Essays Low‑impact

This page is within the scope of WikiProject Wikipedia essays, a collaborative effort to organise and monitor the impact of Wikipedia essays. If you would like to participate, please visit the project page, where you can join the discussion. For a listing of essays see the essay directory.EssaysWikipedia:WikiProject EssaysTemplate:WikiProject EssaysWikiProject Wikipedia essays pages Low This page has been rated as Low-impact on the project's impact scale. The above rating was automatically assessed using data on pageviews, watchers, and incoming links.

Mentions

• Wikipedia:Motto of the day/January 25, 2014
• Wikipedia:Tip of the day/November 24, 2015

It is time to launch this essay I started working on in January 2011. I hope that it will inspire editors to think about how they create and edit leads. A lead can make or break an article by inspiring or discouraging editors to/from reading further. This works both ways because thinking systematically about the organization of the lead can help editors discover faults in the article and guide them in rearranging and improving the article itself.

I hope you enjoy this essay and find it interesting. I'd appreciate feedback here, but also on my talk page, since I have over 8,000 articles, plus their talk pages, on my watchlist. Comments on my talk page will be noticed. You can also get my attention anywhere by pinging me. Start your comment with this code if you want my immediate attention: ((ping|BullRangifer)). -- Brangifer (talk) 05:33, 21 September 2013 (UTC)[reply]

I would like to invite interested parties to comment on this essay:

1. Do you think it is in harmony with the guideline?
2. Do you think there is anything which contradicts the guideline?
3. Is there anything in the essay which would be good to include in the guideline?

If it can be improved, I'd like to do that, so your thoughts will be appreciated. -- Brangifer (talk) 08:33, 29 November 2013 (UTC)[reply]

→ We ascribe beauty to that which is simple; which has no superfluous parts; which exactly answers its end; which stands related to all things; which is the mean of many extremes.

The lead section of a Wikipedia article is the section before the first heading. The table of contents, if displayed, generally appears between the lead section and the first subheading.

Rule of thumb: If a topic deserves a heading or subheading, then it deserves short mention in the lead.

The lead section should contain up to four paragraphs, depending on the length of the article, and should provide a preview of the main points the article will make, summarizing the primary reasons the subject matter is interesting or notable. The lead should be capable of standing alone as a concise overview of the article, should be written in a clear and accessible style, should be carefully sourced like the rest of the text, and should encourage the reader to want to read more. The following table has some general guidelines for the length of the lead section:

The Rule of thumb has been modified:

• "If a topic deserves a heading, then it deserves short mention in the lead according to its real due weight."

Mentioning every subsection in the lead can be too much. -- BullRangifer (talk) 17:17, 25 June 2019 (UTC)[reply]

Moved from Talk:Donald Trump#Leadrefs

Pings:

• User:Andrevan
• User:Space4Time3Continuum2x
• User:Zaathras
• User:Khajidha
• User:Bob K31416
• User:Mandruss
• User:Slatersteven

Valjean (talk) (PING me) 19:05, 15 October 2022 (UTC)[reply]

We could avoid all these problems if we adopted my desired method of writing a lead with Lead "section references". See Wikipedia:How to create and manage a good lead section#Lead "section references" and see how they work in the lead there. They are added to most sentences or phrases in the lead and take one directly to the section on which that lead content is based, right where the citations are located. -- Valjean (talk) (PING me) 18:50, 12 October 2022 (UTC)[reply]

If you think it's good, do it. Bob K31416 (talk) 22:30, 12 October 2022 (UTC)[reply]

It would first need to be adopted as part of our PAG. I'm not even sure where to propose the idea for community input. -- Valjean (talk) (PING me) 22:56, 12 October 2022 (UTC)[reply]

It's not an entirely bad idea, although it would increase the workload in an active article like this one. Section headings change, and editors rarely add the code to retain the old heading as an alternate target. There would always have to be people watching for broken "#'s" in the lead and fixing them. Unlike page links, section links do not turn red when they're broken; the only way to identify a broken section link is to click on it and see what happens.

A template would be in order for ease of coding. This would reduce <sup>[[#First impeachment|#]]</sup> to ((leadref|First impeachment)).

