall that you moved it without discussion, but I may not be remembering the discussion. I don't know how I would write it, but that long addition was too much. Even as written it's not great, but I'd need to think about it. The only point of the section is to remind people that cites need to be near the cited material or in some way flagged as belonging to it, so we don't need a lot of detail. Sarah (SV) (talk) 22:02, 5 March 2015 (UTC)
- Basically I agree with that, and I agree with PBS's point. Editors should be aware that simply placing a cite at the end of the sentence can be unclear. Another idea might be to take a hint from the next section (Bundling citations) and say that the supported fact(s) could be mentioned explicitly in the footnote, like
<ref>For the Moon's size, see {{cite book|...
. – Margin1522 (talk) 23:15, 5 March 2015 (UTC)
- Basically I agree with that, and I agree with PBS's point. Editors should be aware that simply placing a cite at the end of the sentence can be unclear. Another idea might be to take a hint from the next section (Bundling citations) and say that the supported fact(s) could be mentioned explicitly in the footnote, like
Margin1522, the section used to do that. It said:
When using inline citations, it is important to maintain text-source integrity. The point of an inline citation is to allow readers and other editors to check that the material is sourced; that point is lost if the citation is not clearly placed. The following inline citation, for example, is not helpful, because the reader does not know whether each source supports the material, or each source supports part of it, or whether just one source supports it with the others added as further reading:
Delia Smith is the UK's best-selling cookery writer.[1][2][3][4]
The distance between material and its source is a matter of editorial judgment, but adding text without placing its source clearly can lead to allegations of original research, violations of the sourcing policy, and even plagiarism. Where you are using multiple sources for one sentence, or where you prefer not to add each source directly after the phrase or sentence it supports, consider bundling citations at the end of the paragraph with an explanation in the footnote regarding which source supports which point; see below for how to do that. Editors should exercise caution when rearranging material to ensure that the text-source relationship is maintained.
I can't remember when or why it was changed, but I think I prefer (something like) the above. Sarah (SV) (talk) 16:18, 6 March 2015 (UTC)
Adding ping to PBS. Sarah (SV) (talk) 16:22, 6 March 2015 (UTC)
Proposal: Encourage reference information in the Reference section
As the Avoiding clutter subsection notes, "inline references can significantly bloat the wikitext in the edit window and can become difficult and confusing." When trying to edit an article with a significant number of references, such as the Boston Marathon bombings article, it is very difficult to read and work in the edit space when all the reference information is in the midst of the text. Contrast that with the Knights of Columbus article. There are over 130 references, but all the data is at the very end, in the References section. I propose that we encourage the use of WP:NAMEDREFS in the body of the article, and place the full references in the References section. --Briancua (talk) 15:15, 6 March 2015 (UTC)
- I was with you all the way to "namedrefs" — whoa, all stop! You have identified a problem, but your specified solution (which implies list-defined references) is not the only one, and is not entirely acceptable because of certain problems. A better way (IMHO) is to use short cites in the text, but this is opposed by many editors who choke on the implied use of {{Harv}} templates. So where do you want to go with this? ~ J. Johnson (JJ) (talk) 22:05, 11 March 2015 (UTC)
- Named refs suck because it is hard for the editor to know the name of a given reference or to find it if it is invoked several times and you don't know in which one the reference is defined. Harv shortrefs with the {{sfn|author|year|page}} template are much easier to work with and so are inline parenthetical refs. Also we do not have a house citation style and we dont need one.·maunus · snunɐɯ· 22:19, 11 March 2015 (UTC)
Citing republished pieces; What source do we cite?
Following from a recent debate, a few quick questions;
Regarding republished sources (e.g. an AP news piece republished by CNN), several questions.
- When citing republished pieces, is there any requirement to cite the original source? In other words, should one cite AP if AP is the original source?
- Is there a policy anywhere which specifically calls out how to deal with citing republished sources?
