Wikipedia talk:Did you know

From Wikipedia, the free encyclopedia
  (Redirected from Template talk:Did you know/Queue)
Jump to: navigation, search


"Did you know...?" template
Discussion WT:DYK
Rules WP:DYK
Supplementary rules WP:DYKSG
Nominations WP:DYKN
Reviewing guide WP:DYKR
Preps & Queues T:DYK/Q
Currently on Main Page
Main Page errors WP:ERRORS
Removed hooks WP:DYKREMOVED
Archive of DYKs WP:DYKA
Stats WP:DYKSTATS


This is where the Did you know section on the main page, its policies and the featured items can be discussed.

Transclusions on nomination page, suggested solution[edit]

background discussion from WT Village Pump (technical)

This problem has existed for a few months on Template talk:Did you know. Once you get down to the newest subsection dates, the templates don't transclude very well. We were told back in September that the problem was that page is exceeding Template limits Post expand include size. At that time, we had a large special occasion holding area for various special events. The holding area has very little in it now, and the number of nominations we have are otherwise a lot less. The problem is worse than ever. Regardless of what is causing this, can it be fixed? As the internet expands, so does the size of everything programmed into it, and DYK won't be the only ones this happens to. How do we fix it for the future? — Maile (talk) 21:52, 30 October 2016 (UTC)

Abandoning Template limits would be a decision that would need to be taken at WMF level, and they're vanishingly unlikely to authorise it since it's not a bug, it's an intentional feature to prevent DDOS attacks. The way around it is to use fewer transclusions; remember that each DYK nomination includes {{DYK conditions}}, {{DYK nompage links}}, {{main page image}}, {{DYKsubpage}} and {{DYKmake}} plus whatever else the reviewing bot adds, so each transcluded nomination counts as six or more transcluded templates. ‑ Iridescent 22:02, 30 October 2016 (UTC)
The standard fix for template size problems is to substitute templates and to remove any nested transclusions. Jo-Jo Eumerus (talk, contributions) 22:13, 30 October 2016 (UTC)
Exactly how would DYK go about that? — Maile (talk) 22:23, 30 October 2016 (UTC)
On a quick skim, the {{DYK conditions}} template doesn't appear to have any great use and has three nested templates of its own, so getting rid of that would save four templates-per-nomination immediately (with the current 53 nominations, that's an instant saving of over 200 templates, which will probably solve the problem on its own). Basically, go through the five templates I list above, and anything that's not actually both essential to your process, and essential that it remains unsubstituted, think about whether it would be possible to do without it or enforce substitution of it. You could also probably shave quite a bit off by ruthlessly enforcing a "no untranscluded templates in discussions" rule, and clamping down on anyone who uses {{od}}, {{tq}}, {{done}} etc in discussions. ‑ Iridescent 22:41, 30 October 2016 (UTC)
Neither {{DYKmake}} nor {{DYKnom}} should be of concern, since they're commented out. I imagine that increased use of the {{DYK checklist}} for reviews is also contributing to the problem. Does the use of the {{*}} template contribute to the problem or not? It's currently being used by the DYKReviewBot. One template that we absolutely need to retain is the {{DYKsubpage}} template, since it is the final substitution of that template that closes the nomination. BlueMoonset (talk) 04:14, 31 October 2016 (UTC)
Every time this happens I hope it will finally be the motivating factor to do the seemingly obvious and move the reviewed/approved nominations to a different page. DYK that nobody can read that thing on a phone? Opabinia regalis (talk) 05:12, 31 October 2016 (UTC)
Oh, and the answer is yes, templates that are actually transcluded all count, so if there's a bunch of templated bullets then that's definitely contributing. Opabinia regalis (talk) 05:19, 31 October 2016 (UTC)
Then calling Intelligentsium, to see whether the templated bullets can come out of the reviews done by the DYKReviewBot, and any other avoidable templates. Also pinging John Cline, who created {{DYK conditions}}, to see whether there is some way to get the job done more efficiently templatewise, assuming that the job still needs to be done. I have no idea whether the 2015 conversion of {{NewDYKnomination}} to invoke a Module with the same name rather than do the work in a template would have affected the need for DYK conditions or not. BlueMoonset (talk) 07:52, 31 October 2016 (UTC)
Thank you BlueMoonset for your kindness and astute manner; inviting me to join this discussion. I was not aware of it until now, nor did I know anything of the circumstances forbearing it. I am therefore disadvantaged from giving an answer; ore the research I've yet to do.
When I catch up with the topic, however, I am confident that the answers being sought will be found.
If I wasn't so Spock-like, I can imagine myself getting all butt-hurt about not being notified of questions being asked of these templates, perhaps others as well. I was told in the past, things about my style in writing; and before that, of many ill effects that style was cursed to engender. Here, it seems that enduring months of template malfeasance was preferable to enduring discussion where I would invariably be. Being all Spock-like; and all: I feel terrible that this may in fact be. I really do.--John Cline (talk) 11:24, 31 October 2016 (UTC)

