Jump to content

Wikipedia talk:Manual of Style/Lead section

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia
< Wikipedia talk:Manual of Style

This is an old revision of this page, as edited by Koncorde (talk | contribs) at 23:25, 28 August 2024 (→MOS:CITELEAD). The present address (URL) is a permanent link to this revision, which may differ significantly from the current revision.

Citizenship in the lede

We are having a debate on Talk:Stevie Wonder#Citizenship over his newly acquired dual citizenship status. It doesn't seem as though there is a standard to go by on how to refer to a person with dual citizenship; for instance Tina Turner lived in Switzerland for many years, yet she was not referred to as "American-Swiss". Also, Tom Hanks is American-Greek and this is not reflected in the lede. The infoboxes of these articles do point out additional citizenships, and I made sure this was the case on Stevie Wonder. However, several editors insist that his dual citizenship status must be mentioned in the lede despite him being an American citizen for almost the entirety of his life. dekema (Formerly Buffaboy) (talk) 05:24, 16 May 2024 (UTC)[reply]

MOS:LEADLANG

Thoughts on adding the proposed sentence (shown below in bold so editors can readily see the change):

Do not include foreign equivalents in the text of the lead sentence for alternative names or for particularly lengthy names, as this clutters the lead sentence and impairs readability. Do not include foreign equivalents in the lead sentence just to show etymology. Do not include in the text of the lead foreign equivalents written in non-Roman script, as this is unhelpful to the non-specialist reader. Foreign-language names should be moved to a footnote or elsewhere in the article if they would otherwise clutter the first sentence.[A]

Some examples where non-Roman script provides clutter to all readers, but probably helps only a very tiny minority of readers:

These would be better as footnotes or moved elsewhere in the article, and not in the text of the lead. Thoughts? CUA 27 (talk) 20:02, 7 June 2024 (UTC)[reply]

Checking to see if anyone wants to weigh in? CUA 27 (talk) 12:46, 11 June 2024 (UTC)[reply]

I don't think this change is necessary. There is nothing wrong with having a fairly short expression in another script in the lead sentence. Excessive cases are to be avoided, but that's common sense and doesn't need a particular rule. Plus the rule against clutter is already there. Gawaon (talk) 12:55, 11 June 2024 (UTC)[reply]
I wonder if there's a way to put these in an infobox instead. There's nothing inherently wrong with giving the name of a city in all the official/practical local scripts, but I don't necessarily want to see more than one or two in the first sentence itself. WhatamIdoing (talk) 22:37, 25 August 2024 (UTC)[reply]

Worth implementing short description guidance, e.g. WP:SDNONE?

While the mobile apps continue to terrorize the world by not making this at all visible to anyone using them, WP:SDNONE is effectively treated like a MOS guideline: we should probably mention it briefly, right? Remsense诉 20:44, 14 July 2024 (UTC)[reply]

Guideline on citations in the lede

This guideline is a horrible idea. It radically decreases transparency and greatly weakens the verifiability requirement. Editors and readers are expected to either just trust the implicit anonymous claim that every single assertion found in the lede is sourced somewhere in the article, or else read the entire article, which may be very long, just to find where the claim in question and its sourcing could possibly be located. This makes verification far more difficult than under normal circumstances. Given how often unsourced information is added, incorrect conclusions are made from sources, sources are misrepresented (unconsciously or not) or are unreliable, making part of the article - and the most prominent and important one to boot! - extremely difficult to verify is a horrible idea. The benefit is unclear - adding a few citations in the lede, even if they are also found in the body of the article, is not difficult and does not harm readability significantly - and certainly can't compensate for the harm caused. 'Just trust us, it's there somewhere' is an unacceptable principle for an encyclopedia whose articles are produced collectively and anonymously.

Locating the source might be easier when there is an editor that monitors the article constantly, is very familiar with its content and structure and can point the user demanding a citation to the right place in the article body. This may be the true in some cases, but it normally isn't and our official guidelines cannot be based on the assumption that it always is. It also requires the user who placed the tag to take the extra steps of coming back to the article and having a conversation with the editor in question before they can even check whether the source is reliable and says what it is cited for, and the user cannot be reasonably expected to do that each time. Anonymous44 (talk) 12:52, 24 July 2024 (UTC)[reply]