- If there is no existing policy, should we have one? Is this topic worth an RfC?
Thanks in advance for weighing in! NickCT (talk) 15:57, 11 March 2015 (UTC)
- I believe this is already addressed by WP:SAYWHEREYOUGOTIT (which is followed by the also-somewhat-relevant section titled "Reprints of older publications"), which is a part of this guideline?—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); March 11, 2015; 16:04 (UTC)
- @Ezhiki: - Good reference cite. But I'm having trouble interpreting it for this situation. If I read an AP news article on CNN, then I feel as though I've got the news from CNN, not AP. So would I cite CNN? NickCT (talk) 16:23, 11 March 2015 (UTC)
- The cite, whatever form it takes, should mention both AP (since that's the source) and CNN (since that's where you got the news from). The actual implementation doesn't matter that much; as long as the formatting matches that of other cites in the article, it would be acceptable. Does this help?—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); March 11, 2015; 16:42 (UTC)
- Thanks for the reply. In terms of the mechanics of our citation template slots, how do we document both? in this example |publisher=CNN |?=AP? Thanks again. Hugh (talk) 16:49, 11 March 2015 (UTC)
|work=CNN
|agency=AP
- —Trappist the monk (talk) 17:00, 11 March 2015 (UTC)
- Thanks Trappist the monk. Can you look at this article. Would that be
|work=NBC
|agency=The Center for Public Integrity
, or something else? NickCT (talk) 17:14, 11 March 2015 (UTC)- Yeah, I think so.
- —Trappist the monk (talk) 17:51, 11 March 2015 (UTC)
- Thanks Trappist the monk. Can you look at this article. Would that be
- WP:NEWSORG: "Most newspapers reprint items from news agencies such as BBC News, Reuters, Agence France-Presse or the Associated Press, which are responsible for the accuracy. The agency should be cited in addition to the newspaper that reprinted it." Hugh (talk) 14:15, 12 March 2015 (UTC)
- Thanks for the reply. In terms of the mechanics of our citation template slots, how do we document both? in this example |publisher=CNN |?=AP? Thanks again. Hugh (talk) 16:49, 11 March 2015 (UTC)
- The cite, whatever form it takes, should mention both AP (since that's the source) and CNN (since that's where you got the news from). The actual implementation doesn't matter that much; as long as the formatting matches that of other cites in the article, it would be acceptable. Does this help?—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); March 11, 2015; 16:42 (UTC)
- @Ezhiki: - Good reference cite. But I'm having trouble interpreting it for this situation. If I read an AP news article on CNN, then I feel as though I've got the news from CNN, not AP. So would I cite CNN? NickCT (talk) 16:23, 11 March 2015 (UTC)
- Ezhiki, I don't think SAYWHEREYOUGOTIT quite applies. That guideline is just for when one source is cited by another source, and you've read the citING source but not the citED source. Here we have a different situation. The two sources are identical to each other, like a syndication, and we've read both of them. SAYWHEREYOUGOTIT would suggest (second para) that the original source may be cited directly without credit. But Nick's question is, which source is preferable? I submit the original source is preferable, as it allows readers to more easily assess reliability. --Dr. Fleischman (talk) 20:54, 11 March 2015 (UTC)
- @Ezhiki and DrFleischman: - Not sure about that Fleis. If for nothing else than the practical concern that the original source might not be available. If I get an AP piece off of Fox News, and the original AP material isn't readily available, what am I meant to do? Cite something I can't see? If we follow SAYWHEREYOUGOTIT, at least we guarantee that an editor is citing material they have in front of them. NickCT (talk) 21:03, 11 March 2015 (UTC)
- You certainly aren't supposed to cite what you can't see. If you did read the news directly from AP, then of course there is no need to include CNN/Fox/ABC/etc. (although you might still choose to do so for convenience, if the AP link is not easily accessible but CNN or some other news outlet is). But if something is coming from AP but you can only read it on CNN/etc., then SAYWHEREYOUGOTIT definitely applies.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); March 11, 2015; 21:11 (UTC)
