Wikipedia talk:Talk page guidelines

The policy on user talk pages (length, organization)

This regards WP:TALKCOND and WP:OWNTALK, respectively.

Sigh. It has recently come to my attention that the edits made to implement this: Wikipedia talk:Talk page guidelines/Archive 13#Summary so far might not work as intended. The intention was to escape the nebulous and ill-defined "rule of thumb" language and to make it clear user talk pages are not bound by restrictions such as 75K length or stale discussion incidence. In other words, that it's nobody's business if I let my own user talk page grow to contain, say, hundreds of entries in the TOC, or if I create, say, a byzantine indexing system that pushes that TOC well below the first screenful, or if I, say, end up having a user talk page several magnitudes bigger than a mere 75K.

Disclaimer: Let it be clear that my own personal opinion is for our policies and guidelines to be clear and concise. If they say talk pages should not contain many stale discussions, or if they say talk pages should not (much) exceed 75K, then it should be possible to gently nudge or template transgressors, and eventually start automatic archiving for them. However, consensus was strongly against this. So I'm settling for the next best thing, to have the guideline at least be clear and upfront about this. There should be no doubt in the readers mind our guideline lets you have any kind of user talk page you wish (as regards length and organization only, of course). Language like "rule of thumb" can be interpreted differently by different editors (is it something to act upon, or is it merely air talk there to impress visitors with no actual significance?) and is therefore something to avoid.

I believed I achieved this with the latest bouts of edits, but maybe I did not? Before making any changes I'd like to first determine if that is so, or if the current guideline accurately reflects this intent. CapnZapp (talk) 09:33, 19 July 2020 (UTC)

I have re-read Wikipedia talk:Talk page guidelines/Archive 13#Summary so far. My understanding of the consensus is that (1) the archiving section of this page applies to all talk namespaces, including user talk, (2) the decision to implement archiving on a user talk page is in the sole discretion of the user to whom it belongs, (3) these two prior statements are wholly consistent with each other, and (4) it is appropriate to use {{uw-archive}} to inform users who may be unaware of the talk page size guideline. If this is not consensus, and consensus is that the archiving section does not apply to user talk pages, then an explicit statement of such a rule needs to be added to the archiving section. But to get there, I'd think someone would have to explain why the basis for the archiving guidelines (the technical limitations of many users to display long wiki pages properly) somehow do not apply to the user talk namespace. --Bsherr (talk) 02:02, 20 July 2020 (UTC)

Thank you. What I'm getting at is that the guideline used to be incredibly unclear as to the question: can and should we make our fellow users prune their talk pages? Specifically, that "rule of thumb" can evidently be read to support your own position, even if that is directly opposed to someone else's interpretation. Saying the rule of thumb applies is incredibly unclear - the guideline ought to clearly state the consensus. As I understand the consensus, this: with the sole exception of gently informing new users of our archival tools, we should not even ask other users to have fast-loading accessible user talk pages. 75K has nothing to do with it, I can have a 854K user talk page and it's nobody's business. The number of stale or resolved discussions is also irrelevant, I can have a TOC with 428 entries in it if I like! I can even fill the top of my talk page with home-made categorizations such that the TOC isn't even visible without scrolling down! Moreover, that if you were to use our templates to tell me to fix my user talk page - even the trivial work needed to set up automatic archiving - you should expect editors and administrators to swoop in like you were a vandal, warning you "not to template the regulars" or face getting banned. In other words, striving to clean up the garbage fires of others is treated as a fairly serious offense. By reading the guideline you would completely miss how your user talk page is your sacred space not to be violated, regardless of length or accessibility. Thus; clarity is needed. Specifically, since my user talk page can be an experience entirely divorced from the guideline under the Archiving section claiming it applies is only misleading and not clear at all! I tried to achieve this using hat notes; now I have been made aware of WP:LEGITHAT (thanks, Bsherr!) which I somehow had missed. Therefore I agree "an explicit statement" is needed, though possibly for a different reason. CapnZapp (talk) 11:24, 20 July 2020 (UTC)

This is still not getting anywhere. Either the guidelines on talk page length (75K, stale discussions) apply to user talk pages, or they don't. Consensus clearly is against enforcing the guidelines, so the logical conclusion is to rewrite the guideline making it clear they don't apply to user talk pages. We don't have guidelines just for them to be ignored. We especially don't have them to give new users something to believe in, while secretly they don't apply to users experienced enough to know which guidelines that can safely be ignored. The guideline needs to clearly explain to the reader what applies equally to everybody. This can be anything, as long as its written down, so it can be read and known and challenged. If the consensus is somehow for the limitations (75K, stale discussions) to apply, while also letting users off the hook, fine. But then this needs to be addressed openly and explicitly. In other words, there needs to be an explanation for how both of these things get to be true at the same time, and that this explanation is easy to understand. That is, "consensus" needs to assume responsibility for the incongruity here. "Consensus" needs to be held accountable, which it just isn't unless words are put to the page.

If you think you can get a straight answer out of these guys Bsherr, feel free to try, because I couldn't. All I can do is update the actual guideline, since that is so far the only thing that gets them out of the woodwork so they can be forced to engage in consensus-building discussion. That is, I'm not trying to oppose you with my change, I'm merely attempting to jump-start a discussion that "consensus" clearly does not wish to have. If you want to fight for your interpretation, that's fine, but then you also need to explain to me how exactly your statement (3) works, so we can explain it to the reader, and not just assume it is understood. CapnZapp (talk) 09:51, 5 August 2020 (UTC)

Contributions at Wikipedia should aim to improve the encyclopedia or the community that develops it. That means there should be a firm guideline saying long user talk pages should be archived. However, battling editors who for whatever reason want to be different is not useful. Leave this guideline alone. Johnuniq (talk) 11:32, 5 August 2020 (UTC)

The guideline applies to user talk pages. We enforce it by giving a warning to the user whose talk page it is. (We don't implement archiving for the user, we don't block or ban the user for not following through, we don't behead the user, and just because we don't do those things doesn't mean we aren't enforcing it.) Does that make sense? --Bsherr (talk) 18:50, 5 August 2020 (UTC)

I really can't believe that this is still being pursued. This is a guideline that offers advice, not an excuse for beating up any particular editors. If anyone chooses not to follow advice then that is their prerogative. The consensus is more than clear that this is a non-issue. Phil Bridger (talk) 19:04, 5 August 2020 (UTC)

Johnuniq, at this stage I'm not trying to change the guideline. I'm merely trying to make it (much) more clear that while it is permissible to warn users (especially new users), there is no actual enforcement. CapnZapp (talk) 18:27, 8 August 2020 (UTC)

Bsherr, you're explaining to me your position. I'm not asking for a personal explanation - I want the guideline to be clearer, so no explanation is needed. How would you suggest we phrase this in the guideline so the next reader understands it? CapnZapp (talk) 18:28, 8 August 2020 (UTC)

Yeah, I think that's the sticking point we have here. You think I am advancing a position, but I think what I am explaining is the consensus reflected in the last discussion, and the guideline as currently written. So, in my view, no major changes to the guideline are needed; just to replace "article" with "subject page" in a few places. At the risk of speaking for them, it seems like Johnuniq and Phil are much on the same page as me. However: I'm not closed-minded to improving the guideline to better explain current consensus. Perhaps you have a proposal to reword the guideline to make it clearer? To do that, two things. Firstly, we'll need to move past our disagreement over whether "enforcement" can mean only giving a user warning (you seem to think it cannot, I and others here seem to think it can). Secondly, I think we should try to avoid redundancy with the Wikipedia definition of a guideline, which already states, "Guidelines are sets of best practices supported by consensus. Editors should attempt to follow guidelines...." So, that being said, how would you improve the guideline better communicate that it applies to user talk pages, that we issue warnings, but that we don't impose archiving on a user's page unilaterally? --Bsherr (talk) 19:07, 8 August 2020 (UTC)