We have to do something soon. The nominations page is quickly dissolving into nothing but wikilinks with no transclusions. Yes, I know the Prep/Queue page has always been used as the holding area. We cannot control how other people edit nomination templates - i.e. large amounts of text, template comments, additional image suggestions. The way it has always been is not the way that will work for the future.

Below is my suggestion. — Maile (talk) 16:59, 1 November 2016 (UTC)

Suggested solution[edit]

  • The nomination page stays but only includes those which have received no approval whatsoever.
    • Reviewers who only are only interested in non-problem hooks have less to scroll through to find something of interest.
    • This would make a cleaner page for first-time reviewers who get confused by the glut we now have.
  • The Prep/Queue page stays exactly like it is, nothing changes about how it works.
  • A new subpage is created where any nomination that receives an approval is moved there by a bot (or human).
    • Special occasion holding areas, including April Fools' Day, appears at the bottom of this page. It stays consistently as is, in the fact that hooks are only moved here after approved on the main nominations page.
    • Prep promoters draw from this page.
    • Reviewers who like to check for problem areas on approved nominations look here.
    • Any disputed approval and any post-approval ALT hooks added are worked out on this subpage
    • Any hooks pulled from Prep, Queue, or the main page are put back here.

Please add comments below

Comments[edit]

  • Yep, sounds like an excellent idea to me too. Opabinia regalis (talk) 21:44, 31 October 2016 (UTC)
    The bot will now used the substed the template {{*}} - it's weird that the page exceeds the transclusion limit so easily though. The previous time involved {{hat}}, {{hab}} which were being used more than once per nomination, and had several transclusions underneath as well, whereas {{*}} seems to be just a Unicode character. However I think it may be a bit of a hassle to move hooks between two pages - if you move them the moment they are seen by a human, you would probably quickly get the same problem on the second page, but moving them back and forth would be a huge hassle. Intelligentsium 00:59, 1 November 2016 (UTC)
    I know you have worked hard on the bot, but we didn't have this problem before it was activated. If the problems with it can't be ironed out soon, I think we are just going to have to retire it. That would surely be a better solution than having two separate nomination pages. Gatoclass (talk) 17:30, 1 November 2016 (UTC)
    I agree it's more trouble, but I think having a place where approved noms are gathered, for further intense scrutiny by the "eagle eyes", will extremely helpful, as well as solving the overflow problem. EEng 18:58, 1 November 2016 (UTC)
    The bot had been down for a few weeks, and this problem continued even in its absence. — Maile (talk) 19:07, 1 November 2016 (UTC)
  • Support, and... can we add the provision that nom page stays open until the bot closes it (maybe at the moment the hook moves to the main page, or -- better -- at the moment the hook comes off the main page)? EEng 18:58, 1 November 2016 (UTC)
    And have any dialogue on pulled hooks happen there, so that any nominator, reviewer, or other participant on that nomination would be aware of it as long as they watch-listed the open template. I don't know the mechanics of having a bot close the nomination, but it's worth asking Shubinator if that's possible to do in conjunction with whatever else DYKupdatebot does. — Maile (talk) 21:43, 1 November 2016 (UTC)
    Exactly. I'd also like to add a further suggestion that adding the green tick (which is presumably what will trigger the bot moving the nom page to this new "approved area") should always be accompanied by a tentative designation of exactly one of the (possibly several) ALTs as the one to used. Further discussion in the "approved area" might change that, but this way once the nom moves to the "approved area" there's just a single ALT that the "eagle eyes" (our precious editors who focus on quality control) will have to focus on checking. EEng 01:01, 2 November 2016 (UTC)
    If closing a nomination just involves subst'ing the DYKsubpage template and marking it as passed (with humans responsible for moving the noms between the various pages, except for queue -> main page), DYKUpdateBot can do this while promoting the set (not while taking it down). As BlueMoonset noted, the bot will not know about comments that should go into the "2" field. With this model, how will folks know which admin promoted the nomination into the queue? Shubinator (talk) 01:16, 2 November 2016 (UTC)
    The promoter simply posts on the bottom of the nom page e.g. ALT1 to Prep4 (without image). ~~~~. The bot closes the nom as it swaps the hook set onto Main Page (i.e. at the same time the credit boxes are posted to creator/nominator talk pages) and the 2= could be Swapped onto Main Page 0800 22 Jan 2017 UTC. This way, all concerns prior to the actual main-page appearance can be discussed on the still-open nom page, where it belongs; concerns arising after that time have to go through ERRORS as now.