I don't feel similarly about this guideline, but I do note that inexperienced editors will frequently add new information (sometimes cited) directly to the lead rather than under a more appropriate subheading, possibly because the "edit lead section" icon is so prominent in comparison to the rest. By consequence, in practice this guideline is rarely followed except at well-established and stable articles. Folly Mox (talk) 14:15, 24 July 2024 (UTC)[reply]
The lead is meant to be a summary of the rest of the article. If it has its own citations and referecing, the possibilities of original research and/or synthesis increase 10 fold. But as FM says above, this is a rather "late-stage" guideline, which (in practice) is really only applied to rather well written articles. – Aza24 (talk) 18:02, 26 July 2024 (UTC)[reply]
I disagree totally, echoing points above. Moreover, contrary to your vibe check, when this guideline is actually considered or enforced, it only ever makes articles better. To me, it's not really under question that summary paragraphs of cited material do not require their own inline citations. I will also insist that a requirement would mostly result in lazy, less helpful, distracting, and misleading citations. Moreover, it would result in worse prose, as editors would be more likely just to copy-paste to ensure their summary doesn't need to grab citations from anywhere else in the article.
Think about it like this: when this guideline is invoked, it has the effect of making editors look the lead over, and reconsider how well it represents the article as a whole. A much better article usually results from this. Remsense诉 07:36, 31 July 2024 (UTC)[reply]
I agree with Remsense here. Something that's in the lead without being expanded upon (and sourced) in the body can usually be challenged and removed on that reason alone. The lead is meant to be a summary of the main points of the body, therefore in theory it would never need any references of its own (direct quotes exempted). Gawaon (talk) 09:37, 31 July 2024 (UTC)[reply]

when this guideline is actually considered or enforced

I think this is exactly what's wrong with MOS:CITELEAD. This guideline has never even been read by anyone, nevermind actually being considered or enforced. If anyone did read it, they'd notice it says the exact opposite of what everyone linking to it claims:

The lead must conform to verifiability,[...] "all quotations, and any material whose verifiability has been challenged or is likely to be challenged, must include an inline citation to a reliable source that directly supports it". [...] There is no exception to citation requirements specific to leads.[...] The presence of citations in the lead is [not] prohibited in any article.