- @Ezhiki: - Hmmmm.... Anticipating the follow up question; What if you can read it on AP, even though you originally read it on CNN? Is there obligation to go to the original source? NickCT (talk) 21:15, 11 March 2015 (UTC)
- Right, we're talking about cases when both sources are available online, no paywalls, and you've confirmed they have exactly the same content. --Dr. Fleischman (talk) 21:22, 11 March 2015 (UTC)
- While we're at it, why don't we go ahead and add another wrinkle to our scenario, say a wide differential in noteworthiness, say one version was on the front page of the NYT and another was on a news agency website. Hugh (talk) 21:29, 11 March 2015 (UTC)
- "the original source is preferable, as it allows readers to more easily assess reliability" Long term, which is the hat we are expected to wear, easier access is afforded by the most noteworthy version, which would be most likely to be indexed and archived in an EBSCO or a ProQuest. The biggest agencies such as McClatrey may be indexed and archived, regardless of whether or not a paper picks them up, but in general a smaller news agency is not. Hugh (talk) 21:36, 11 March 2015 (UTC)
- The original source is the best; "most noteworthy" is an inherently non-objective standard that is likely to bring on endless disputes. "The WaPo is more important than NYT!" or "ABC news is more noteworthy than Brookings." Endless and unneccesary. Capitalismojo (talk) 01:12, 12 March 2015 (UTC)
- Guys, I think you are losing the big picture here. As long as both the original and the republisher are reliable sources and with all other things being equal, the choice of one over the other is guided primarily by practical concerns. Wikipedia does not employ citations to collect the "best" (or the "most noteworthy") versions out there; citations are used first and foremost to enable readers to verify the statements being made in the article. And if republished content is more easily accessible and comes from a reliable source, there is no reason why it can't be used over the original, but perhaps no longer easily accessible source (and that's one possible scenario where SAYWHEREYOUGOTIT comes into play). The needs of the readers normally outweigh more esoteric concerns (such as which of the two equally acceptable reliable sources is "better", and especially why it is better).
To return to Dr. Fleischman's scenario (when both sources are available online, no paywalls, and you've confirmed they have exactly the same content), in that case it would be just a personal preference. As long as a statement made in the article can be verified with one or another reliable source, your job is done. If both sources are equally accessible, editors shouldn't be changing one to another simply because they like the other one better. If circumstances change, then it's a different story, of course.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); March 12, 2015; 13:22 (UTC) - @Capitalismojo: Circulation is on our RS checklist WP:RSVETTING. Hugh (talk) 14:49, 12 March 2015 (UTC)
- This is not about identifying RS, if it is we are all at the wrong noticeboard. Capitalismojo (talk) 14:51, 12 March 2015 (UTC)
- "As long as a statement made in the article can be verified with one or another reliable source, your job is done." Thanks for your reply. I see what you are saying, but in practice in WP, once you get passed the reliability of sources, and the verifiability of proposed content, might you run into issues of noteworthiness, at which point there may well be a difference in noteworthiness between two similar/identical sources? Thanks again. Hugh (talk) 16:31, 12 March 2015 (UTC)
- "Noteworthiness" of a source is not a criterion Wikipedians should consider; only reliability and suitability in a given context. If two otherwise reliable sources are similar/identical, one may still be somewhat superior for practical reasons (e.g., one is the original and the other one is a re-publication; one can be linked to and the other can't; etc.), but barring that, it makes no difference which one you end up using.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); March 12, 2015; 16:40 (UTC)
- Thanks for your patience. I understand that in terms of reliability and verifiability, the topics of this talk page, other practical things being equal, it makes no difference, but respectfully beyond that, might "no difference" be just a little too strong, in that beyond reliability and verifiability, proposed content may have to jump through a hoop of noteworthiness AKA due weight, off-topic here perhaps, but would you acknowledge that in the context of a weight discussion, the front page of the NYT is different from a local/regional newspaper or wire service website? Thanks again. Hugh (talk) 16:56, 12 March 2015 (UTC)