I'll add also that I'm not taking a position on the merits of the current consensus. Perhaps that should change, or perhaps not. But Wikipedia:Templates for discussion/Log/2020 July 16#Template:Uw-archive and this discussion were initiated to address whether these pages adequately state the current consensus, so I'm limiting my scope to that question. --Bsherr (talk) 19:15, 8 August 2020 (UTC)

Phil Bridger: Did you read my post (starting with This is still not getting anywhere). Do you still believe I'm looking for excuses to "beating up any particular editors"? CapnZapp (talk) 18:32, 8 August 2020 (UTC)

I did read it and stand by the position that I stated. Just stop flogging this dead horse. As I have said previously, if you think that someone is unaware of how to reduce the size of their User Talk page then give them a friendly message about it, not some sort of warning. If they are aware but choose not to do anything then just drop it. Phil Bridger (talk) 18:43, 8 August 2020 (UTC)

It appears that everybody is fine with the guidelines clearly saying you should archive and such, while at the same time being fine with this to be trivially ignored. This reeks of hypocrisy to me, since if you don't know better you get the impression the guideline is to be followed, but if you try to act upon it you're the one getting admonished. It's a clear case of double standards. Wanting to have the cake while eating it (newbs fixing their talk pages while I don't have to). Other guidelines don't work this way. If the consensus is that it's a recommendation only that you can freely choose to not follow, that is of course fine - but it is also spelled out for everyone to realize. But not in this case. Read the guideline and walk away with a completely different impression than if you "just know".

That's it. If anyone care to bring this up to a higher level, to break this local cabal of a consensus that flies in the face of what Wikipedia usually is, feel free. CapnZapp (talk) 18:29, 27 August 2020 (UTC)

How should "latest topic" be defined for WP:BOTTOMPOST?

WP:BOTTOMPOST says: "The latest topic should be the one at the bottom of the page, then the next post will go underneath yours and so on. This makes it easy to see the chronological order of posts."

"Latest topic" isn't defined, and could be interpreted one of two ways:

1. The latest topic the last one created.
2. The latest topic is the one with the most recent activity.

I propose that we codify that both ways are acceptable.

The reason the second version is important is because sometimes a topic is still actively being discussed while several new topics below it are dead and aren't being discussed any more. In that case, the topic that's actively being discussed is stuck in the middle (or top) of the Talk page, and it isn't apparent when perusing the Talk page that there's an active discussion there. I think there's an expectation that the most recent activity will be at the bottom, or near the bottom of the Talk page, but that doesn't happen when active discussions get orphaned in the middle. As an example, see WP Talk Verifiability. The active discussion topic "RfC: Let's define self-published sources" has seven topics below it that have been dead for as long as five weeks.

I'm not suggesting to make it a requirement to move active topics to the bottom of the pages, but it should at least be acceptable. I've tried to move the active RfC to the bottom of the Talk page, which I think is in keeping with the spirit of WP:BOTTOMPOST, but another user keeps reverting me, claiming his interpretation of WP:BOTTOMPOST is the only correct one. But it's not really defined, so I'm putting the question to the community.

So, on the question of both methods being acceptable, Support, Oppose, or Tweak? -MichaelBluejay (talk) 05:01, 17 August 2020 (UTC)

I would find it disorienting for sections to move. I prefer that they stay in their original order. isaacl (talk) 05:33, 17 August 2020 (UTC)

The text should be reworded to more clearer match the practice that has been standard for a long time (probably forever). Namely, new topics are started at the bottom, otherwise topics don't move. The other possibility is just begging for editors to play games with topics moving around. Zero^talk 05:36, 17 August 2020 (UTC)

The quoted text means to add a new topic to the bottom of the page, not to move topics every time a reply is made. It's clear in context: right before the text quoted, it says (bold as in original) "Start new topics at the bottom of the page: If you put a post at the top of the page, it is confusing and can easily be overlooked. The latest topic should be the one at the bottom of the page, then the next post will go underneath yours and so on."

Doing otherwise would require editing the entire page for each new comment, and could not have been what was intended. If the guideline is being misunderstood, I wouldn't object to making it more clear, but it seems pretty clear to me. TJRC (talk) 05:43, 17 August 2020 (UTC)

Example: Talk:Ayurveda#A lead paragraph without the whitewashing I made a proposal for a lead paragraph. put it on hold until an RfC closed, then made the edit. The section was in the middle of the page, but I knew that as soon as I made the edit there would be a lot of discussion, so I moved it to the bottom. This should be a judgement call. --Guy Macon (talk) 11:27, 17 August 2020 (UTC)

Whether that example was good, I'm not sure, but I don't think this guideline should provide it as an option. IAR is there for rare exceptions. Zero^talk 12:05, 17 August 2020 (UTC)

Whilst Michaelbluejay (talk · contribs) has written As an example, see WP Talk Verifiability., they fail to disclose a number of key facts: (a) which particular edits at WT:V this relates to (for the record, they are: 1 & 2 and my revert; 3 and my revert); (b) this was discussed at my talk page a few days ago, where Izno (talk · contribs) also commented; and (c) it is related to an ongoing RfD (since closed as delete --Redrose64 🌹 (talk) 19:20, 20 August 2020 (UTC)).

My main objections to the proposed change are twofold: (i) it would require everybody who posts to a thread that was not at the bottom of the page to move it there, and that way lies chaos (consider a page having two active discussions - they would frequently need to exchange their relative positions); (ii) as may be seen from the diffs that I have posted, it makes it very difficult to see what the real change was. For instance, in this edit (that I already mentioned), how many of you spotted the alteration of "adopting Masem's definition?" to "adopting Masem's definition of SPS?" --Redrose64 🌹 (talk) 20:47, 17 August 2020 (UTC)

I could go with something like "move sections with care" and "make the section move a separate edit from any content change" Moving the currently active discussion to the bottom is a useful tool that can improve the discussion, but should be rarely used and only by experienced editors. -Guy Macon (talk) 23:52, 17 August 2020 (UTC)

I think once I moved a new post from top to bottom on a talk page. That makes sense. I never felt like moving an old discussion, but a few times started a new discussion that might have been related to an old one, and put that at the bottom. Most of the time I think that is better than moving an old discussion. Gah4 (talk) 00:52, 18 August 2020 (UTC)

I could live with that. The minor amount of extra work might be a good trade-off for people not being allowed to game the system to give their comments extra prominence. --Guy Macon (talk) 02:00, 18 August 2020 (UTC)

Thinking about it further, I think having two sections where people are discussing the same thing could be a problem. --Guy Macon (talk) 13:38, 18 August 2020 (UTC)

There shouldn't be two identical sections actively discussing the same thing. In your cited example you should have simply closed the initial section in the middle of the page and started a new section on the subject, taking the RFC outcome into account, on the bottom of the page.Tvx1 14:03, 18 August 2020 (UTC)

Even better! I propose that the policy presents as an option (not required!) closing the old discussion (but only if there have been no comments in the previous four days) and opening up a new discussion at the bottom with a link to the old discussion. That seems better than moving sections, and that's what I plan on doing from now on for those rare times when I think moving a section is appropriate. --Guy Macon (talk) 22:53, 18 August 2020 (UTC)

---

The section has been changed by recent editing. Flyer22 Frozen (talk) 04:52, 18 August 2020 (UTC)