Here's what everyone linking to it claims it does:
  1. Removes the need for inline citations
  2. Makes exception to the citation requirement specific to leads
  3. Prohibits the presence of citations in the lead
Frankly I'm fine with throwing out the current policy. I'm also cool with keeping the policy as written and telling people who think references are ugly that this plugin exists.
But after having one GAN rejected for uncited claims in the lead, and a renomination rejected for citing these claims, I'll go ahead and say the current policy is awfully-written and unclear. I've removed the weird parts of the policy that subtly imply you should remove references (without ever outright stating this). If someone wants to open an RfC to get consensus on changing this, please go ahead and be my guest. Closed Limelike Curves (talk) 01:52, 26 August 2024 (UTC)[reply]
I have no idea why you think that other people just don't see where it plainly says direct quotations et al. in the lead still require inline citation. Your GAN got rejected either because the review was incompetent or because the lead just didn't do its job of summarizing the body with nothing more. To proclaim based on that experience that nobody else understands what this guideline plainly says and does not say is specious and insulting. Remsense ‥ 论 02:03, 26 August 2024 (UTC)[reply]
The lead summarizes the body. The two reviewers (presumably) disagree about whether the material I linked is controversial or technical enough to require citations in the lead. Closed Limelike Curves (talk) 19:29, 26 August 2024 (UTC)[reply]
In re the (intentionally hyperbolic) statement that This guideline has never even been read by anyone: section-specific shortcuts are getting about 250 page views per month, which suggests some interest in its contents.
My own experience is that the line about Although the presence of citations in the lead is neither required in every article nor prohibited in any article is frequently followed by a paragraph or to that amounts to so you are required to do it my way.
The questions I asked below are relevant. Specifically: If the material in question "must include an inline citation", and that material appears in more than one place, is that material required to have an inline citation in each location? Or is once enough? And if once is enough, does it have to be the first?
For example: I might someday clean up the slightly self-contradictory information in Cancer about risk factors. The article should say (in some form) that most cancers and most cancer deaths are not actually preventable by individuals. That is: About 40% of the (significant) cancers being diagnosed today could have been prevented through lifestyle changes, and the first, second, and third most important lifestyle changes are "Stop smoking already", "No, really, smoking kills", and "You moron, I told you not to smoke". This one risk factor is half of the individually preventable 40% of cancers. So if you're a lifelong non-smoker, and you get cancer, there's a ~67% chance that there was nothing you could do to prevent it. Maybe others (e.g., pollution regulators) could have prevented it, but the odds are that there was nothing you could have done as an individual. Smoking is three times as bad as obesity, four times as bad as moderate alcohol consumption, etc. If you are (or were) a smoker, and you get diagnosed with cancer, there's a 25% that the cancer was caused by smoking.
Something about risk factors should be in the lead, and it should actually say that most cancers can't be prevented through lifestyle changes. This is, in my experience, very surprising to people. My friends are all convinced that not eating enough fruits and vegetables is the #1 cause of cancer, and it's just not true. So I consider this Wikipedia:Likely to be challenged, and it will be cited. But: If the lead says "Most cancers can't be prevented through lifestyle changes", and the article has a whole section, with "twenty-seven 8-by-10 color glossy pictures with circles and arrows and a paragraph on the back of each one explaining what each one was, to be used as evidence", does the article also need to have some of those citations duplicated up in the lead?
LEADCITE doesn't answer that question either way. LEADCITE says stuff has to be cited, but it doesn't say that the stuff has to be cited repeatedly. WhatamIdoing (talk) 03:07, 26 August 2024 (UTC)[reply]
Perhaps these questions would help:
  1. As a general rule, do you personally believe that if the same facts appear multiple times in the same article, they should be cited every time (e.g., copy/paste the same ref to the end of each sentence), or do you personally believe that once is usually enough?
  2. WP:BLP requires an inline citation for contentious matter about living people. If contentious matter about a BLP appears in both the lead and also in the body (a very common situation), do you believe that it should be cited in both places, or is once usually enough?
  3. WP:V requires require an inline citation for direct quotations. If the quotation appears both in the lead and also in the body (a rare situation), do you believe that it should be cited in both places, or is once usually enough?
  4. If once is usually enough, should that once be the "first" mention (e.g., lead) or the "biggest" mention (e.g., a whole section)?
WhatamIdoing (talk) 22:35, 25 August 2024 (UTC)[reply]
I don't care as long as we pick one and stick to it consistently, or just clearly and explicitly state that this is up to whoever creates the article and can't be changed (just like current WP:CITEVAR policy).
Closed Limelike Curves (talk) 01:54, 26 August 2024 (UTC)[reply]
You might be looking for WP:STYLEVAR, which is the generic version for style questions. CITEVAR allows changes (you just need to have a quick chat on the talk page, to make sure nobody will be mad about it). I haven't looked at STYLEVAR in years, but I assume it's a similar approach. WhatamIdoing (talk) 03:10, 26 August 2024 (UTC)[reply]
In practice, if there are already citations in the lead, other editors are unlikely to remove them – why should they bother? In that sense, the first person to write a proper lead (not necessary the first article author) might have a good say in actually getting them added to the existing lead text. However, this cannot extend as far as making a requirement that newly added or significantly rewritten lead paragraphs must be referenced as well. Consider this case (fairly typical in my experience): There's a long and detailed section with 37 references. In the lead it's summarized in just one sentence, or at most a paragraph. So how should we reference that one? Repeating all 37 refs would clearly be excessive and make the lead much less readable, but cutting it down to just 2 or 3, while still getting everything mentioned properly referenced, would at least require detailed knowledge of all the references, and might in many cases be simply impossible. So "here's the summary, see below for details and references" is the most feasible solution in such cases. It's a good thing that CITELEAD allows this, and no first or other editor should be able to forbid it. Gawaon (talk) 06:16, 26 August 2024 (UTC)[reply]
I usually would remove them if they're clearly tacked on by a third-party editor who thought they were needed, or are of clearly lower quality than those actually citing the material in the body being summarized. Remsense ‥ 论 06:43, 26 August 2024 (UTC)[reply]
Same here. I routinely remove them, especially when they are of poorer quality than the sources in the article body. If it isn't in the body then it cannot be in the lead. Hawkeye7 (discuss) 07:21, 26 August 2024 (UTC)[reply]
That rule doesn't apply to sources, and it doesn't apply absolutely to lead content anyway. Sometimes a simple classification, or a statement about related/confuse-able subjects needs to be in the lead, but there isn't anything else to be said, and redundancy is both bad writing and also inconsistent with encyclopedic style. Simvastatin needs to begin with a sentence like "Simvastatin, sold under the brand name Zocor among others, is a statin, a type of lipid-lowering medication", but the article does not need a ==Classification== section that repeats the fact that it's classified as a statin. Once is enough for that kind of super basic fact. WhatamIdoing (talk) 15:34, 26 August 2024 (UTC)[reply]