- I kind of bundled (un)due weight and other similar concerns under "given context". It's not uncommon for a source to be reliable but not acceptable in some contexts (for example, a specific writer's biography is not a good source to use in articles about geography or history, even if it does mention some geographic/historical details in passing). In a similar fashion, I can very well imagine a situation when a local paper would be a better choice than the front page of the NYT. But these are the situations when other things are not being equal, and that always warrants some discussion.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); March 12, 2015; 18:38 (UTC)
- Thanks for your patience. I understand that in terms of reliability and verifiability, the topics of this talk page, other practical things being equal, it makes no difference, but respectfully beyond that, might "no difference" be just a little too strong, in that beyond reliability and verifiability, proposed content may have to jump through a hoop of noteworthiness AKA due weight, off-topic here perhaps, but would you acknowledge that in the context of a weight discussion, the front page of the NYT is different from a local/regional newspaper or wire service website? Thanks again. Hugh (talk) 16:56, 12 March 2015 (UTC)
- Guys, I think you are losing the big picture here. As long as both the original and the republisher are reliable sources and with all other things being equal, the choice of one over the other is guided primarily by practical concerns. Wikipedia does not employ citations to collect the "best" (or the "most noteworthy") versions out there; citations are used first and foremost to enable readers to verify the statements being made in the article. And if republished content is more easily accessible and comes from a reliable source, there is no reason why it can't be used over the original, but perhaps no longer easily accessible source (and that's one possible scenario where SAYWHEREYOUGOTIT comes into play). The needs of the readers normally outweigh more esoteric concerns (such as which of the two equally acceptable reliable sources is "better", and especially why it is better).
- The original source is the best; "most noteworthy" is an inherently non-objective standard that is likely to bring on endless disputes. "The WaPo is more important than NYT!" or "ABC news is more noteworthy than Brookings." Endless and unneccesary. Capitalismojo (talk) 01:12, 12 March 2015 (UTC)
- @Ezhiki: - Hmmmm.... Anticipating the follow up question; What if you can read it on AP, even though you originally read it on CNN? Is there obligation to go to the original source? NickCT (talk) 21:15, 11 March 2015 (UTC)
- You certainly aren't supposed to cite what you can't see. If you did read the news directly from AP, then of course there is no need to include CNN/Fox/ABC/etc. (although you might still choose to do so for convenience, if the AP link is not easily accessible but CNN or some other news outlet is). But if something is coming from AP but you can only read it on CNN/etc., then SAYWHEREYOUGOTIT definitely applies.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); March 11, 2015; 21:11 (UTC)
- @Ezhiki and DrFleischman: - Not sure about that Fleis. If for nothing else than the practical concern that the original source might not be available. If I get an AP piece off of Fox News, and the original AP material isn't readily available, what am I meant to do? Cite something I can't see? If we follow SAYWHEREYOUGOTIT, at least we guarantee that an editor is citing material they have in front of them. NickCT (talk) 21:03, 11 March 2015 (UTC)
- For scientific papers, more and more are first put up on a preprint server like arXiv, and then peer-reviewed republished in a journal. Or they are first published at a workshop, and then published in a journal special edition. In those cases, I would nearly always cite the final version, as it has had additional quality control. I think that is a general situation - the same article may not meet WP:RS if self-published, but will meet WP:RS when picked up by a reliable publisher. --Stephan Schulz (talk) 13:36, 12 March 2015 (UTC)