We have historically had trouble making sure that all lead content is backed up by body content. A solution like this might make that easier, as the absence of a leadref would be a flag that someone should check the body (then add a leadref, add body content and a leadref, or remove the lead content).

Since the "#" has only essay status, we could consider using section sign instead. This would be consistent with Template:Section link, as well as looking better in my opinion.

If eliminating clutter in the lead is a main goal, I'm not sure that a bunch of #'s or §'s would be significantly less clutter than a bunch of [n]'s. But that isn't the only goal.

I don't think you necessarily need community permission to do this. PAGs are supposed to arise from common practices, not the other way around, and common practices always begin with a single use case. Just Do It and see who complains. (FKA 68.97.42.64, FKA Mandruss) 161.97.225.237 (talk) 03:01, 13 October 2022 (UTC)[reply]

Thanks for understanding my intent and for the constructive suggestions. Is there a project where experts might make a serious attempt to develop and improve a testable way to do this? I need to move this away from this page. -- Valjean (talk) (PING me) 03:11, 13 October 2022 (UTC)[reply]

I agree with Mandruss: go ahead and do it. Andre🚐 03:12, 13 October 2022 (UTC)[reply]

I'm not sure what you mean by "testable". There's no question it would work from a technical standpoint, as we know section links work and we see this working in the essay. The only "experts" you need are for creating the template, and that could wait a little while to see if the concept flies at all without a template. I'd add the first one to get it started, but since my "retirement" I no longer edit AP2 articles and try to be logged out for any edits. (FKA 68.97.42.64, FKA Mandruss) 161.97.225.237 (talk) 03:53, 13 October 2022 (UTC)[reply]

I'd be happy to help you set up the template @Valjean, it should be pretty easy. I set up a basic proof-of-concept here, let me know if that's what you had in mind. Andre🚐 04:38, 13 October 2022 (UTC)[reply]

@Andrevan: For what we have in mind, see the essay that Valjean linked in his opening comment. We don't want something that looks like a citation number when it's not one. And the lead content will often summarize information from multiple sources cited in the body, often non-adjacent in the target section. Furthermore, your technique would require named references in the body. As it is, we'll have some cases summarizing body content from multiple sections, requiring multiple leadrefs in the lead; let's not make that worse by requiring a separate template for each citation. I suggest a template that does what I described above with the coding examples. (FKA 68.97.42.64, FKA Mandruss) 161.97.225.237 (talk) 05:02, 13 October 2022 (UTC)[reply]

The one as used in the example above is extremely simple: <sup>[[#(({1|))}|#]]</sup> It's at User:Andrevan/Simple_Leadref, feel free to move or copy or edit as you would like. I was trying to come up with a way to automatically have it figure out which ref it is in sequence if you haven't named the refs. It can probably be done with a Lua module. Andre🚐 05:09, 13 October 2022 (UTC)[reply]

Not sure what is meant by I was trying to come up with a way to automatically have it figure out which ref it is in sequence if you haven't named the refs. Your Simple_Leadref is the ticket, but we shouldn't use it in the article from user space. To get it into template space requires an editor with the template editor user right, and doing that before the concept is accepted in actual practice would be cart before horse (partly because of the extra effort required to delete a template). Just add a few without the template and see if it's accepted. If so, add the rest. Then wait a few weeks before having some authorized editor add the template in template space, and then convert everything to use it. (FKA 68.97.42.64, FKA Mandruss) 161.97.225.237 (talk) 05:25, 13 October 2022 (UTC)[reply]

Yes I agree we shouldn't use the templates from userspace, I was merely demonstrating how it works, and feel free to change it around. Yes someone with rights would have to create it in mainspace, and I agree you could simply do it by hand for now. I agree generally with your idea of how this could be adopted. All I meant about having it calculate the sequence would be to number the links. If you use Wikipedia's built-in linking or references it numbers them for you: [1] [2] ^[1][2]. I couldn't quite figure out how that works though I'm sure there is a way to do it, possibly through hacky text processing, so I just made the number a parameter for now. I borrowed the section sign idea as well for my example. Using the User:Andrevan/Leadref approach, the target is a reference superscript itself, which is how the backlinking works from ((reflist)). This superscript ^⇣2 links to the reference superscript named bar, but it also works for unnamed references as well ^⇣1. The trick is figuring out how to automatically update the numbers for named refs. I guess you can also call it a refref vs a sectionref^⇣999 Andre🚐 15:19, 13 October 2022 (UTC)[reply]

