User talk:Cessaune/Lead to body link proposal

Copy editing

Any objection to some copy editing? Nothing really substantive. Revertable of course. ―Mandruss ☎ 03:24, 1 November 2024 (UTC)[reply]

No. Cessaune [talk] 20:47, 1 November 2024 (UTC)[reply]

"Names and locations will remain standard wikilinks."

Strong oppose. Those names and locations can be linked in the body, just like everything else. The only exception to the "rule" I see is president of the United States, which has no related body content to link to. ―Mandruss ☎ 03:37, 1 November 2024 (UTC)[reply]

1) This proposal is bigger than Trump, so bear that in mind. 2) My philosophy is always to link to names and locations. Think about it in the context of bypassing the lead. If I'm clicking on a name I'm probably looking for information on that person specifically. Same goes for a location. We should only body link to events because those are the major things that are getting bypassed. Cessaune [talk] 20:29, 1 November 2024 (UTC)[reply]

So you would link Clinton, Gorsuch, Kavanaugh, Barrett, Kim, and Biden in the lead.

Locations? Most locations in the Trump lead are countries and U.S. states unlinked per WP:OL. University of Pennsylvania? U.S. Capitol?

I'm thinking cost (in added complexity) exceeds reader benefit. In any case, as a practical matter, I think "we" should be on the same page before taking the proposal to the community—else the proposal is not fully formed—and we need that reflected in the sandbox. Space4Time3Continuum2x and Valjean: What do you think? ―Mandruss ☎ 22:42, 1 November 2024 (UTC)[reply]

(I'm lost — where is the proposal?) Do "names" include political parties, organizations/institutions such as UPenn, Trump Organization, U.S. Capitol (which is both a name and a location); TV shows? MOS:UL says "proper names that are likely to be unfamiliar to readers" should be linked. Who defines "likely" and "readers"? Space4Time3Continuum2x 🖖 15:35, 2 November 2024 (UTC)[reply]

I would suggest that Amy Coney Barrett and U.S. Capitol are the type of things to just outright link. Cessaune [talk] 16:36, 2 November 2024 (UTC)[reply]

@Space4Time3Continuum2x: (I'm lost — where is the proposal?) The proposal, very rough at this time, is at User:Cessaune/Lead to body link proposal, the adjunct to this page. The heading of this section is a quote from that page. By "names", Cessaune means names of people. ―Mandruss ☎ 20:54, 2 November 2024 (UTC)[reply]

My goodness, don't recommend making section headings into page links. That idea was proposed before & the response was 'nope'. GoodDay (talk) 22:29, 3 November 2024 (UTC)[reply]

No idea what you're talking about. Nobody is suggesting making section headings into page links. The idea being developed here is quite different from what has been proposed before. You're opposing a proposal that hasn't even been made yet, and this is not the place. Unless you can contribute constructively to the development of the proposal, which is the purpose of this page, I "recommend" you wait until it's brought to the community and then either support or oppose. ―Mandruss ☎ 22:40, 3 November 2024 (UTC)[reply]

BTW, you made the same mistake back in January, here, and apparently learned nothing from that experience. Have to say it's getting annoying. ―Mandruss ☎ 23:02, 3 November 2024 (UTC)[reply]

Dark/light mode: we need to rectify this before we can move on.

We need to solve the light/dark ~~theme~~ mode (correct term) error. Ever since the official dark mode has come out, it's all I use, and black text is not okay on a dark gray background. This is the current state of the template, but my preferred design, this, is not feasible if we can't grab user's mode choice and adjust the color accordingly. Unless we all collectively agree to stick to a design that keeps the links colored normally, this needs to be figured out before we can proceed. I'll see what I can do. Cessaune [talk] 21:11, 3 November 2024 (UTC)[reply]

Again, how can I turn on this "dark mode"? ―Mandruss ☎ 21:23, 3 November 2024 (UTC)[reply]

Next to your username, there is a glasses icon. Click on it and it should be right there. Cessaune [talk] 21:25, 3 November 2024 (UTC)[reply]

I see no glasses icon. My username is all over the place; where are you referring to? ―Mandruss ☎ 21:28, 3 November 2024 (UTC)[reply]

Ah. It's only applicable to V22 and you're probably using V10? Cessaune [talk] 21:30, 3 November 2024 (UTC)[reply]

Lol. I haven't a clue what I'm using. But I followed the instructions at Wikipedia:Dark mode (gadget) and this added a "Dark mode" link at the top of the page. I guess that's the same thing. ―Mandruss ☎ 21:34, 3 November 2024 (UTC)[reply]

So I take it you're looking for a way to make the text white when dark mode is enabled, and, barring that, we're stuck with blue. Got it, I think. ―Mandruss ☎ 21:37, 3 November 2024 (UTC)[reply]

Yep. Exactly. It's possible with CSS, methinks...