"Doing otherwise would require editing the entire page for each new comment, and could not have been what was intended." (groan) As I stated explicitly, "I'm not suggesting to make it a requirement to move active topics to the bottom of the pages, but it should at least be acceptable." In any event, it's clear that we're not going to get anywhere close to consensus for my proposal, so I withdraw it. -MichaelBluejay (talk) 19:27, 18 August 2020 (UTC)

"My main objections to the proposed change are twofold: (i) it would require everybody who posts to a thread that was not at the bottom of the page to move it there..." NO, NO, NO, NO, NO, NO! "I'm not suggesting to make it a requirement to move active topics to the bottom of the pages, but it should at least be acceptable." -MichaelBluejay (talk) 19:31, 18 August 2020 (UTC)

Then why did you post this. followed later on by this edit? Clearly, you do want people to move threads to the bottom. --Redrose64 🌹 (talk) 22:27, 18 August 2020 (UTC)

Could you dial back the aggression, please? Telling people that you know what they want better than they themselves do seldom works out well. --Guy Macon (talk) 22:56, 18 August 2020 (UTC)

• Before violence breaks out, could someone please give an example of actual behavior, on an actual talk page, that the contemplated guideline changes are intended to forestall? (If such an example is already hidden somewhere above, my apologies.) EEng 01:12, 19 August 2020 (UTC)

  I have never seen any examples, but then again I am a newbie -- only 14 years and 51,000 edits. :) --Guy Macon (talk) 01:31, 19 August 2020 (UTC)

  Then what is everyone arguing about? EEng 02:45, 19 August 2020 (UTC)

  @EEng and Guy Macon: I gave examples in my post of 20:47, 17 August 2020 (UTC). --Redrose64 🌹 (talk) 22:29, 19 August 2020 (UTC)

  Um, your link shows someone moved a thread, and I gather from something elsewhere that maybe the move changed something. That's bad. So does the guideline tell people to move threads, and the proposal is to change it to not say that? Or is it something else? EEng 00:06, 20 August 2020 (UTC)

  @EEng: Mathglot put the anchor in the middle of my post, not at the start, so you missed all of the examples that I intended you to read. Anyway, the "someone" who moved a thread is Michaelbluejay (talk · contribs) (yes, the same person who originated this discussion), and as I pointed out elsewhere, it is Michaelbluejay's earlier misinterpretation of the existing guideline that triggered this all off. --Redrose64 🌹 (talk) 19:18, 20 August 2020 (UTC)

  Oh, no; mea culpa! Sorry, Redrose64, that was inadvertent; and sorry everybody for some unintentional misdirection. My intent was transparency, it's unfortunate that it just made a muddle of things. (Walks away sheepishly...) Mathglot (talk) 19:41, 20 August 2020 (UTC)

• Per Zero0000: new topics are started at the bottom, otherwise topics don't move.

I get Guy Macon's dormant section update concern, although I haven't felt that to be a concern myself; but let me see if I can address it.

As I understand it, Guy, the use case is roughly this: section "Middle" goes quiescent for a while; meanwhile, five or ten new discussions get added to the bottom of the page. Now, you want to reply to "Middle", and not have to worry that it won't be noticed halfway up the page. If I misunderstood, please correct me.

My proposed solution: start section "Middle prime" at bottom of page, and start it with some boilerplate, e.g. "Following up on the issue in section [[#Middle]] above, I wanted to propose that...". (You could instead use {{Discussion moved from}}.) Concomitantly, up in section "Middle", you append a {{Discussion moved to}} template, or just a link to "[[#Middle prime]]". (To underscore that the follow-up is in the bottom section exclusively, you could Close the discussion.)

Does that work for you, Guy? WP:REFACTOR discusses the possibility of moving discussions around, as you mentioned, but it's a bit of a bold move, as REFACTOR implicitly acknowledges, saying: "If another editor objects to refactoring then the changes should be reverted." But continuing a dormant section in a new, bottom-of-page section would never be off-limits, afaict; and I believe it handles your "dormant-section-update-findability" concern, as well as my "dont-move-the-furniture-around-while-I'm-asleep" concern. What do you think?

Oh, and as to the original point of adding a definition: I don't think it's necessary here. It's hard to imagine anyone interpreting "Last" other than bottom of page. If there's a counterexample demonstrating confusion on someone's part, please add a diff link. We shouldn't be fixing problems that don't exist. Mathglot (talk) 06:25, 19 August 2020 (UTC)

Repairing faulty ping: User:Zero0000. Mathglot (talk) 06:27, 19 August 2020 (UTC)

@Mathglot: In my post of 20:47, 17 August 2020 (UTC), I linked to User talk:Redrose64#Misapplying WP:BOTTOMUP, where confusion is shown over what constitutes the "latest topic". --Redrose64 🌹 (talk) 22:29, 19 August 2020 (UTC)

Lol, well that's a technical win for you, Redrose64. But you have to know the backstory, which is that MichaelBlueJay is half Vulcan, and his logic sometimes leads him to a conclusion that mere humans like us would never imagine. So let me amend my, "hard to imagine anyone" to "hard to imagine anythree interpreting...". (Michael, no canvassing T'Pol, pls.) Thanks, Mathglot (talk) 23:22, 19 August 2020 (UTC)

• Thanks for the suggestion but this guideline will never say anything that might encourage people to move sections on a talk page. That way lies madness with hard-to-check diffs (did the new user correctly move sections, or was something lost?) and pointless turmoil (I think a certain section should be at the bottom but you think it shouldn't) and edit conflicts. If there is a dormant section below something important, either archive the dormant section or, as Mathglot says, start a new section that references the hard-to-see earlier discussion. All of that is for experienced editors only—by definition, if you don't know how to do it or whether it would be desirable, you should leave it alone. If really essential (it isn't), ask at WP:Teahouse for assistance. By the way, WP:BOTTOMPOST has been fixed (thanks EEng). Johnuniq (talk) 07:21, 19 August 2020 (UTC)
• I count 114 section headings or dot points in this article, which must be a record for all guideline pages. I think that is way over the top and hardly any editors will take the time to study it. Zero^talk 07:47, 19 August 2020 (UTC)
• To expand on my previous comment: for most pages I use my watchlist to be notified of changes and then look at the history diff between the last edit I read and the current edit. For this, sections moving around introduces a lot of noise and makes it harder for me to find the latest changes. For high traffic pages such as the incidents noticeboard (ANI), I skim through the sections I'm interested in looking for changes, generally based on timestamp. Moving sections around would make it harder for me to find sections of interest. In addition, the hard changes to find are those in the middle of a section, thanks to the multi-branching discussion convention used on Wikipedia, and moving sections to the bottom doesn't help me at all in this case. So for me personally, moving a section to the end would not help me find changes in it. Does someone have another mode of operation that would be aided by this proposal? isaacl (talk) 10:20, 19 August 2020 (UTC)

Re: "Does that work for you, Guy?", not only does Mathglot's "My proposed solution" above work for me, but it is clearly superior to what I have done in the past. I have already decided to do the above in those rare cases where I previously would have (carefully) moved a section. I would like to see an explanatory supplement on this so other editors can benefit by knowing the preferred method. --Guy Macon (talk) 15:00, 19 August 2020 (UTC)

I'm sorry, but if we're talking about continuing a discussion (that's in an old section in the middle of the page) by creating a new "Middle prime" section at bottom of the page, with these pointers back and forth, that's a terrible idea that fragments the discussion. Just continue the thread in the section it's in, wherever that is (maybe starting a ===-level subsection to highlight that discussion's being resumed). If there are stale threads below the thread you're resuming, then archive them; if all the threads below are still active, well then your thread is just one more of the active threads at the end of the page. And people who really care should be clicking on X changes since my last visit in their watchlists.