- If we are talking material that is just republished, not re-edited or changed in any way? First version or re-published? Capitalismojo (talk) 14:40, 12 March 2015 (UTC)
- It depends. Most often the journal version will have minor or even major changes, but not necessarily so. Even if identical, the later version will have more layers of quality control. --Stephan Schulz (talk) 18:04, 12 March 2015 (UTC)
- If we are talking material that is just republished, not re-edited or changed in any way? First version or re-published? Capitalismojo (talk) 14:40, 12 March 2015 (UTC)
I concur with Ezhiki; Wikipedia does not have a "best evidence rule", nor should it. Editors should feel free to cite to whichever reliable version of the source they wish. (I qualify that with "reliable" version because I also agree with Stephan Schulz that self-published drafts generally fail WP:RS, even if they are identical to a final version published by third parties.) –Prototime (talk · contribs) 15:36, 12 March 2015 (UTC)
- And if a dispute arises over which of two identical reliable sources is better, what then? It's a wash? --Dr. Fleischman (talk) 16:34, 12 March 2015 (UTC)
- As long as its reliable... One thing with wire-service stories is that newspapers typically edit them. So the versions of AP stories that you see in different papers will often be slightly different. So if you get it from a paper you should cite the paper. And set the agency= parameter to indicate that it was a wire-service story. – Margin1522 (talk) 16:41, 12 March 2015 (UTC)
- Margin1522, I'm no journalism expert but that's news to me. I always thought syndicated stories were identical in every newspaper, but I have no evidence to back that up; it was just an assumption. Can you point me to something suggesting otherwise? --Dr. Fleischman (talk) 17:33, 12 March 2015 (UTC)
- @DrFleischman: Mainly experience. The headlines of course are different, and often part of the story gets left out due to space constraints. I used to read a copy editors' mailing list (Copyediting-L) and they would talk about how sometimes they would change this or that. Bill Walsh's site (The Slot) is great for the things that a good copy editor will do with a story. – Margin1522 (talk) 20:03, 12 March 2015 (UTC)
- Margin1522, I'm no journalism expert but that's news to me. I always thought syndicated stories were identical in every newspaper, but I have no evidence to back that up; it was just an assumption. Can you point me to something suggesting otherwise? --Dr. Fleischman (talk) 17:33, 12 March 2015 (UTC)
- Concur with Prototime & Margin1522 & Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky). And strongly agree with the comment "As long as both .. are reliable sources .. the choice .. is guided primarily by practical concerns." Frankly, I don't think this concept is explicitly called out in Wikipedia:Citing_sources#Say_where_you_read_it. Can we add it? Something like;
- When referencing republished materials or materials duplicated at two or more different sources (e.g. a news article from an agency duplicated by various news outlets, exerts from books reprinted in other media, etc), as long as both the sources are reliable, editors should cite the source they originally got the material from, unless there is good reason to believe that one source is significantly easier to access than another. There is no need to cite the most original source.
- Anyone think this would be worthwhile? NickCT (talk) 16:50, 12 March 2015 (UTC)
- I don't understand the logic of this proposal. Why is "the source they originally got the material from" preferable? WP:SAYWHEREYOUGOTIT is ultimately about verifiability; when both sources have been verified there's no longer a verifiability issue. --Dr. Fleischman (talk) 17:36, 12 March 2015 (UTC)
- (edit conflict) @DrFleischman: I'm not sure why there would be a dispute over which of "two identical reliable sources is better" (except, perhaps, that one source is more accessible than another). So long as a reliable source is cited, the material it supports is verifiable, satisfying WP:V. Whether the reliable source cited is an original version or a republished version doesn't affect the verifiability of the claim. If there ever were such a dispute (and it wasn't about accessibility), the original citation should be maintained because there would be no reason to disturb it.