I am unsure this will work, as you say in the thread above they will still object to its inclusion. Slatersteven (talk) 10:54, 13 October 2022 (UTC)[reply]

In that regard, it may be useful to hear from User:SPECIFICO, User:Space4Time3Continuum2x, and User:Khajidha. Bob K31416 (talk) 15:42, 13 October 2022 (UTC)[reply]

I just tested <sup>[[#First impeachment|§ he was impeached]]</sup> and <sup>[[#First impeachment|# he was impeached]]</sup> (§, #). It adds the § and # symbol, respectively, and if you hover over the link you see the section it links to. Either one would require an explanation at the top (Click § to view article section, or s.th. similar). I'd be all for it. Space4Time3Continuum2x (talk) 17:30, 14 October 2022 (UTC) Valjean, have you seen this RfC? Space4Time3Continuum2x (talk) 18:39, 14 October 2022 (UTC)[reply]

Did you notice that the text "he was impeached" was superscripted? Not acceptable, obviously. And removing the superscripting would put a non-superscripted symbol in the middle of the prose, impeding readability. I don't know why people keep trying to deviate from the technique shown in the essay, which works fine. (FKA 68.97.42.64, FKA Mandruss) 161.97.225.237 (talk) 23:07, 14 October 2022 (UTC)[reply]

Space4Time3Continuum2x, sorry for the delay in replying. I was not aware of that RfC. Thanks for the heads up. Here's the opening of the closing decision: "There is a consensus that self-links within prose should be allowed and that linking should be based on editorial discretion." I agree with that decision. There is a sentence at the end of the lead in the Steele dossier article which does that: "The dossier is a factor in several conspiracy theories promoted by Trump^[3]" A slightly related, but different, type of example is found in that article, where there are links between specific Steele dossier#Allegations sections and the relevant specific Steele dossier#Veracity and corroboration status of specific allegations sections, making it easier for readers to learn what RS say about the status of each allegation, which is often a mixture of opposing opinions and interpretations (when RS disagree, we are supposed to document both sides), but sometimes a clear "confirmed", "unconfirmed", or "disproven" exists. (I use synonyms right there.) Whenever RS appear with new information of relevance, that content is updated. (In spite of that, there are fringe editors who still insist on applying some synonym of "unproven" or "disproven" to the whole dossier and to all allegations, which is false and contrary to the treatment in RS, because a number of allegations are confirmed true.) This subject of "self-links within prose" is a tangentially related topic to this thread, so, to not create a distraction, I'll end my reply here. If we need to continue, maybe we should move this to a new thread. -- Valjean (talk) (PING me) 15:47, 15 October 2022 (UTC)[reply]

In April, as an alternative to citations, I replaced some of the links in the lead with links to the respective body sections and got reverted pronto. One of the arguments at the time was that "readers would probably expect to be taken to another article entirely." How do we know that? Has there been a poll, not to mention a reliable poll, of readers’ expectations? I don’t remember what I expected — other than more info — when clicking on an inline link before I started editing on WP. (Also, quite a few of the articles on Trump have a quality problem.) Another argument was not wanting "people confused if sometimes they click a link and it goes to another article, and other times they click a link and just jump around in the same article." Are WP readers unfamiliar with the revert arrow in their browsers?

MOS is style, not policy (It is a generally accepted standard that editors should attempt to follow, though it is best treated with common sense, and occasional exceptions may apply.) And we shouldn’t presume that WP readers are easily confused or unfamiliar with the revert arrow of their browsers, i.e., less competent than us genius editors. In a lead and an article of this size, I think the lead should be a summary of article content without citations or inline links to other articles - want more info, look at the table of contents and/or click on the inline link to read the section in the body with the citations and the links to other articles (yeah, when pigs fly). The lead, not just in this article, seems to be turning into an article in its own right. Space4Time3Continuum2x (talk) 16:16, 14 October 2022 (UTC)[reply]