I'd be against anything recommending this create-a-new-section-with-pointers idea. EEng 15:23, 19 August 2020 (UTC)

@EEng:, That's a disingenuous misdirection, I'm afraid. You're way too smart a guy not to know what fragmenting a discussion is about. Actual fragmentation by cross-posting the same discussion on two pages lessens coherence and ability to follow a discussion, because it goes on separately in two places. This proposal improves coherence and transparency, by preventing fragmentation. Whatever its other disadvantages may be, fragmentation is certainly not one of them. Mathglot (talk) 19:58, 20 August 2020 (UTC)

I appreciate the compliment, but I'm afraid I'm not smart enough to understand what you're saying. The proposal seems to be that a thread in the middle of a page might be continued by opening a new section at the bottom of the page, with some kind of pointer saying "OK, here we'll continue the earlier thread named Foo above on this page". (If I misunderstand, my apologies.) That breaks the discussion into two parts for no reason I can see, and that's fragmenting it (even if not in the WP:CROSS-POST sense.) EEng 20:37, 20 August 2020 (UTC)

Telling newbies to post at the bottom without telling them why

EEng, regarding this? Why tell editors to post at the bottom and not state why we are telling them that? Do you really believe that is the best route, or are you reverting just to revert me? You really think I need to provide diffs of newbies wondering why they are being told to post to the bottom or repeatedly posting to the top after being told to post to the bottom...when many of us have experienced this?

And, Francis Schonken, regarding this? My reason was not "Oh, I talked it over with Johnuniq, and our opinions trump others' opinions." It's that Johnuniq is the one who initially removed that part, I asked him about it via email, and he was clear that he went overboard on that matter and he agrees that this piece should stay. Flyer22 Frozen (talk) 03:50, 20 August 2020 (UTC)

What valid reason is there to remove this long-standing short piece from the guideline? Flyer22 Frozen (talk) 03:52, 20 August 2020 (UTC)

• without telling them why – What are you talking about? My text [1] was

  Start new topics at the bottom of the page, where they are most visible. Use the "New section" button (at the top of the page) to do this.

So it does tell them why: it says that the bottom is the most visible location – surely a strong motivation. And yes, you claimed [2] that "countless times" newbies have explicitly asked "Why do I have to post at the bottom of the page?", which I think is hyperbole at best. But if it's really true then you should have no trouble providing a few diffs; asking for evidence that there's a problem to be solved is standard when discussing guidelines, to counter bloat.

And I'd further request that you produce such diffs from newbies who went on to be productive editors. Let's say the professor says, "Time's up. Place your bluebooks in this box." One kid says, "Why should I?" How long do you think he'll remain in good standing? Same principle applies here.

What valid reason is there to remove this long-standing short piece – The reason is that this page is 30% longer than it needs to be because of all the soporific blah blah blah blah blah blah blah, so nobody reads it and the rules we're trying to inculcate slumber here unspied upon, magnificently impotent. That's why. EEng 04:14, 20 August 2020 (UTC)

What am I talking about? The following is telling them why: "If you put a post at the top of the page, it is confusing and can also get easily overlooked." Your "where they are most visible" wording is an unnecessary, watered down cut. No, I'm not going to provide you diffs on a common occurrence that many long-time Wikipedia editors, including myself, have experienced, as I though I am lying. You stated, "[I]f it's really true then [I] should have no trouble providing a few diffs." Oh, yes, because I should think back on the encounters I've had, trying to remember which newbies I interacted with or saw do this when interacting with others. I should do this...when I and others have dealt with so many newbies. Nope. I wasn't inclined to provide examples of editors breaking up others' posts when I made my argument on that matter either. Others provided examples. This "telling newbies why they should post to the bottom" thing? I'm not going to debate something that should obviously be stated in the guideline. Like I've done for years, I'd rather just point to that section instead of tell a newbie, "The reason you should post at the bottom is because it's standard to post that way, and so posts at the top will be considered old and may therefore be overlooked." When I made this clarifying edit summary, I figured that, just to aggravate, you would predictably remove the piece I restored. Or do you just think that only you get a say in how these guidelines are worded? Considering this wholesale revert, it seems to me that you should have left it alone. Flyer22 Frozen (talk) 04:48, 20 August 2020 (UTC) Tweaked post. Flyer22 Frozen (talk) 04:57, 20 August 2020 (UTC)

Please AGF. I didn't intentionally remove your text. It turns out that there are two bullet points headed Start new topics at the bottom of the page – one in the Layout section and one in the New topics and headings on talk pages section. As a result it seems we've been talking at cross purposes, and in the change you complain about [3] I was adding the "where it's most visible" text, to address the concern expressed in your edit summary. Only if you scroll down in the diff does it become apparent that, through some accident of which revision I grabbed, I was (unintentionally) also removing something in the second Start new topics bullet later on the page. OK? EEng 12:49, 20 August 2020 (UTC)

Okay, EEng. We do have a somewhat tempestuous history; you know that. So it's not unreasonable for me to consider that you were just poking me. But, as seen above, I also wondered why you didn't want us telling newbies why we want them posting at the bottom. Flyer22 Frozen (talk) 02:29, 21 August 2020 (UTC)

Without my glasses I though for a moment it said temptress history. EEng 19:40, 28 August 2020 (UTC)

Whether newly created discussion sections are put at the bottom or the top is a matter of convention within a given community. Those that are used to having them on top would find that to be most visible or most likely to not be overlooked, and vice versa for those used to having them on the bottom. We only need to say adding new threads at the bottom is the convention in English Wikipedia. isaacl (talk) 06:12, 20 August 2020 (UTC)

Exactly. EEng 12:49, 20 August 2020 (UTC)

• The most visible part of a page is the top, not the bottom. That's why, for example, we put the title and lead of an article at the top, rather than the bottom. So, we should not use visibility as an explanation because it would be wrong. Bottom posting is just a convention and it is not always followed on Wikipedia. Places where new entries are top-posted include WP:AFD, WP:ITN and WP:RFA. Presumably, that's because in those areas, they want the newest item in the most visible position – the top. Andrew🐉(talk) 06:56, 20 August 2020 (UTC)

  Most people coming to an article haven't seen it before, and start reading at the beginning; that's why the lead is at the top. Most (or many, anyway) people coming to a talk page want to know what's new, and start at the bottom; thus the bottom is arguably the most visible location (and of course it's well known that talk page headers, FAQs, and so on are overlooked in the main). The old text was

  Start new topics at the bottom of the page: If you put a post at the top of the page, it is confusing and can easily be overlooked

  so if you like we can avoid the "most visible" issue by simply saying

  Start new topics at the bottom of the page where they won't be be overlooked

  Though as I mention somewhere above, it turns out there are two Start new topics at the bottom of the page bullets on the page, and we've all been trapped in a comedy of errors. EEng 12:49, 20 August 2020 (UTC)

• In both mentions of posting at the bottom it should say why. That's not "bloat", it's brief and motivates good behavior. And posting at the bottom makes it more visible to Wikipedians, so we should say it makes it more visible. Most of us scroll past the old junk at the top of many talk pages, or have seen newbies post there. Crossroads ^-talk- 17:59, 20 August 2020 (UTC)
  • I would have agreed with that but Andrew's "most visible part of a page is the top" post above pretty well demolishes the argument. If a reason is to be given, it should be that other editors expect new topics to be at the bottom without entering the rabbit hole of whether bottom-posting is more visible than top-posting. All I want is for the guideline to start by telling people what to do (click the "new section" button) and use minimum waffle so that message is not diluted. Johnuniq (talk) 02:13, 21 August 2020 (UTC)

    That very last thing you said is the golden principle of which this page so very much needs generous and frequent application. EEng 02:47, 21 August 2020 (UTC)