I think the comments above show this is common. It's also routinely brought up at WP:FAC and WP:GAC, where editors demand you mechanically remove all sources from the lead, regardless of objections or past controversy on a topic.
I agree with you that newly-added or substantially-rewritten leads shouldn't be required to stick to the old citation style. Changing the style as part of a substantial rewrite is allowed by WP:VAR, so long as you provide notice on the talk page. Closed Limelike Curves (talk) 16:43, 26 August 2024 (UTC)[reply]
Please note that this is not a question of style. A good lead is for the most part a summary of the body. As such, it should not require citations, because they can be found in more detail in the body. Gawaon (talk) 17:58, 26 August 2024 (UTC)[reply]
More strongly, when a lead does require citations, it is a hint that some of its content may really belong in the body of the article instead. So omitting the lead citations serves a useful purpose as a flag to editors for that sort of issue.
Additionally, when a lead does properly summarize sourced and expanded body content, it may be very difficult to find a single source that summarizes the same material in the same way, so if we were to cite leads we might be faced with citation overkill when we repeated all the sources needed for the material that the lead summarized. —David Eppstein (talk) 18:20, 26 August 2024 (UTC)[reply]
So:
  • If the lead requires citations, that might indicate a problem.
  • Due to the summative nature of leads (and our rules that we have to Wikipedia:Use our own words), it may be difficult to find a single source that Wikipedia:Directly supports the summary (e.g., "She was an educated woman", even if you have sources that spend pages and pages on the details of her education and how unusual that was for a woman in her culture).
But perhaps the question is closer to:
  • If the lead doesn't require citations, and you want to add them anyway, should a GA or FA reviewer be able to force their exclusion?
WhatamIdoing (talk) 18:42, 26 August 2024 (UTC)[reply]
We're talking about the WP:Manual of Style/Lead section page, so I think it is a question of style. WhatamIdoing (talk) 18:37, 26 August 2024 (UTC)[reply]
@Gawaon: I'm not making radical changes to the policy, or any changes at all. I'm keeping the policy exactly the same but wording it clearly. Right now, the policy says "Citations are permitted but not required in the lead", but is worded so confusingly that everyone linking it claims the opposite.
My edits are just clarifying the current policy:
  1. The manual of style discusses citations in leads, but makes no recommendation for or against including them.
  2. Therefore, WP:STYLEVAR currently applies, because there's no established rule about citations in leads. You can change the style if you have a good reason, but not just because you feel like it.
Now, if you want to change the policy to enforce one way or the other, go ahead. You can rewrite this to say "Citations are forbidden in leads, except if X, Y, Z" or to say "Citations are always required in leads", or you can rewrite it to say "you can include or exclude citations in leads whenever you want, but don't change it without a good reason because edit warring over this is stupid". I personally won't revert regardless of what you do, since I don't care. Closed Limelike Curves (talk) 19:21, 26 August 2024 (UTC)[reply]
I suggest simply leaving the policy as is. Nobody except you so far has even expressed a desire for changes, as far as I can see. So if you're happy with its content too, what would be the point of changing anything? Gawaon (talk) 19:55, 26 August 2024 (UTC)[reply]
I'm OK with leaving the policy, i.e. its content, as is. I think the original commenter is the only person interested in a change (they want to require citations in leads). But the phrasing on this page is confusing and easy to misunderstand.
In fact, so far as I can tell, 2 editors commenting here have explicitly said they want to enforce a policy that does not allow citations in leads, outside exceptional cases (e.g. BLPs). 2 others have implied the same. I think that's fine and dandy, but it's a completely different policy from the one actually written on this page, which waffles between requiring citations and allowing but not requiring them. – Closed Limelike Curves (talk) 20:11, 26 August 2024 (UTC)[reply]
BTW, I looked at 10 current FACs, and 90% of them have no citations in the lead. 20% had an Wikipedia:Explanatory footnote. There may be a perception that FAC "expects" uncited leads, even if there is no actual rule. WhatamIdoing (talk) 18:38, 26 August 2024 (UTC)[reply]
The guideline waffles and ultimately ducks the issue by saying "The necessity for citations in a lead should be determined on a case-by-case basis by editorial consensus." So, there's clearly no hard rule and the OP's point is moot.
Myself, I usually start articles as small stubs and don't create a summary lead until the size of the article seems to warrant one. Citations will therefore appear throughout.
As a related point, note that Wikipedia's main page, which is its primary showcase, never has any citations at all. They are obviously not essential. What matters more is that the statements are accurate and correct. But even, so there's still a disclaimer on that and every page which emphatically advises the reader that Wikipedia cannot guarantee the validity of the information found here.
Andrew🐉(talk) 09:37, 26 August 2024 (UTC)[reply]
  • So it's quite common to see citations in the lead during the early stages of article development, especially in the absence of experienced editor participation. If and when the citations are pulled from the lead as the article progresses, we get a chance to see if this creates any issues. Is there something highly controversial in the lead that warrants a citation? Do random editors repeatedly challenge a claim in this section, despite the lead qualifying as a legitimate summary of the article?
    If you find yourself constantly reminding others that, "Hey, it's in the body right here, complete with inline citations", well, maybe that's when you know you need to make an exception. Add a citation or {{Efn}} to the lead where there are issues, or even consider a rewrite. Exceptions should generally be a rare occurrence. Not sure there's a good way to codify that in the MoS, but MOS:CITELEAD isn't likely to settle the dispute for you. --GoneIn60 (talk) 19:20, 26 August 2024 (UTC)[reply]
  • It radically decreases transparency and greatly weakens the verifiability requirement.
    Honestly, I disagree, A lead is meant to be a summary of content that's cited elsewhere on the page. While it might be nice to have citations in the lead (especially for more contentious articles), we shouldn't be requiring citations in the lead. For some articles, it may be helpful; for others, it defeats the point of having the lead be a concise summary. This is why WP:CITELEAD, as currently written, neither requires nor bans citations in the lead—editors may put as many or as few citations in the lead as they like.
    I'm just speaking from personal experience here, but for well-written articles, I can typically Ctrl+F the claim and find the citation within 3-5 seconds. If editors and readers can't do that, then it's not a well-written lead. – Epicgenius (talk) 21:34, 26 August 2024 (UTC)[reply]
  • The lead should never have citations in it (barring quotations, which should probably not be in the lead anyway). The article is the source for the lead and it is right there for anyone willing to actually read. --User:Khajidha (talk) (contributions) 12:00, 27 August 2024 (UTC)[reply]
    @Khajidha, in the RFC below, does your view align with Wikipedia talk:Manual of Style/Lead section#Citations should be removed from article leads, except for BLP or direct quotations? WhatamIdoing (talk) 17:02, 27 August 2024 (UTC)[reply]
  • I think citations should be avoided in the lead and only put in if contentious. The body of the article is the source. The lead should summarize the article, putting in citations seems to stop people doing that and can make the lead say things which aren't in the article. A citation in the lead indicates to me that probably someone has stuck stuff in the lead without checking the article and making it fit in properly. NadVolum (talk) 21:44, 28 August 2024 (UTC)[reply]