That's a good reason to avoid a link that looks anything like a link to another article. A lone superscripted symbol (^§) would have only one use, to link from the lead to a section heading in the same article. I don't think an "explanation at the top" is necessary. I think most readers would figure it out by clicking on one of the symbols, and in my experience they rarely take the time to read explanations at the top anyway. With any luck, the usage would eventually become widespread enough that readers could transfer that knowledge to other articles, particularly the more contentious ones. We might add something about it to the FAQ on this page, but even that would have questionable value. (FKA 68.97.42.64, FKA Mandruss) 161.97.225.237 (talk) 23:07, 14 October 2022 (UTC)[reply]

I agree that leads should not link to other articles, and I and others have said so in the past. That tends to encourage readers to bypass the body content, raising the question of what it's for. Readers should be steered to our body first, and only then to other articles if they want even more detail. When combined with the lone-symbol leadref concept described here, a pleasant side effect is the elimination of all blue prose in the leads, further improving readability. (FKA 68.97.42.64, FKA Mandruss) 161.97.225.237 (talk) 00:23, 15 October 2022 (UTC)[reply]

Mandruss, that's so well said. I love it. As much as possible, it should be "lead --> body --> other articles, but we can't do it completely. That makes so much sense. Using discrete leadref symbols would indeed help to eliminate many, but not all, blue links in the lead. There is a difference between a blue wikilink and "non-blue" lead content with a leadref symbol after it. We're aiming to reduce the number of refs in the lead. -- Valjean (talk) (PING me) 15:59, 15 October 2022 (UTC)[reply]

I misunderstood and did this: <sup>[[#First impeachment|§ he was impeached]]</sup>. Is this the clickable section reference Valjean’s proposal is referring to: he was impeached<sup>[[#First impeachment|§]]</sup>? I’m not quite clear on the concept of where to add the symbol. The sentence in the example has six links to other pages (and I just noticed that two are MOS:SOB). The two linking to "event" pages are no problem but what about the other four? "Abuse of power" and "obstruction of Congress" are also mentioned in First impeachment, of course, so the links could be located there, but what about Biden and the House of Representatives? (The specific charges could also be replaced by "two charges", IMO.)

After he pressured Ukraine to investigate<sup>[[#First impeachment|§]]</sup> Biden in 2019, he was impeached<sup>[[#First impeachment|§]]</sup> by the House of Representatives for abuse of power<sup>[[#First impeachment|§]]</sup> and obstruction of Congress<sup>[[#First impeachment|§]]</sup> in December.

"After he pressured Ukraine^§ to investigate Biden in 2019, he was impeached^§ by the House of Representatives for abuse of power^§ and obstruction of Congress^§ in December." Space4Time3Continuum2x (talk) 13:48, 15 October 2022 (UTC)[reply]

I really like this idea and this specific implementation. --User:Khajidha (talk) (contributions) 17:24, 15 October 2022 (UTC)[reply]

Replying to Khajidha: I like it, too, and would be willing to give it a try. Space4Time3Continuum2x (talk) 17:48, 15 October 2022 (UTC)[reply]

Has this format been used in any other live article yet? Zaathras (talk) 13:58, 15 October 2022 (UTC)[reply]

Not that I know of. -- Valjean (talk) (PING me) 16:15, 15 October 2022 (UTC)[reply]

Replying to Zaathras: If I understood editor 161.97.225.237 FKA Mandruss correctly, we would be groundbreakers. Space4Time3Continuum2x (talk) 17:48, 15 October 2022 (UTC)[reply]

An important concept to make this work best is my belief in a tight/mirrored connection between each part of the lead and each section in the body, IOW mirroring the ORDER each appears in the body. The lead should mirror the body, also in the order in which things are mentioned. Wikipedia:How to create and manage a good lead section describes this. This concept IS NOT an absolute requirement for using leadrefs. I mention this just as an FYI. My essay deals with much more than just the lead or just leadrefs. -- Valjean (talk) (PING me) 16:15, 15 October 2022 (UTC)[reply]

Ok, groundbreakers. Is that how things should work, though? I'd think a new referencing system would start somewhere at the Village Pump rather than an article talk page that is only frequented by a handful of politics-oriented editors. There's also the "why" to consider, is this truly to inform the reader, or assuage an even tinier handful of gadflies who want every Trump-critical thing in the lede referenced? Zaathras (talk)