    I think it would be ideal if, while we're at it, we changed the bot actions of posting credits to editor Talk, and closing the nom pages, to the moment the hook set is swapped off of the main page. Then the nom page really stays open for the entire life of the hook, "cradle to grave". But I recognize this might be more complex to do. EEng 02:14, 2 November 2016 (UTC)
    Mark me as opposing the further suggestion: the reviewer should feel free to check and approve as many interesting hooks as seem appropriate and are properly support in both article and sources, but not all reviewers are the best judges of which is the best, and sometimes the person assembling a prep set will pick one good hook over another good hook because it better balances the prep set. To limit it to exactly one hook of the reviewer's choice also reverses the deference we've given to the nominator regarding proposed hooks.
    As for the promoter, may I suggest that the promoter be required to fill in the 2 field with their promotion message? The bot's closing of the page will cause the time of closure to be added to the page. BlueMoonset (talk) 03:46, 2 November 2016 (UTC)
    The reviewer can approve as many ALTs as he or she wants, but (my suggestion is) that just one of them will be designated, tentatively, as the one that will appear. Further discussion might change that, selecting a different ALT, but starting at this point there would be only one ALT on the table at a given time for a given nom, so that attention can focus on it for error-checking and so on. To increase quality and reduce errors appearing on Main Page, it's essential that the checking process begin further upstream than it does currently i.e. currently this doesn't start until Prep, and now it can start when the nom is moved to this new "approved area". But it needs to focus on one potential hook at a time; if multiple hooks are in play, the checking just can't be thorough. I don't buy that this constrains prep set assembly enough to outweigh the advantages, and again I say that the designation of a single hook is only tentative, subject to change. EEng 05:40, 2 November 2016 (UTC)
  • Support. It will make it so much easier to scroll through the set of approved hooks when building prep sets. Yoninah (talk) 19:01, 1 November 2016 (UTC)
  • Support. This is an excellent suggestion. -- Notecardforfree (talk) 19:10, 1 November 2016 (UTC)
  • Support sounds fine LavaBaron (talk) 01:55, 2 November 2016 (UTC)
  • Support, since I have no major objections. I do have a couple of doubts though. First, my understanding of the technicalities is not great, but if this problem is arising from people using too many templates without substituting them, it would seem that this is relatively a small fix: and that unregulated use of templates in the review process is going to create a problem again sooner or later. So, wouldn't it make sense to create some guidelines for folks editing the nomination pages, to help with this? Second, I find that very many of the hooks that need reviewing at any given time, and indeed the ones requiring the most attention, are not "fresh" nominations, but those that have been reviewed already, but require a new reviewer for whatever reason. @Maile66: where would these fit in your scheme? Vanamonde (talk) 04:20, 2 November 2016 (UTC)
    Vanamonde93 Regarding the guidelines, it does begin to be instruction creep. We cannot control what editors really do, no matter how many guidelines we write. As we experience on this talk page, a lot of editors aren't reading the guidelines anyway. So, we can spend a lot of time spinning our wheels and complaining on the talk page about those who do what they want, but we cannot control others. As to your second question, perhaps I wasn't clear. The minute a nomination receives a passing tic, it gets moved to the new page. There it stays, and any further issues or comments happen on that page. That means turn-around ticks on review questions, pulled hooks that were already promoted. Anything. EEng has suggested we keep the template open until when/if the nomination is off the Main page. Keeping it on that page does not close out the nomination, but leaves it there in a way that anyone with a given nom template on their watch-list will be aware it needs attention. New (first time) reviewers will have an easier time with unreviewed templates than figuring out why an already approved nom is in the midst of revision for one thing or another. — Maile (talk) 13:20, 2 November 2016 (UTC)
    @Mail66 and Gatoclass: I think you're right about the guideline creep, but I didn't necessarily mean another page or another bullet point in the current set. What I mean is that we can do minor things that should still add up to something substantive. For instance, some folks mentioned templates (DYK checklist) that are only used at DYK: we can add a note to the documentation saying that they must be substituted, and also possibly have a bot substitute them every time. We can add to the DYK template edit notice, asking people to minimize their use of templates. And so forth. I imagine that other folks can think of other options. Vanamonde (talk) 16:03, 2 November 2016 (UTC)