• As several others have pointed out, it is not true that the bottom of the page is the most visible. The most visible place is where your browser first takes you when you visit. For all of my browsers, that is the top. It is true that posting at the top will confuse other people, but that's only because other people know the convention is to post at the bottom. In summary, the only real reason that new topics should go at that bottom is that that is what we do. Editors are not too stupid to understand that, and they don't need more reasons than that, especially not false reasons. Zero^talk 12:18, 21 August 2020 (UTC)
  • Everyone agrees that posting a new comment in a thread that is in the middle of a large number of threads and which hasn't had a new comment in months is not particularly visible. The question is waht, if anything, to do about it. --Guy Macon (talk) 12:43, 27 August 2020 (UTC)
    • As I said before close the discussion in the middle of the page and start a fresh one at the bottom.Tvx1 12:58, 27 August 2020 (UTC)

Here are the options

Consider the situation where there is an active talk page with many sections. You see a thread in the middle of the page that has not had a new comment in weeks. You wish to add a comment. Your options are:

1. Just add your comment, sit back, and watch as nobody notices it. Not popular among those who leave comments in the middle of the page and get zero responses.
2. Start a new section at the bottom, possibly with two-way links between the old and new sections. Not popular among those who don't want to see the same thing discussed in several sections.
3. Move the existing section to the bottom in one edit, them add your comment in a separate edit. Not popular among those who dislike the idea of allowing editors to move sections.

Did I miss any?

So, which unpopular option do you want to pick? --Guy Macon (talk) 14:43, 27 August 2020 (UTC)

I dispute the premise "watch as nobody notices it." Most editors see comments not because they're made at the top or or at the bottom of the page, but because the article is on their watchlist and they are presented with a diff. I don't believe that the position of the new comment on the page has any significant bearing on whether it is noticed.

It is true that many new comments added to very old threads are ignored, but that's not because the comment was made in an area that is not noticed, but because the thread is long-dead and no one has any interest is starting to flog a long-dead horse. TJRC (talk) 18:39, 27 August 2020 (UTC)

Exactly. If an editor comes to the page via their watchlist, the diff tells them where to look. If an editor just wanders onto the page fresh, never having seen it before, then the whole page and its constituent sections are new to them explore as they will, and there's no way to overlook the new post because it's because it's all new. EEng 21:06, 27 August 2020 (UTC)

And yet, if the exact same "very old thread" is moved to the bottom with a new comment, for some inexplicable reason other editors do tend to "have an interest in starting to flog a long-dead horse". And on those pages where the long-dead threads don't have a bunch of other threads below them, once again the other editors tend to reply to the new comment. It's almost as if your theory about why comments made in the middle of a page with many threads don't get many responses isn't the correct one. --Guy Macon (talk) 19:59, 27 August 2020 (UTC)

I'd like to see evidence of this claim. EEng 21:06, 27 August 2020 (UTC)

It's hard to see how your theory or my theory could be confirmed. I can easily find examples of comments in the middle being ignored and comments with moved sections getting replies, but I can also easily find examples showing just the opposite. Clearly anecdotes are not good enough. One would have to do some statistical number crunching, and even then the numbers would be suspect -- there may be something different about users who comment on old threads in the middle of long talk pages. --Guy Macon (talk) 02:55, 28 August 2020 (UTC)

hard to see how your theory or my theory could be confirmed – Right, but I'm not proposing an elaboration of the already-complex guidelines in order to solve a problem merely conjectured. EEng 03:22, 28 August 2020 (UTC)

I made no proposal. I commented on a few proposals by others but the closest thing to a proposal that I wrote was "This should be a judgement call" at 11:27, 17 August 2020 (UTC). --Guy Macon (talk) 09:34, 28 August 2020 (UTC)

My apologies. You will understand how easily one becomes confused in the thicket that is this discussion. EEng 12:52, 30 August 2020 (UTC)

1 or 2, but not 3. Simple. CapnZapp (talk) 18:12, 28 August 2020 (UTC)

To add to that;

1. You can {{ping}} relevant users to ensure they don't miss a comment made way up in the middle of a large talk page.
2. if you have a good reason (meaning you're not just beating a dead horse) to "restart" a slumbering discussion that's "lost" "up page" (maybe you're late to the party), you can create a new talk section (and link to it at the end of the previous one) for maximum visibility. By adding that link (for instance "Discussion continued below") you have effectively staved off the "same thing discussed in several sections" concern, since your link effectively inactivates the old section - any editor that find the old section can (and probably should) follow your link before placing his or her comment. Note: this is exactly the same as when you find an archived discussion you want to revisit - you're instructed to start a new section on the active talk page. If you want/need to repost parts of the previous discussion, you copy (or link to) it, you never move it.

Cheers CapnZapp (talk) 08:37, 30 August 2020 (UTC)

Moving parts of a section is already permitted

"Sectioning: If a thread has developed new subjects, it may be desirable to split it into separate discussions with their own headings or subheadings. When a topic is split into two topics, rather than sub-sectioned, it is often useful for there to be a link from the new topic to the original and vice versa. A common way of doing this is noting the change at the [then-]end of the original thread, and adding an unobtrusive note under the new heading, e.g., :This topic was split off from #FOOBAR, above.. Some reformatting may be necessary to maintain the sense of the discussion to date and to preserve attribution. It is essential that splitting does not inadvertently alter the meaning of any comments. Very long discussions may also be divided into sub-sections."

While the guideline says "The latest topic should be the one at the bottom of the page, then the next post will go underneath yours and so on. This makes it easy to see the chronological order of posts", I seriously doubt that anyone would report a violation if the person doing the splitting kept the two sections together where the one section used to be -- especially if it was split into a subsection.

Despite the above wording allowing the moving of part of a section, and contrary to what some editors on this page have implied, I could find no language that specifically allows or disallows moving an entire section section. (Feel free to correct me if I missed it.) I am unconvinced that we need to modify the existing guidelines to address this. We already have ways of dealing with disruptive edits without trying to specify all the ways someone can be disruptive. --Guy Macon (talk) 09:52, 28 August 2020 (UTC)

Remember, this talk section is about MichaelBluejay's proposal to allow moving talk sections (hopefully opposed by the consensus by now). What you're quoting isn't permitting moving sections: it means to create a new fork, not moving the existing ones. Essentially, you're answering your own question: if you "break the rules" and move a section for a really good reason, noone will likely intervene. Worst thing is someone will revert you, and in that case, you'll have to settle for interlinking. And that's it. Explicitly allowing section moves (even only in certain cases) will do much more bad than good, imo. CapnZapp (talk) 18:23, 28 August 2020 (UTC)

I see no reason why this subthread, which contains no material about MichaelBluejay's proposal, has to be limited to being "about MichaelBluejay's proposal".