I have asked about a better place to do this. It has implications and usefulness in myriad articles. It just happened to start as a discussion here. -- Valjean (talk) (PING me) 18:09, 15 October 2022 (UTC)[reply]

Quoting Mandruss again: Readers should be steered to our body first, and only then to other articles if they want even more detail. Space4Time3Continuum2x (talk) 18:12, 15 October 2022 (UTC)[reply]

@Zaathras: Quoting myself: I don't think you necessarily need community permission to do this. PAGs are supposed to arise from common practices, not the other way around, and common practices always begin with a single use case. There is no community consensus that precludes this. If one wanted to kill this on the vine, the most effective way would be to go ask for prior community permission. You'll get more community support if the community can see the concept working in actual practice, rather than discussing things in the abstract.

I don't even know why this needed to be moved off the Trump talk page. For now, it's about improving that article, which is the purpose of that talk page. (FKA 68.97.42.64, FKA Mandruss) 161.97.225.237 (talk) 19:31, 15 October 2022 (UTC)[reply]

Feel free to unhat it at Talk:Donald Trump if that's better. See my comment below. -- Valjean (talk) (PING me) 19:42, 15 October 2022 (UTC)[reply]

I did the deed, per Just add a few without the template and see if it's accepted. What's the worst than can happen? I get reverted, and it's not as if that would be a new experience. Space4Time3Continuum2x (talk) 18:07, 15 October 2022 (UTC)[reply]

"(Replacing Wikilinks with section references — see discussion at Talk)" That's an application of Mandruss's idea about fewer blue links, whereas my idea of leadrefs is to substute them for actual citation references. These are two different concepts. -- Valjean (talk) (PING me) 18:16, 15 October 2022 (UTC)[reply]

This is premature, I think. The idea was to try this out locally to gauge the reaction before taking it WP-wide or not. If you don't mean s.th. similar to the clickable "§", what kind of clickable "section references" are you suggesting? Space4Time3Continuum2x (talk) 19:24, 15 October 2022 (UTC)[reply]

This has nothing to do with trying it out locally. That can be done anywhere, and right now you have tried it at Donald Trump, which is great. Here I'd like to see development and improvement, maybe using templates like Andre is developing at User talk:Andrevan/Sectionref. My original idea here seems to work, but maybe another way might work better? The discussion has been good, so continue. It was just off-topic at the talk page, and others requested it be closed or moved, so that's what I've done.

It can also be unhatted there, but then limited to what's relevant to the Trump article. I'm not sure if I'm making myself clear, but do whatever works. -- Valjean (talk) (PING me) 19:37, 15 October 2022 (UTC)[reply]

Space4Time3Continuum2x, I'm referring to the clickable "§" that replace citations/references in the lead, as I do at the lead for this essay. They take one to sections, which is why I call them "lead section references" or "leadrefs". Using them to replace blue wikilinks is something else, but maybe both ideas will work. This all started because of objections to all the citations/refs in leads. We want to avoid them as much as possible. Doing this to replace blue wikilinks in the lead is a totally new and different idea that has been piggybacked onto my idea, which is fine. -- Valjean (talk) (PING me) 19:48, 15 October 2022 (UTC)[reply]

I unhatted the last part of the discussion at the Donald Trump Talk page so that editors looking at the lead will know what went on there. Space4Time3Continuum2x (talk) 19:54, 15 October 2022 (UTC)[reply]

That makes sense. -- Valjean (talk) (PING me) 20:04, 15 October 2022 (UTC)[reply]

Your clickable "§" is the exact same one it took my a while to understand (and which is now implemented in the last paragraph of the lead of Donald Trump, except for the last sentence — I didn't dare mess with consensus item 58:) Being able to eliminate the blue wikilinks is the pleasant side effect Mandruss mentioned. Two birds with one stone, and IMO the side effect is actually the bigger selling point. What would also be good is a button after each section title to take the reader back to the lead. Space4Time3Continuum2x (talk) 20:02, 15 October 2022 (UTC)[reply]

That's three birds! Go for it! -- Valjean (talk) (PING me) 20:13, 15 October 2022 (UTC)[reply]

Testing this and getting it accepted on the Trump article will likely ensure acceptance everywhere. -- Valjean (talk) (PING me) 20:15, 15 October 2022 (UTC)[reply]