- That's a decent proposal, NickCT. In response to HughD's concern, NickCT's proposal states what is already the rule; since there's nothing in Wikipedia policies or guidelines that establishes a best evidence rule, NickCT's proposal is not making a new rule. But I do think there's valuing in stating the rule outright so that editors have clarity when this situation arises. I'd support adding something like NickCT's proposal to this guideline. –Prototime (talk · contribs) 17:47, 12 March 2015 (UTC)
- In response to Dr. Fleischman's newest comment (didn't see due to the edit conflict): I agree with you that NickCT's proposal isn't necessarily justified by WP:SAYWHEREYOUGOTIT, since multiple editors could read and cite different versions of the same source; it's more of a WP:IFITAINTBROKE issue. –Prototime (talk · contribs) 17:58, 12 March 2015 (UTC)
- "I'm not sure why there would be a dispute over which of "two identical reliable sources is better" (except, perhaps, that one source is more accessible than another)" Even after reliability and verifiability and accessibility issues are settled, proposed content also has to pass muster in terms of due weight. The context of a weight discussion is an example of when two similar sources may be different in noteworthiness. Hugh (talk) 18:14, 12 March 2015 (UTC)
- @Hugh: Due weight requires claims be presented "in proportion to their representation in reliable sources"; it does not speak the nature of the reliable sources themselves. If due weight is a concern, it would make more sense to cite to both versions of the source than to cite to any one particular version (and more sense still to cite to multiple sources that are completely different). –Prototime (talk · contribs) 18:45, 12 March 2015 (UTC)
- (edit conflict)There is a dispute, and I do believe there's a small but real benefit to citing the original source, other issues (such as access) being equal. By citing the original source we make it clearer who was responsible for producing the content (i.e. who the author and editor worked for). Although there is consensus the outlet for the original source is reliable, there are sometimes questions raised about its bias, including possible COI (see WP:QS). Making it clear who produced the content helps readers to evaluate this for themselves. As I see it, following WP:SAYWHEREYOUGOTIT just because it's related and already in the guideline would be a triumph of rote bureaucracy to the (slight) detriment of the encyclopedia. --Dr. Fleischman (talk) 18:08, 12 March 2015 (UTC)
- @DrFleischman: - re "sometimes questions raised about its bias" - If the original source is biased, and republisher isn't biased, why is the republisher publishing something that they feel is biased and/or inaccurate material? If you said something that I felt was biased, I wouldn't present that thing to some third party as my own idea. NickCT (talk) 18:32, 12 March 2015 (UTC)
- I don't know what the republisher's thought process is; maybe they think the original source is unbiased or maybe they don't. I'm not saying the original source is biased. I'm saying questions have been raised about the content producer's bias, and readers are less able to evaluate those questions for themselves when the identity of the content producer is obscured. --Dr. Fleischman (talk) 18:41, 12 March 2015 (UTC)
- @DrFleischman: - re "sometimes questions raised about its bias" - If the original source is biased, and republisher isn't biased, why is the republisher publishing something that they feel is biased and/or inaccurate material? If you said something that I felt was biased, I wouldn't present that thing to some third party as my own idea. NickCT (talk) 18:32, 12 March 2015 (UTC)
- @Dr. Fleischman: If there is a COI that destroys a source's reliability, then per WP:QS the source is unreliable (except as a source about itself) regardless of which version is cited to, and thus the citation should be removed completely, not replaced with a citation to the original version. –Prototime (talk · contribs) 18:45, 12 March 2015 (UTC)
- I'm not saying there's a COI. Concerns were raised about COI but they haven't been fully investigated. But that shouldn't matter, as it's a mere illustration of why this matters. My point is that original source should always be cited in these otherwise-equal situations so that readers know who the content producer is and can evaluate its bias for themselves. --Dr. Fleischman (talk) 18:52, 12 March 2015 (UTC)
- We've been discussing this assuming that the source, regardless of the version cited to, is reliable. But you seem to suggest that reliability cannot be adequately determined without reviewing the original version; is that right? That would create a very high threshold for determining reliability. And I'm not sure how that principle, if adopted, could be applied only in "otherwise-equal" situations given that reliability concerns trump access concerns. –Prototime (talk · contribs) 19:13, 12 March 2015 (UTC)