That's a rather interesting interpretation of "If a thread has developed new subjects, it may be desirable to split it into separate discussions with their own headings or subheadings". That sure sounds like starting with an existing section and splitting it into two sections, and not at all like "creating a new fork, not moving the existing ones". --Guy Macon (talk) 18:39, 28 August 2020 (UTC)

I've lost track of this discussion but would suggest that moving stuff around should be done rarely and only by experienced editors. Further, there should not be edit wars if another experienced editor reverts. That can't be codified in a guideline and should be (rarely) discussed on a case-by-case basis. That is, I think I'm agreeing with Guy Macon that the guideline should not attempt to tell people moving stuff is ok because it's too complex to explain when it would be reasonable and how to respond if the move is challenged. If it's in the guideline, misguided people can keep their dead-horse topic alive indefinitely by moving it to the bottom with a repetition of their against-consensus opinion. Johnuniq (talk) 00:05, 29 August 2020 (UTC)

I meant no offense, User:Guy Macon. But the section is rather confused and sprawling as is it, so maybe you'd be better off starting a brand new section where you clearly introduce us to your problem (no previous reading assumed). Cheers CapnZapp (talk) 08:09, 30 August 2020 (UTC)

I see what you did there. (smile) Suggesting starting a new section on the same subject during a discussion where some are in favor of starting a new section on the same subject and some are opposed to starting a new section on the same subject -- brilliant! --Guy Macon (talk) 12:33, 30 August 2020 (UTC)

I completely agree with Johnuniq, Guy: moving stuff around should be done rarely and only by experienced editors and That can't be codified in a guideline. Specifically that it can't and is not codified in our current guidelines. This should not change. Yes, exceptions occur. No, that still doesn't mean "Moving parts of a section is already permitted" - I've explained how a split/fork does not suggest/encourage movement of existing content. Thanks, CapnZapp (talk) 09:39, 31 August 2020 (UTC)

On topic, I'm not sure what to say. Whatever you may think, no guideline ever on Wikipedia encourages editors to move around/rearrange/reorder the talk comments of others (with few exceptions*). If (and only if**) what you're trying to get to is that it is possible to interpret this language as such, then you're welcome to suggest a reworded phrasing that makes it more clear that, no, just because a discussion develops a new subject does not mean it's cool to touch what's already been said. What this guideline means is that it's perfectly okay to split that out into its own (sub) header with or without links back and forth, still not moving any existing content. CapnZapp (talk) 08:21, 30 August 2020 (UTC)

I never implied that a guideline Wikipedia encourages editors to move around/rearrange/reorder the talk comments of others. I did point out (with a direct quote to the guideline) that one kind of "moving around/rearranging/reordering" (splitting) is explicitly allowed. --Guy Macon (talk) 12:33, 30 August 2020 (UTC)

You interpret it as "explicitly allowing" movement. I hope to have shown you how you can (and should) interpret it as not suggesting any movement of existing content. CapnZapp (talk) 09:44, 31 August 2020 (UTC)

*) even then, you only remove, archive or hide (e.g. {{cot}}/{{cob}}) stuff. Actually moving (rearranging) talk comments is almost never needed or accepted.

**) I have a feeling you want editors to be able to move stuff around, and if so, please disregard. Cheers CapnZapp (talk) 08:21, 30 August 2020 (UTC)

That's an interesting opinion, but it conflicts with the established fact (which I pointed out with a direct quote to the guideline) that one kind of moving (splitting) is explicitly allowed. --Guy Macon (talk) 12:33, 30 August 2020 (UTC)

Such feelings are in direct conflict with my clearly worded statements that I do not think that there needs to be any change in any of our current policies or guidelines on this. Let me be even blunter: Our existing policies and guidelines are not broken and do not need fixing. Avoiding WP:CREEP in this area has served us well. --Guy Macon (talk) 12:33, 30 August 2020 (UTC)

I agree about there not being a need for change, but that is because your view, that "Moving parts of a section is already permitted", is false. Not sure what else I can say, Guy. CapnZapp (talk) 09:29, 31 August 2020 (UTC) Well, that we should not confuse exceptions for rules - I have moved the comments of other editors myself, but not because specific policy allows it. Instead because we do not encode every exception, and when something is clearly misplaced, nobody will object if you make a neutral clean-up edit. This does not mean our policies actively encourage editors to move stuff around for any purpose. This leaves you with two choices:

1) silently disagree. When you act upon your reading of the guidelines, be prepared to yield if and when opposed, since the other editor will likely not share your interpretation. If you "get away" with it, congratulations - you have graduated into being a pro editor who knows when rules must be broken :)

2) actively propose a change to make it clear (or, I guess, more clear, anyway) movement is allowed (per some conditions, I'm sure).

Regards, CapnZapp (talk) 09:29, 31 August 2020 (UTC)

Redirects & talk page archives

Some talk page archives (which are subpages, i.e. "Talk:Foo/Archive 1") have their subject pages (i.e. "Foo/Archive 1") as redirects to their base pages (i.e. "Foo"). The argument in favor is that this provides a convenient manner of linking to the base page rather than a red link. The arguments against are that this pollutes search results, that the most relevant breadcrumb trail back to the base talk page is already part of the interface, and that this creates a lot of redirects with little utility. For occurrences in the article namespace, there has been a consensus at WP:RFD to delete. (See Wikipedia:Redirects for discussion/Log/2019 November 2#Neutral country/Archive 01, Wikipedia:Redirects for discussion/Log/2019 November 10#Mainspace archive subpages, Wikipedia:Redirects for discussion/Log/2020 January 27#Mainspace archive subpages, etc.) There is also an RfD concluding in deletion for a similar redirect in the Help namespace (with an ironic topic) (Wikipedia:Redirects for discussion/Log/2016 July 3#Help:Archiving a talk page/Archive 1). I recently commenced an RfD for one in the Wikipedia namespace (Wikipedia:Redirects for discussion/Log/2020 September 23#Wikipedia:Template index/Cleanup/Archive 1), but it occurs to me that it would be useful to form a consensus about a uniform practice, and then to document it in an appropriate guideline. So, should these redirects be created or not (and should this vary by namespace)? --Bsherr (talk) 15:21, 23 September 2020 (UTC)

archive editing and reflist-talk

I know in general we aren't supposed to edit archive pages. There is one that should have a {{reflist-talk}} that doesn't have one, so the references come out in the wrong place. Is it too strange to edit, so they come out right? Gah4 (talk) 00:34, 6 November 2020 (UTC)

There are hundreds (if not thousands) of such archive pages. Which one are you thinking of? --Redrose64 🌹 (talk) 21:15, 6 November 2020 (UTC)

Specifically Talk:Proton/Archive_1, but mostly making sure that there isn't a rule against it. Gah4 (talk) 21:41, 6 November 2020 (UTC)

As Redrose64 is implying, there is not much point in fixing a minor issue in one archive because there are many more. In the absence of a really compelling reason, please do not edit archives. Doing so complicates future attempts to investigate past discussions because an edit history casts doubt on what the archive was like at the time it was archived. That can be worked out, but it's simpler to leave it alone. Johnuniq (talk) 23:31, 6 November 2020 (UTC)

OK, that is what I suspected. Yes it probably doesn't matter much, but it does complicate reading them when the references come out in the wrong place. Thanks. Gah4 (talk) 00:18, 7 November 2020 (UTC)

Order of replies

Flyer22 Frozen and I are having a disagreement at Talk:Transsexual about the normal order of replies to a given talk page comment. My belief, based on 6 years of participating in discussions on Wikipedia, is that replies to the same comment, at the same level of indentation, are generally added one beneath the other chronologically, to preserve the natural flow of discussion. Flyer22 Frozen disagrees. Who's right? —Sangdeboeuf (talk) 06:27, 2 December 2020 (UTC)

As usual, the above editor is misrepresenting what I've argued. SMcCandlish, can we finally get your commentary on this non-issue? Flyer22 Frozen (talk) 06:30, 2 December 2020 (UTC)

At 05:46 I stated, "replies to a given comment (as indicated by indentation level) should be ordered chronologically". You replied at 06:04 stating, "I disagree with your analysis ... Whether or not it's clearer in chronological order is obviously subject to disagreement'. What am I misrepresenting exactly? —Sangdeboeuf (talk) 06:41, 2 December 2020 (UTC)

Anyone who takes the time to read that entire off-topic thread will be able to see what you are misrepresenting, and that includes taking my "Whether or not it's clearer in chronological order is obviously subject to disagreement." sentence out of context. You are done wasting my time on this. Do propose something about chronological order with regard to what you've argued, and see where that goes. It will be just like I stated. Flyer22 Frozen (talk) 06:43, 2 December 2020 (UTC)