MOS:CITELEAD

Should citations be included for most claims in an article's lead, just like in the body? 19:51, 26 August 2024 (UTC)

Citations should be included in article leads just like in the body

  1. Oppose. Most featured articles do not have citations in leads and I do not want to be the one to correct that.– Closed Limelike Curves (talk) 19:51, 26 August 2024 (UTC)[reply]
  2. Oppose; at best, this will clutter the lead, and at worst, it forces editors into an unpalatable WP:CITEBOMB situation. This is also far beyond what WP:V requires—the policy does not say that a claim must be cited every time it is mentioned, only that claims be cited somewhere in the article. (See footnote [a], which says: The location of any citation—including whether one is present in the article at all—is unrelated to whether a source directly supports the material.) If a claim is not sourced in the lead, it should be tagged with {{citation needed lead}} or removed. Epicgenius (talk) 21:15, 26 August 2024 (UTC)[reply]
  3. Oppose As argued above, this oversteps policy and would invite overloading intros with little blue clicky linky numbers that don't actually help. XOR'easter (talk) 03:32, 27 August 2024 (UTC)[reply]

Citations should be removed from article leads, except for BLP or direct quotations

  1. Oppose. Most featured articles do not have citations in leads, but some do. A campaign to retroactively delete those citations will be a pain in the ass.– Closed Limelike Curves (talk) 19:51, 26 August 2024 (UTC)[reply]
  2. Oppose. This goes too far in the other direction, and in some cases (e.g. basically any article about a controversial subject), there may be a good reason to put citations in the lead. Epicgenius (talk) 21:15, 26 August 2024 (UTC)[reply]
  3. Oppose We don't need a blanket rule in favor of inclusion or in favor of removal. There are more occasions than just direct quotations and BLPs where citations are good. For example, in an article about a constant of nature (e.g., Planck constant), it makes sense to have the numerical value for that constant and a supporting citation to a standard reference work in the lede. Biographies of deceased but still controversial figures are another case where attaching citations directly to a description can back up a statement that partisans would want to dispute. XOR'easter (talk) 03:40, 27 August 2024 (UTC)[reply]

Current policy (editors may do either, but should not switch without a good reason)

  1. Support. This is by far the easiest approach. – Closed Limelike Curves (talk) 19:51, 26 August 2024 (UTC)[reply]
  2. Weak support, although not sure on "should not switch without a good reason". We realistically should make it clear that citations are optional only if the information in question is already cited in the body. If a claim is not sourced or mentioned in the body, it should be sourced in the lead, added to the body with a source, tagged with {{citation needed lead}}, or removed. Epicgenius (talk) 21:15, 26 August 2024 (UTC)[reply]
    I'd definitely qualify that as falling under "good reason"! When I say "without a good reason" I'm just reiterating Wikipedia:STYLEVAR here, which is the current guidance and therefore policy. – Closed Limelike Curves (talk) 00:12, 27 August 2024 (UTC)[reply]
  • Comment: That the "current policy" is "editors may do either, but should not switch without a good reason" is disputed, as MOS:CITELEAD nowhere actually says so. Gawaon (talk) 18:56, 27 August 2024 (UTC)[reply]