If I knew how to write code I'd probably be making a lot more money than my employer is paying me. I'm afraid the button is way beyond me. Space4Time3Continuum2x (talk) 21:58, 15 October 2022 (UTC)[reply]

I suspect Andre can do it. -- Valjean (talk) (PING me) 22:02, 15 October 2022 (UTC)[reply]

Yes that can be done, check out ((User:Andrevan/Back to Top)):

🔝

Andre🚐 23:10, 15 October 2022 (UTC)[reply]

On my platform, Windows/Firefox, I can easily get "back to top" with one mouse movement, one keystroke, and one mouse click. It works for a scrollable page of any type, Wikipedia or otherwise, so 90%+ of readers already know how to do it. I probably do it a hundred times a day. Is there a platform that makes that significantly more difficult? If not, it's hard to justify the additional clutter. You might as well add a "Scroll Down" button. Avoid reinventing wheels and providing redundant functionalities. (FKA 68.97.42.64, FKA Mandruss) 161.97.225.237 (talk) 05:42, 16 October 2022 (UTC)[reply]

This isn’t about reinventing the wheel, it’s about putting a tire on it that will make the ride less bumpy. "§" is a "jump to" button, so why not add a "jump back to top" button, ideally on the right side of the section heading line. I don’t know how the equivalent of one mouse movement, one keystroke, and one mouse click would work for the dab and swipe folks when they're way down at the end of a long article - the closest I’ve come to a smartphone is a large tablet, and I avoid using it as much as I can. I added a couple of "back to top" buttons here and here. They work but I had to add them to the heading, so they showed up in the TOC. Unforeseen consequence: when I click the button in the TOC, it links to the body section, so I deleted them. (I probably should have foreseen that, it being the TOC.) Space4Time3Continuum2x (talk) 11:22, 16 October 2022 (UTC)[reply]

I'll take a look and see how it could work better in a heading. Andre🚐 16:52, 16 October 2022 (UTC)[reply]

If a "back to top" function is accepted, it should be part of the wiki default settings when creating a section heading, IOW made by Wikimedia programmers. Right now the wiki default for a section heading is with a blank line below the heading and a blank space on each side of the wording in the heading. That means the heading should always have a blank line above and below it. That makes it much easier to find section headings when quickly scanning pages while editing. If there is only one blank line, it looks like any other blank line between paragraphs and one can whizz right on past it without noticing the heading. -- Valjean (talk) (PING me) 17:03, 16 October 2022 (UTC)[reply]

Yes, if we had the assistance of the developers, it would be trivial to append a back to top button to the div .mw-editlinks that includes the [edit] button for the section, but then you'd need a way to enable it selectively. Probably the way we could hack it to work, so it doesn't show up in the TOC but just for the section header, would be to add it on the next line but then use css tricks to make it appear on the previous line. Andre🚐 17:38, 16 October 2022 (UTC)[reply]

I'm not sure if it's worth the effort at the moment, let's see first whether the section references will be accepted by the community. Space4Time3Continuum2x (talk) 18:09, 16 October 2022 (UTC)[reply]

Yes, no worries. I assume if everyone accepts this process, others who are more keyed in to the state of Wikipedia templates and development will have ideas and come up with more authoritative solutions. FWIW, though, I was able to add the negative margins and it does work for a proof of concept, check it out on User:Andrevan/Leadref Demo. Andre🚐 18:21, 16 October 2022 (UTC)[reply]

"but then you'd need a way to enable it selectively" Why would it be selectively enabled? Having this as a standard part of a section header would seem to be useful in and of itself. --User:Khajidha (talk) (contributions) 11:07, 17 October 2022 (UTC)[reply]

That's true Andre🚐 14:13, 17 October 2022 (UTC)[reply]

PLEASE DON'T add or re-add those to any page. They're visually distracting, to say the least. GoodDay (talk) 01:32, 20 October 2022 (UTC)[reply]

You're the only one of about 5 editors objecting to them. Andre🚐 01:36, 20 October 2022 (UTC)[reply]

I'm opening an RFC on this big change. It shouldn't be up to only five editors. GoodDay (talk) 01:38, 20 October 2022 (UTC)[reply]