There are three main "skins" that people use: Monobook, which was the default when I was still learning how to count, Vector 2010, which was the default for ages, and Vector 2022, which is the current default. The dark mode gadget in V10 (Vector 2010) works completely differently to the actual dark mode in V22 (Vector 2022), which means that it probably needs to be factored in also. Cessaune [talk] 21:39, 3 November 2024 (UTC)[reply]

I have managed to confuse myself again (it don't take much). I currently have dark mode enabled, and examples 2, 4, 6, and 8 at User:Cessaune/Lead to body link proposal#Technical have the text white. ―Mandruss ☎ 21:58, 3 November 2024 (UTC)[reply]

You may be onto something here. It seems that the dark mode gadget must force all black text styles to be white. This means that there might be a way to do the same with code. Cessaune [talk] 22:05, 3 November 2024 (UTC)[reply]

Seems to me you could just output what you have there already, and call it a day. ―Mandruss ☎ 22:07, 3 November 2024 (UTC)[reply]

That won't work for all use cases though, and it won't work for the default, which is a no-go. Cessaune [talk] 22:09, 3 November 2024 (UTC)[reply]

Ya lost me, starting with "what default"? ―Mandruss ☎ 22:10, 3 November 2024 (UTC)[reply]

The default skin Vector 2022, and its associated dark mode. Cessaune [talk] 22:11, 3 November 2024 (UTC)[reply]

The thot plickens. ―Mandruss ☎ 22:12, 3 November 2024 (UTC)[reply]

CHAT I'VE DONE IT at least for Vector 2022. Now I need to do it for Timeless (the skin)... and I think that the dark mode gadget takes care of it for V10. Cessaune [talk] 22:33, 3 November 2024 (UTC)[reply]

"CHAT"?

I'm still getting blue text in the sandbox while in dark mode. Don't know if that's what you expect. ―Mandruss ☎ 19:40, 9 November 2024 (UTC)[reply]

On Vector 2022, it works when in the dark mode and light mode, but not in the automatic mode. I have to figure out how to fix that. It currently does not work at all for Vector 2010 and Timeless. Something to do with the order of how styles load in. Just means it's become far more complex than it seemed initially.

"Chat" in this context is an indefinite pronoun born from the livestreaming era, where 'streamers' will refer to people watching their stream and commenting on their stream (people in the chat) as the singular term 'chat'. For example, a streamer might ask 'Chat, is this real?' and they would be asking the people watching to fact-check something. So when I use "chat" in this context, it's as if I'm talking to someone sitting right next to me, except they aren't sitting right next to me and I'm not talking to a distinct person: I'm talking to the air, and within the air are people such as yourself who are reading, making you a part of the "chat". Cessaune [talk] 23:30, 9 November 2024 (UTC)[reply]

Who woulda thunk something this simple would be so difficult. ―Mandruss ☎ 11:18, 11 November 2024 (UTC)[reply]

Lead-to-body links and lead citations

A benefit that I don't think has been mentioned in all this: Lblinks go a long way toward addressing concerns about the lack of citations in the lead. That's much less a concern if it's very easy for readers to find the related citations. You might add this to your case for lblinks. ―Mandruss ☎ 19:27, 9 November 2024 (UTC)[reply]

Timing

I'm thinking it will be at least March before things have quieted down enough to bring this proposal. His first weeks in office will be pretty much consuming editors' attention. No problem, that just gives us plenty of time to refine the proposal. ―Mandruss ☎ 19:34, 9 November 2024 (UTC)[reply]

Calling Rollinginhisgrave

@Rollinginhisgrave: Since you're a smart guy, I wonder if you have the time to look at this proposal and offer some feedback here. See the associated page and the illustrative sandbox. ―Mandruss ☎ 11:00, 11 November 2024 (UTC)[reply]

Thanks for the ping Mandruss. Some thoughts:

Is there a reason this isn't going through Wikipedia:Village pump (idea lab)?
This seems to duplicate some functionality of the table of contents.
If linked articles are too dense (e.g. Iran nuclear deal), the solution should be to apply the guideline Wikipedia:Make technical articles understandable: It is especially important to make the lead section understandable using plain language
If readers bypass the article's body by clicking on links in the lead, what even is the point of writing it? No comment on whether readers go beyond the lede, but a primary purpose of the lede is to summarize the body. It would be very hard to provide a summary of the topic without having first written a body.