No change, leave MOS:CITELEAD as it is

  1. Support. None of the other options makes sense under all circumstances, and no change is needed. Gawaon (talk) 21:17, 26 August 2024 (UTC)[reply]
  2. Support. I think a clarification of the MOS:CITELEAD guideline might be warranted (see my comment under "Editors may do either, but should not switch without a good reason"), but I don't think either requiring or banning lead citations would be helpful at this time. Epicgenius (talk) 21:25, 26 August 2024 (UTC)[reply]
  3. Support. Current wording provides adequate guidance. No change is needed. —David Eppstein (talk) 21:42, 26 August 2024 (UTC)[reply]
  4. Support clearly this is the communal norm.Moxy🍁 21:55, 26 August 2024 (UTC)[reply]
  5. Support It ain't broken. XOR'easter (talk) 03:43, 27 August 2024 (UTC)[reply]
  6. Support But more obviously the primary instance I have seen of citations in leads becoming unwieldy is when there is contentious content. At one point I think it was Gamergate had 17 citations for one simple assertion that Gamergate "harassed" some people because of bad-faith arguments and persistent edit warring. Adding all the citations was overkill - but it made the arguments in defence of that sentences inclusion ironclad. As such I feel there is value to allowing the inclusion of some sources - but they should always be used and included within the article also. Orphan lead citations I disapprove of. Koncorde (talk) 23:25, 28 August 2024 (UTC)[reply]

Discussion

How about SNOW-closing this? An RfC whose OP opposes all suggested options obviously won't go anywhere (leaving aside the fact that none of the options makes much sense). Gawaon (talk) 20:20, 26 August 2024 (UTC)[reply]

If you want to propose some other option, feel free to add it. I initially opposed all because I don't care that much, but I've edited my comment to support the option I think makes most sense. – Closed Limelike Curves (talk) 20:25, 26 August 2024 (UTC)[reply]
@Gawaon I have no objections to including additional options, but duplicating an existing option just confuses people. I've merged both options, but I don't like the confusion this might lead to. – Closed Limelike Curves (talk) 22:38, 26 August 2024 (UTC)[reply]

I don't really see this leading directly to a change in this guideline, but it might be interesting to see what the community thinks overall and see whether some adjustments should be considered. There are editors who believe that "one sentence, one citation" is the ideal, and they might support a change. There are also editors who think that the fewer rules we have, the better.

In case the current state of citation density interests anyone, based on some work done recently by BilledMammal, the middle half of our articles, sorted by length, have about 5–30 sentences, and the middle half when sorted by number of refs have about 2–10 inline citations (about 35% of articles meet both of these conditions). We're running around one ref per four sentences for all articles, and one ref per 3.5 sentences for the middle-ish articles. Stubs (when defined as ≤250 words [mean of 110 words] or as ≤10 sentences) tend to have one ref per two sentences. WhatamIdoing (talk) 21:17, 26 August 2024 (UTC)[reply]