The misrepresentation appears to be that she does not disagree that such is the "normal" or "general" practice as you said in your opening post here, but does disagree with you regarding exceptions, saying they can be made. And they can. Crossroads ^-talk- 06:51, 2 December 2020 (UTC)

For context, a user replied to a comment of mine here, after which Flyer22 Frozen replied to the same comment, but above the first user's comment, here. The first user then moved their comment back to where it was originally. Both were replying to the same comment of mine, at the same indentation level. My question is, was Flyer22 Frozen "correct" to put their comment above the earlier comment, as argued here? —Sangdeboeuf (talk) 07:20, 2 December 2020 (UTC)

It's courteous to place your reply below any preceding replies. Earlier replies, being closer to the original comment, are generally easier to understand in context than later replies. It's uncollaborative for editors commenting afterwards to assume that their comments should be placed in a more advantageous position. isaacl (talk) 07:26, 2 December 2020 (UTC)

Seems a bit unconventional, but it didn't really break anything. When "injecting" like that, it seems to me to be customary to indent one level extra, to make it very clear that it's a later injection. This is how I do it, in the rare cases I think it's necessary, and no one's groused at me about it for a long time. I tend to only do this when I think it's important (e.g. to correct a false statement, with a diff proving it); when it's seriously clarifying for all (e.g. to provide a link to a more pertinent policypage than someone cited: "I think you meant MOS:THECAPS; WP:THE is about titles, not body copy."); or when it's utterly trivial, like a one-liner joke (which I usually do in {{small}}). If it's a regular post that is fairly likely to generate a reply to it, I'll almost always bottom-post it (quoting with {{tq}} as seems needed, if the thread is long and my reply is a screenful or more away from what I'm replying to). In this specific instance, it looks like Flyer was making a substantive point, while the other was just providing some trivial one-liner update, so the injection might subjectively have been considered reasonable. I would not have done it at the same indent level, but one deeper:

    OP says yadda yadda. -PrimoPoster
     ::I interject! -InternetLoper
    :The original first reply. -QuickOnTheBall
 — SMcCandlish ☏ ¢ 😼  07:41, 2 December 2020 (UTC)

OK then, what happens when the original reply and the "injection" produce further replies, generating entire new sub-threads, such as:

    OP says yadda yadda. -PrimoPoster
     ::I interject! -InternetLoper
    :::Reply to Loper. -PrimoPoster
     ::::Reply to Primo. -InternetLoper
    :::::Reply to Loper. -PrimoPoster
     ::::::Reply to Primo. -InternetLoper
    :The original first reply. -QuickOnTheBall
     ::Some stuff. -PrimoPoster
    :::Some more stuff. -QuickOnTheBall
     ::::Even more stuff. -PrimoPoster
This seems like it could seriously break the flow of discussion, especially if the lower sub-thread actually came first. —Sangdeboeuf (talk) 08:30, 2 December 2020 (UTC)

Introducing an extra indent level causes extra announcements by screen readers, so is generally not desirable. isaacl (talk) 09:04, 2 December 2020 (UTC)

If it were done frequently, I would agree (to the extent our talk pages can be sensibly parsed by screen readers at all). At this point, I'm pretty sure screen-reader users know what is going on in an occasional case of "injecting", just as they know what's going on when they hit a list gap or hit a bunch of lines with :* markup that should have been **, and so on; they can tell by overall familiarity and previous problem-solving with it, like people who live in San Francisco can tell where they are even when the fog is thick. FWIW, I have repeatedly pestered the MW devs, at every version of their bug-tracking system over the years, to reparse talk pages to stop treating : and * as list markup and do something else with them. There are a lot of potential solutions, but the devs just DGaF about this.

Geekery: One approach would be using <article>, or trusty ol' <div> with some classes. For intentional "lists proper", like examples of article text one is working on with a list in it, a <notalk> wrapper could make that material be parsed the same way it is in main or project space. And there could be an __ISTALK__ magic word that forced the parser to treat a non-talk page, like ANI or VPPOL, as a talk page for parsing purposes. (There's already some more internal way of doing this, and that's why ANI has a "New section" button in addition to "Edit"; but there's no reason it shouldn't be doable just with an in-page bit of code, I would think). None of this is terribly difficult; there's just no will at WMF to do anything about talk pages being a "user-hateful" experience for the visually impaired.

PS, if I may ramble about technically related matters: Similarly, the parser needs to stop ever, on any page, treating : or ; as d-list markup, unless they actually form a valid d-list. According to the HTML specs, any <dt> must be followed by another (for multi-term entries) or by a <dd>; and a <dd> must be preceded by another (for multi-definition entries) or by a <dt>; ergo, any purported d-list that does not have both at least one <dt> (;) and at least one <dd> (:) is not a valid <dl>, and instead should be re-parsed as <div>s with styling to produce boldface or indentation, respectively. This is actually a pretty simple coding job, but after just shy of 20 years, it's clear we're never going to get it, at least not without major staffing and priority changes on the development side. There are many other "basic HTML screwups" going in MW that many of us have flagged for repair since the mid-2000s, with no action on them.
 — SMcCandlish ☏ ¢ 😼  20:59, 2 December 2020 (UTC)

"This seems like it could seriously break the flow": Sure, and this is why it's not advisable to do it with anything likely to generate responses. E.g., I just did an "injection" a few minutes ago [4], a one-liner note that requires no reply and isn't likely to generate one. If it did, then it would be sensible to refactor it back into the conventional article flow (adjusting the indent level), and if that put it way below another long branch of the thread, maybe refactor one's initially injected comment to have a quote making the context clearer (especially if that once-injected post was pretty context-free in its content, having at the time been immediately after the OP); this post is itself an example of that quote-for-context style. :-)  — SMcCandlish ☏ ¢ 😼  20:59, 2 December 2020 (UTC)

Sure, but all that refactoring seems like a lot of work that could be avoided by just putting replies in chronological order to begin with...hence the kind of thing most users won't bother doing. [N]ot advisable to do it with anything likely to generate responses seems to rule out the idea of interjecting with a substantive point, as you said Flyer22 Frozen did here. Substantive replies often generate more replies, that's kind of the point. —Sangdeboeuf (talk) 22:30, 2 December 2020 (UTC)

So, isaacl, going by your "uncollaborative" logic (which would apply to a whole lot of people who are actually being collaborative at various talk pages), the editor who made this post should not have placed it there? It should have been after my second post? And when replying to you right now, this post should be where it is instead of being ahead of SMcCandlish's post with correct indentation...even though placing it ahead of SMcCandlish's post with correct indentation would be clearer and I would not have needed to address you by your username? No need to ping me if you reply. In fact, please don't. Flyer22 Frozen (talk) 07:49, 2 December 2020 (UTC)

No, your post should have gone above SMcCandlish's (or "ahead of", if you prefer), with correct indentation, just as you say. That's why I specified that posts with the same indentation level should be in chronological order. By replying to Isaacl, you would be adding another level of indentation. This is shown in the example at Help:Using talk pages#Indentation. Sangdeboeuf (talk) 08:44, 2 December 2020 (UTC)

As per Help:Using talk pages § Indentation, since you were replying to me, your reply should be one nesting level deeper than my reply, immediately after my comment. SMcCandlish was responding to Sangdeboeuf's comment, after I had already replied, and so placed his comment below mine (and, if there had been any replies to my comment at that point, below those hypothetical replies). I didn't go through this aspect, since you had already mentioned Wikipedia:Indentation in the talk page discussion linked to in the original post. It's not necessary to ping me. I'm not sure why you only asked me not to ping you. isaacl (talk) 09:00, 2 December 2020 (UTC)