Rollinginhisgrave (talk) 20:14, 11 November 2024 (UTC)[reply]

@Rollinginhisgrave:

Idea lab: We have four experienced editors involved, not including you. We think we have the competence to handle this. Too many cooks spoil the broth, as we see demonstrated over and over again at Wikipedia. A sea of broth has been spoiled over the years.
ToC: I spake the following on October 28 UTC, still on the Trump TP. I doubt I could say it any better.
Even if they choose not to read the body content, it's usually only one more click to reach the relevant other article. That effort may be compared to the effort of searching this massive table of contents for the body content elaborating on (and supporting) something you read in the lead. You think that's easy? Pretend you're new to the article and its ToC, forget everything you know about them, and try it for a few cases (no cherry picking). I think you'll find it's much harder than clicking a link in a hatnote at the top of a section you were just directed to. This equation may be different in shorter articles, which is why lead-to-body links should be nothing more than a local option; but they are sorely needed at at least one article—this one—and very likely others.
Linked articles too dense: Maybe, but that's an enormous job that probably won't be addressed in the foreseeable future. It's not the main point of the proposal.
Ok, so you've answered the question "What is the point of writing the body?" Try this one instead: "What is the point of writing the body's additional detail?" ―Mandruss ☎ 21:37, 11 November 2024 (UTC)[reply]

Idea lab: I see your point. It will need to go through WP:Village pump at some point to avoid being local consensus.
ToC: When you have some text, i.e. "Iran nuclear deal", you will only be able to link it to one of Donald Trump#Iran or Iran nuclear deal. Editors can access the latter by clicking through to Donald Trump#Iran and then finding a link within to Iran nuclear deal. But it still is creating a trade-off by making more work for editors who just wanted to go straight to Iran nuclear deal. If the ToC is accessible, there doesn't have to be a trade-off. Independent of intra-article links, the article and headings should be reorganized to ToC bloat and navigation difficulties. A small example of how this can be done: [1].
Linked articles too dense: I mentioned this because this proposal discusses Template:Main, so the simple summary should already in large part be already contained in this article.
Purpose of body: I think this is a particularly bad article to make that case on. If 1% of readers are going beyond the lede, that's still a quarter of a million readers per year.

I'll have a think about these issues and perhaps make some edits to the proposal to try to address them if I can. Your replies to these further comments would be valuable to that end. Rollinginhisgrave (talk) 22:29, 11 November 2024 (UTC)[reply]

@Rollinginhisgrave: I/we think you're misinterpreting LOCALCON for this case. There is no community consensus against lead-to-body links. We are permitted to innovate. See the consensus list, which didn't go through WP:Village pump; we Just Did It and it stuck; it has been a major success story and has been adopted at seven other articles at my last count (with no advocacy on my part). Opposition to change very often stifles evolutionary improvement and harms the encyclopedia.

If lead-to-body links were never adopted outside of Donald Trump, I would have no problem with that. It would take readers about half a minute to figure out what they're all about. Click on one and see what happens. Done. That said, I think there are plenty of large articles that would benefit.

I hope we can gain your support, as any change of this magnitude is going to be a hard sell regardless of the actual merits. ―Mandruss ☎ 23:46, 11 November 2024 (UTC)[reply]

The relevant guideline is Manual of Style/Linking: It is bad practice to create links in article text using the format Article#Section; navigation then becomes difficult if the section is expanded into a new article. Instead, link using a redirect to the main topic. I don't intend to be legalistic, it is cheap to ask on the talk page if this goes against the spirit of the guideline, in which case ignore all rules applies.

I may unfortunately be the wrong person to ask at this time to support; I'm perhaps more conservative in my approach. Wish you the best, I'll have more of a think. Rollinginhisgrave (talk) 01:19, 12 November 2024 (UTC)[reply]

Referenced guideline refers to inter-article section linking, not intra-article section linking. Editors of an article can't be expected to know what's going on in other articles as to section is expanded into a new article. So both the letter and the spirit fail to preclude this change, and any argument otherwise will just be a disguise for resistance to change. I've seen it a thousand times. But I'm happy with an IAR argument if it works. ―Mandruss ☎ 01:51, 12 November 2024 (UTC)[reply]

I think that's not at odds with my read: the reason why its important editors see if a new page is created is because it's worse to link to a section than to an article covering the topic. Rollinginhisgrave (talk) 02:09, 12 November 2024 (UTC)[reply]

Correct me if I'm wrong, but this is addressed in the proposal: In theory, and oftentimes in practice, there will either be a hatnote (Main article: Economic policy of the Donald Trump administration) or an immediately visible phrase in prose (Trump was inaugurated) linking the reader to a more specific article. This creates an elbow: when the reader clicks on a link in the lead, they are routed through the body. Cessaune [talk] 07:37, 15 November 2024 (UTC)[reply]

I was using the example given in the proposal. If I click "Iran nuclear deal" and it takes me to Donald Trump#Iran, I would have to figure out that the the deal is officially known as Joint Comprehensive Plan of Action before I could click.

I think the proposal could work as an elbow if links were restricted to subheadings using template:main. Rollinginhisgrave (talk) 07:54, 15 November 2024 (UTC)[reply]

Hmmm. How exactly would we go about doing that? In practice, I always imagined that that's how things would happen inevitably. Cessaune [talk] 07:57, 15 November 2024 (UTC)[reply]

In that case, using Iran nuclear deal was probably not the best example :). I am not a technical editor so I couldn't comment on how it could be done. Rollinginhisgrave (talk) 08:31, 15 November 2024 (UTC)[reply]