As if the RFC were not already dubious and compromised enough, in Special:Diff/1242462087, Closed Limelike Curves attempted to refactor it into three questions instead of four, merging two groups of questions that had already received separate answers. CLC: Do not change the questions in active RFCs. That is not how RFCs work. —David Eppstein (talk) 22:53, 26 August 2024 (UTC)[reply]
As a note, I tried to clarify choices 3 and 4 are the same, with both referring to current policy (MOS specifies both styles are permissible). I'm of the opinion the confusion caused by duplicate options is more dubious than just combining them, but if you disagree, I'm fine with leaving it. – Closed Limelike Curves (talk) 00:30, 27 August 2024 (UTC)[reply]
That is not how RFCs work. ...but it might be how RFCs happen all the time. @Gawaon boldly changed the third one after two people had replied there, seemingly to draw a distinction between "current policy" and "no change", or perhaps to indicate a belief that the current rules aren't that editors may do either (or perhaps that they shouldn't switch without good reason?). The practical difference between those two is unclear to me. WhatamIdoing (talk) 01:11, 27 August 2024 (UTC)[reply]
Woah... I thought something was off with my question title, but I figured I must have been misremembering what I wrote somehow.
@Gawaon If you object to the characterization of this as current policy that's fine. However, in the future, please provide a ping and explanation of the change (like I did when I merged the two), so someone can dispute this. I'd be happy to see a proposal from you clarifying exactly what you think the current policy says; given both I and @WhatamIdoing seem to think the description I gave of current policy is accurate, but you think it's inaccurate, I think it's pretty clear there's some kind of ambiguity here. – Closed Limelike Curves (talk) 02:32, 27 August 2024 (UTC)[reply]
You had explicitly written "If you want to propose some other option, feel free to add it", so that's what I did. MOS:CITELEAD says "editors should balance the desire to avoid redundant citations in the lead with the desire to aid readers in locating sources for challengeable material", which is far more nuanced than "Editors may do either, but should not switch without a good reason". Hence I don't think your third option is an adequate summary of the status quo. If you meant it as "don't change MOS:CITELEAD at all", well ... in that case my apologies to you, but you certainly didn't write that, and I didn't read it as such. Hence I added "No change" as a fourth option since it's the only one I can really support. And from the comments this RfC has attracted so far, other editors seem to think the same. Gawaon (talk) 06:36, 27 August 2024 (UTC)[reply]
I think the concern is that you not only added "some other option", but you also "changed an option people had already voted on", potentially changing how other people would interpret their votes. WhatamIdoing (talk) 17:05, 27 August 2024 (UTC)[reply]
I merely changed "Current policy (editors may do either, but should not switch without a good reason)" to "Editors may do either, but should not switch without a good reason" and considering that "Editors may do either, but should not switch without a good reason" is very clearly NOT the current policy (which doesn't address the question of switching at all), I think that change was indeed called for to prevent confusion. The only person, besides Closed Limelike Curves themselves, that had supported that option at that time was Epicgenius. They later change their position to "Weak support" and added their support for the "No change" option, indicating that "Current policy" rather than "Editors may do either, but should not switch without a good reason" had supposedly been their preferred position all along. Gawaon (talk) 18:10, 27 August 2024 (UTC)[reply]
I absolutely wrote that, as @WhatamIdoing showed. My original title explicitly described the option clearly and unambiguously as maintaining current policy. It included a description of what this policy is as well, unlike the newly-inserted option, which does not adequately summarize current policy (contra WP:RfC guidelines, which state each choice should be adequately summarized and explained).
From the comments this RfC has attracted so far, other editors seem just as confused by this. The question just asks:
Should citations be included for most claims in an article's lead, just like in the body?
In other words, both options are the same with regards to the question being asked. Both involve maintaining the status quo that citations are optional, and there are no situations where they are prohibited (which by MOS:STYLEVAR and other guidelines means users should not attempt to remove them without consensus). – Closed Limelike Curves (talk) 18:05, 27 August 2024 (UTC)[reply]
If you really believe that "Editors may do either, but should not switch without a good reason" is the current policy, then I challenge you to prove it by citing the exact sentence(s) in MOS:CITELEAD that say so. Gawaon (talk) 18:12, 27 August 2024 (UTC)[reply]
Although the presence of citations in the lead is neither required in every article nor prohibited in any article, there is no exception to citation requirements specific to leads. The necessity for citations in a lead should be determined on a case-by-case basis by editorial consensus.
– Closed Limelike Curves (talk) 18:26, 27 August 2024 (UTC)[reply]
If you'd like to propose a change exempting this one specific guideline from the declaration at [[MOS:]] that:
Where more than one style or format is acceptable under the MoS, one should be used consistently within an article and should not be changed without good reason.
Then you're free to propose doing so as an additional option. – Closed Limelike Curves (talk) 18:51, 27 August 2024 (UTC)[reply]
Ah, so you admit that MOS:CITELEAD doesn't say so itself, but you take it from elsewhere in the MOS. However, note that the presence or absence of citations is not a "stylistic choice", so that rule doesn't apply here. Gawaon (talk) 19:01, 27 August 2024 (UTC)[reply]
I did not say that. If the Manual of Style explicitly states "you can do either this or that", that's clearly a stylistic choice. I cited the exact lines clearly stating that this is a stylistic choice.
Although the presence of citations in the lead is neither required in every article nor prohibited in any article, there is no exception to citation requirements specific to leads. The necessity for citations in a lead should be determined on a case-by-case basis by editorial consensus.
If you believe this policy is not clear in this way, you may propose a change in the phrasing that would clarify this specific section is exempt from MOS:STYLEVAR. Otherwise, I'm done replying to WP:ICANTHEARYOU. – Closed Limelike Curves (talk) 21:47, 27 August 2024 (UTC)[reply]
MOS:STYLEVAR says: "When either of two styles is acceptable it is generally considered inappropriate for a Wikipedia editor to change from one style to another unless there is some substantial reason for the change." Having citations or not is not a question of style. I rest my case. Gawaon (talk) 22:00, 27 August 2024 (UTC)[reply]
If you believe that the location of citations is not a question of style, then perhaps you will consider making a separate proposal to remove CITELEAD from WP:Manual of Style/Lead section. Any time a time a rule is presented in a page that actually says "Manual of Style" in its title, I think it's fair for editors to assume that it's "a question of style".
Additionally I'm not sure whether you object to:
  • Editors may do either, or
  • but should not switch without a good reason, or
  • both.
Which one(s) of these do you think is not the community's current rule and practice (regardless of where that rule may or may not be documented)? WhatamIdoing (talk) 22:55, 27 August 2024 (UTC)[reply]
I don't think any change is needed, because I don't think the MOS is only about what it refers to as "stylistic choices" in its lead, while STYLEVAR is precisely about such merely stylistic choices – such as the use of DMY vs. MDY dates. The MOS also has a chapter on images, but that doesn't mean that STYLEVAR also applies to the use or non-use of images. Consider: "Editors may or may not add images to articles, but should not switch from one style to the other without a good reason, hence they normally shouldn't add images to articles that don't yet have them." Would you consider that a good summary of our image policy? I for one, would not, despite that fact that that chapter lives indeed in the MOS. Gawaon (talk) 06:12, 28 August 2024 (UTC)[reply]
Has anyone made a decision to not include any images at all in any article, even if a good image became available? If not, then we can ignore that as a strawman.
I've seen editors decide to not include an image in the lead, and I'd usually classify that as a stylistic choice.
I've also seen editors object when someone switching the image from one to another similar image without a good reason. I think that's also a stylistic choice. WhatamIdoing (talk) 06:28, 28 August 2024 (UTC)[reply]
@Gawaon your comments are conflating the inclusion/non-inclusion of citations in an article as a whole (not a stylistic choice) and where exactly they should be located within the text (a stylistic choice). – Closed Limelike Curves (talk) 20:29, 28 August 2024 (UTC)[reply]
It's not merely a stylistic choice, as myself and others have already said. Like, when summarizing a long section with 37 references in a single lead sentence and you add just 3 references to that sentence, then you're potentially misleading your readers, since those 3 references might be insufficient to back everything in the summary. But if you tackle all 37 references to that sentence, unreadability ensures. So the best choice, and not at all stylistic, in such cases is to omit the references in the lead summary altogether, implicitly pointing to where they can actually be found, namely the body. Gawaon (talk) 20:39, 28 August 2024 (UTC)[reply]
@WhatamIdoing: Would it be helpful to get the statistics limited to the lede of articles? BilledMammal (talk) 13:30, 27 August 2024 (UTC)[reply]
@BilledMammal, I think it would be difficult to reliably differentiate a stub from a lede, but if you want to have a go at it, I'd love to see the numbers. Perhaps it would be possible to filter for articles that contain a ==Section== with at least two non-list/non-template sentences in it? (Or some roughly equivalent number of words, if that's easier to calculate?)
If you were to run the first set on just, say, 50 or 100 articles, I'm willing to manually look at them all to see how the filtering is doing. WhatamIdoing (talk) 16:10, 27 August 2024 (UTC)[reply]