- Prototime, I guess I'm not being very clear. The reliability of both sources is established in this scenario. But that doesn't mean they're free of bias, per WP:BIASED. Readers should be informed who the content producer is so they can evaluate their bias. Let's take a hypothetical but concrete example. There is consensus here that Fox News sources are often reliable. But Fox News is a bit of a lightning rod; some readers don't trust Fox News and discount content sourced to it, and they have every right to do so. If Fox News produces a story that's then republished by another outlet, say, I don't know, WSJ, should we be citing Fox News or WSJ? We should cite Fox News, so that readers can make informed decisions. The same thing works vice versa: if Fox News republishes a WSJ story, we should cite the WSJ version. --Dr. Fleischman (talk) 19:42, 12 March 2015 (UTC)
- Your question is essentially about a different topic, the question of neutrality and bias is distinct from the question of which source to cite when a piece is republished. We simply cannot require editors to find the original source, that would be too much of an obstacle for many editors. What we can do is encourage using the original source, because that is simply the best scholarly practice. But the question of bias in either the primary por secondary publisher is not relevant in terms of what source to cite - that is a discussion to be had at the talkpage of the article, and decided in the given context. The function of the citation on the page is to show where the information comes from, we have no obligation to assist the reader in evaluating the source - rather that is an editorial question.·maunus · snunɐɯ· 19:49, 12 March 2015 (UTC)
- Are we talking about citing or republishing? Hugh (talk) 19:46, 12 March 2015 (UTC)
- Prototime, I guess I'm not being very clear. The reliability of both sources is established in this scenario. But that doesn't mean they're free of bias, per WP:BIASED. Readers should be informed who the content producer is so they can evaluate their bias. Let's take a hypothetical but concrete example. There is consensus here that Fox News sources are often reliable. But Fox News is a bit of a lightning rod; some readers don't trust Fox News and discount content sourced to it, and they have every right to do so. If Fox News produces a story that's then republished by another outlet, say, I don't know, WSJ, should we be citing Fox News or WSJ? We should cite Fox News, so that readers can make informed decisions. The same thing works vice versa: if Fox News republishes a WSJ story, we should cite the WSJ version. --Dr. Fleischman (talk) 19:42, 12 March 2015 (UTC)
- We've been discussing this assuming that the source, regardless of the version cited to, is reliable. But you seem to suggest that reliability cannot be adequately determined without reviewing the original version; is that right? That would create a very high threshold for determining reliability. And I'm not sure how that principle, if adopted, could be applied only in "otherwise-equal" situations given that reliability concerns trump access concerns. –Prototime (talk · contribs) 19:13, 12 March 2015 (UTC)
- I'm not saying there's a COI. Concerns were raised about COI but they haven't been fully investigated. But that shouldn't matter, as it's a mere illustration of why this matters. My point is that original source should always be cited in these otherwise-equal situations so that readers know who the content producer is and can evaluate its bias for themselves. --Dr. Fleischman (talk) 18:52, 12 March 2015 (UTC)
- As long as its reliable... One thing with wire-service stories is that newspapers typically edit them. So the versions of AP stories that you see in different papers will often be slightly different. So if you get it from a paper you should cite the paper. And set the agency= parameter to indicate that it was a wire-service story. – Margin1522 (talk) 16:41, 12 March 2015 (UTC)
- "the identity of the content producer" That's what the agency parameter of the citation template is for WP:NEWSORG. Hugh (talk) 19:38, 12 March 2015 (UTC)
- Well that's interesting - I hadn't noticed that tidbit in WP:RS. So that's what Srich was referring to. In our case the content producer isn't exactly a news agency, and I still think if the original source is available it should be used instead of the republished version to avoid confusion. But that definitely mitigates the problem, not perfectly, but it's an improvement. --Dr. Fleischman (talk) 19:50, 12 March 2015 (UTC)
- "the identity of the content producer" That's what the agency parameter of the citation template is for WP:NEWSORG. Hugh (talk) 19:38, 12 March 2015 (UTC)