Isaacl, and this is the type of non-chronological order thing I was talking about and how it's common. I was very clear that non-chronological order is not always followed and pointed to such an example at the Transsexual talk page. Even here in this thread, you and another posted ahead of my "07:54, 2 December 2020 (UTC)" comment below. For another example, we can look at this section of Jytdog's talk page, where there are successive comments. If I wanted to reply to Roxy the dog, I would do so right beneath Roxy the dog's post with correct indentation. Not at the bottom. And for those even higher up, I obviously would not reply to them all the way at the bottom. Posting all the way at the bottom to respond to a comment in cases such as these is not the clearest option. It is not the best practice. That is why SMcCandlish recently posted ahead of another's comment here at Talk:Transsexual. Two posts by the same editor are dated "11:52, 2 December 2020 (UTC)." But did SMcCandlish post behind the second one? No. SMcCandlish cut in between them and responded directly to the one he was focused on. Similarly, at Talk:Transsexual, I see no solid reason why my post needed to go below this post. Sure, there is indentation and I could have used the editor's name to make it clearer who I am responding to. But, other than the indentation of the post I jumped ahead of, that comment seemed like a general statement addressed to everyone. Not just to one editor. And this has been echoed in this thread by SMcCandlish. So I chose to reply ahead of it. Editors do stuff like that all the time without issue. As for asking you not to ping me? It's because I had replied to you and pinged you while doing so. I didn't want to be pinged back. And, of course, I haven't pinged you to this section again since you stated that I don't need to. Flyer22 Frozen (talk) 00:46, 3 December 2020 (UTC)

Yes, when I respond to a specific comment at a specific list indent level, I make my reply as a nested list item within that comment. If there are any existing replies to that specific comment that already started a nested list, I continue the nested list by adding a new list item to the end of it, which is the standard practice as documented at Wikipedia:Indentation. Consider the reverse case: say you've carefully worked out what you feel is a pithy, concise comment in response to an editor. Now another editor has chosen to insert a comment ahead of yours. That other editor is implicitly telling you that they consider your response to be "some trivial one-liner update". Maybe you're right; maybe they're right. Generally, though, it does little harm to leave the first response in its place, thus avoiding acrimony. I'm old school, having started editing before echo notifications. I won't ping editors already participating within a discussion. I was just uncertain why you'd think I'd ping you, but didn't seem to make the same assumption about anyone else.isaacl (talk) 01:28, 3 December 2020 (UTC)

For another example, we can look at this section of Jytdog's talk page ... If I wanted to reply to Roxy the dog, I would do so right beneath Roxy the dog's post with correct indentation ... And for those even higher up, I obviously would not reply to them all the way at the bottom. This is exactly what both Isaacl and I have been saying with regard to level of indentation. No one is suggesting that all new comments should go at the bottom of the page or thread.

Editors do stuff like that all the time without issue. Obviously it was an issue this time, hence this discussion. —Sangdeboeuf (talk) 01:47, 3 December 2020 (UTC) (edited 01:51, 3 December 2020 (UTC))

(edit conflict) Isaacl, I understand what you're saying. My point is that what I did is not some uncommon thing and that it usually does not cause problems. If my comment was some brief update about the status of the article, I maybe wouldn't call it trivial. And I did not state that the update in question is trivial; I agreed with SMcCandlish, who clearly didn't mean "trivial" in some offensive way. But I would not mind someone placing their comment ahead of mine in such a case. And, actually, this has happened times before. I think the type of comment matters. In this case, it's objective to state that the comment was a one-liner update. That's not a subjective view of the comment's value. I also started editing before the WP:Notifications feature was implemented. I remember discussion about it taking place on my talk page. Flyer22 Frozen (talk) 01:49, 3 December 2020 (UTC)

All I'm saying is you've made a decision that your comment has greater primacy such that it should be placed ahead of earlier repliers, and that can create friction, whether or not they choose to state an objection. It's akin to interrupting someone in a meeting, forcing them to wait to contribute. (Yes, it happens in many written and in-person discussions, and will keep happening.) isaacl (talk) 02:30, 3 December 2020 (UTC)

Wasn't about primacy for me. I did not think of the comment in that way. While SMcCandlish stated "trivial", I stated that "other than the indentation of the post I jumped ahead of, that comment seemed like a general statement addressed to everyone. Not just to one editor. [...] So I chose to reply ahead of it." It was about directly responding to the post that concerned my comment, similar to how SMcCandlish choosing to cut in between two posts with the same timestamp by a single editor here here was about him replying to the post that directly relates to what he had to state. If you want to see me jumping ahead of that one-liner update as a primacy matter, we disagree. And I'm sure that SMcCandlish does not view his comments in a primacy way in the instances that he mentioned jumping ahead. Flyer22 Frozen (talk) 02:42, 3 December 2020 (UTC)

You are essentially second-guessing the replier, indicating that you think their chosen indent level was not chosen accurately. Again, consider the reverse: if you've carefully chosen where to place your reply, will you be discontented that someone else is imposing their view on yours? And even if the other person is right, is there any significant benefit that makes the risk of upsetting you worth it? isaacl (talk) 02:51, 3 December 2020 (UTC)

Second-guessing the replier? Imposing my view on their view? What? No. I've considered the reverse, and I already gave you my answer on that. As for "is there any significant benefit that makes the risk of upsetting you worth it?", it was never an issue for me until now. And it wasn't much of an issue before an editor decided to lecture me about it at the article's talk page and later bring to the matter here. Furthermore, the two editors who made it an issue are editors I have a less-than-stellar history with, and not because of jumping ahead of their posts. But given my history with the editor I jumped ahead of in this case, it would have been better that I had not jumped ahead. Add: Also, Isaacl, you and I never agree. So I'm not surprised that you are making a bigger issue out of this than it is. Flyer22 Frozen (talk) 03:02, 3 December 2020 (UTC) Tweaked post. Flyer22 Frozen (talk) 03:07, 3 December 2020 (UTC)

SMcCandlish, you stated, "Flyer was making a substantive point, while the other was just providing some trivial one-liner update." Exactly. That is why I made the choice I did in that case. Flyer22 Frozen (talk) 07:54, 2 December 2020 (UTC)

@Flyer22 Frozen and Isaacl: To make it clear to whom you are replying, the {{reply to}} template may be used, which (if used properly) generates a bell notification. If the user concerned desires not to receive those, the variant {{no ping}} is available, as demonstrated at the start of this post. --Redrose64 🌹 (talk) 16:09, 2 December 2020 (UTC)

Thanks. Yes, I know. I was just caught up in my points. Flyer22 Frozen (talk) 00:46, 3 December 2020 (UTC)

Alternatively, you can just mention the editor by name, without any links or templates. isaacl (talk) 01:28, 3 December 2020 (UTC)

I mentioned that in both my "07:49, 2 December 2020 (UTC)" and "00:46, 3 December 2020 (UTC)" posts. Flyer22 Frozen (talk) 01:57, 3 December 2020 (UTC)

Note my response was to Redrose64, as it is part of the nested list one level in from his comment. (You actually did link to my user name in your 2 December post.) isaacl (talk) 02:39, 3 December 2020 (UTC)

Doesn't negate the fact that I mentioned it as an option. Flyer22 Frozen (talk) 02:45, 3 December 2020 (UTC)

I was replying specifically to Redrose64's last sentence. I was clarifying that there is no need to use the {{noping}} template, if the user name is written without linking to the user's page. For conciseness, I didn't choose to highlight that the text in your 2 December post is in alignment with this (although you did in fact link to my hypothetical user page); my apologies. isaacl (talk) 02:55, 3 December 2020 (UTC)

I heard you the first time. And I never stated or implied that you needed to highlight a thing. Flyer22 Frozen (talk) 03:07, 3 December 2020 (UTC)