Max length guideline similar to the total length guideline

We're often getting instances where a lead is kept to 4 paras, but is almost 700 words e.g. [1]. It's worth strengthening the lead guidance to give a suggested max, like we have for total length: 15,000 words. We state most featured articles have up to 400 words, but this isn't keeping leads to a reasonable length. Surely adding a guideline very similar in wordage to the total length guideline, e.g. "More than 500 words means it almost certainly should be trimmed," would help. What do people think? Tom B (talk) 10:47, 28 August 2024 (UTC)[reply]

I agree such a recommendation would be good to have. My own rule of thumb is to strive for 225–450 words in the lead. Gawaon (talk) 12:49, 28 August 2024 (UTC)[reply]
@Tpbradbury, see Wikipedia talk:Manual of Style/Lead section/Archive 23#FA numbers and Wikipedia talk:Manual of Style/Lead section/Archive 22#Seeking consensus for table modification and probably some others.
In general, I think editors are willing to remove the paragraph count from the table and replace it with a word count. Something approximately in the 200–500 range is suitable for a longer article. Few leads, even on a Start-class article, should be less than 100 words.
However, I think there are two other statements they'd agree with:
  • Most articles should have four or fewer paragraphs.
  • Changing the number of paragraphs by adding/removing line breaks is not the point of this guideline. Having 700 words across five paragraphs is not automatically worse than 700 words across four longer paragraphs. What's usually needed is to not have 700 words.
WhatamIdoing (talk) 17:27, 28 August 2024 (UTC)[reply]
Also, I don't know if this has been discussed before, but I'm not sure that article length is actually the best indicator of lead length. It makes some intuitive sense, but a short article about (e.g.,) a disease needs the same information in the lead (e.g., definition, symptoms, treatment, epidemiology, prognosis) as a long article about the same subject. A long list might need a short lead ("This is the list of widget styles. There are ___ significant variations in widget styles. Blue-green widgets are widely agreed to be the most commercially important category" – a mere three sentences/27 words, but not much else is needed, regardless of whether the number in the blank is 15 or 5,000). Subject complexity might be a better indicator of lead length than the number of words it takes to write the article. Bangladesh is given as an example above, and it seems like an article about a country is a complex subject that would need a lot of room to fully summarize. WhatamIdoing (talk) 18:08, 28 August 2024 (UTC)[reply]


Cite error: There are <ref group=upper-alpha> tags or {{efn-ua}} templates on this page, but the references will not show without a {{reflist|group=upper-alpha}} template or {{notelist-ua}} template (see the help page).

Retrieved from "https://en.wikipedia.org/w/index.php?title=Wikipedia_talk:Manual_of_Style/Lead_section&oldid=1242823911"