  • Comment: Per Vanamonde, I think what we need to be doing is working out why this problem is occurring, and take steps to eliminate or minimize it, because it never used to occur even with 350 nominations and now it's occurring with just 150. If the number of nominations builds up again, the problem will recur. Gatoclass (talk) 06:54, 2 November 2016 (UTC)
  • Comments: Keeping the nom open up until (and even through) the main page appearance seems reasonable to me, so long as the technical template issues can be addressed. I think it is fine and appropriate for a reviewer to choose a hook, but also to leave the choice open to the promoter, but I would like to see some reasoning posted. I've had a few cases where I've wondered why a hook was chosen (or not chosen), which I find frustrating and yet asking the promoter every time could get awfully intrusive given the relatively small group of set builders. Having another approved hook available is also useful in cases where an issue arises, because sometimes swapping hooks rather than pulling might be reasonable and appropriate. I would also like to see an explicit requirement that all ALTs be reviewed because I've had at least one case of offering several and only the first being reviewed / promoted on the presumption it was my preference (an incorrect assumption on that occasion, but understandable and arising from poor communication on my part). EdChem (talk) 07:15, 2 November 2016 (UTC)
    I believe you make a valid point about the ALTs being reviewed. I've noticed the same thing. If all hooks are not reviewed, then the review isn't complete. It does a disservice to both the nominator and the promoter. Also, I have no problem with the promoter leaving a small note on the template about why a given hook among several available was promoted. — Maile (talk) 13:54, 2 November 2016 (UTC)
    Yes, Maile66, with my Timothy N. Philpot nomination, only the ALT0 has been reviewed and it was my fall-back option if all the others (which I think are more interesting) are rejected on undue negativity grounds. So, I posted here at WT:DYK requesting input, but the thread attracted no responses. I'm not sure what to do because the rules technically require all ALTs to be reviewed but making an issue of my case will focus on the reviewer, who is behaving as others do and does a lot of DYK work. EdChem (talk) 14:22, 2 November 2016 (UTC)
  • Support Good idea. There are currently several structural problems and the proposal looks like a sensible way forward. If there isn't one already, it would be good to have a page to document the process flow so that it's clear how a nomination progresses from page to page. Andrew D. (talk) 12:33, 2 November 2016 (UTC)
  • Comment I've just converted Template:DYK checklist to use Module:DYK checklist, which makes each checklist take up about half the post-expand include size that it did before. This has resulted in 12 more nominations being visible at the bottom of Template talk:Did you know, but we are still quite a bit over the limit. — Mr. Stradivarius ♪ talk ♪ 01:40, 7 November 2016 (UTC)
  • Comment I find it worrying that there isn't one bit about why this is happening - I don't mean technically, I mean temporally. The number of new articles continues to decline, there appears to be no (major?) change in the number of noms being posted per day, and I don't see anything about the technical limit being changed. This is the only time I've noticed it - it seems to have happened before but I assume for a short period? So why now, in 2016? Maury Markowitz (talk) 15:38, 9 November 2016 (UTC)
    • @Maury Markowitz: I think the cause of this might be the bot, which adds a bunch to text to every nomination. Pppery 20:17, 9 November 2016 (UTC)
      Ahh, so it's based on the total text, not the number of transclusions? Maury Markowitz (talk) 20:44, 9 November 2016 (UTC)
      Yep. (It's actually more complicated than that, but) Pppery 21:01, 9 November 2016 (UTC)
Maury Markowitz We have actually been discussing this a great deal on this talk page. A year or two ago, our individual reviews weren't so complicated, except in the case of drawn-out threads. Most were pretty brief. But graphics, text, little check templates, and a lot of thing have increased the size of the individual nominations transcluded. We also now have the bot that does a preliminary review. However, that bot was down for several weeks, and the problem continued. When we pushed it to the limits, the visual kind of went kaflooey. Think of what happens with your browser if the cache doesn't get cleared for a long time - eventually things aren't working right on a given page. It's kind of like that. Have you read the green hatted text at the top? We've exceeded our Template limits Post expand include size, and only WMF can give us more. And that isn't likely to happen, because WMF has safeguards in place to prevent a Denial of Service attack. Little things help some, like not putting checkmark templates on the nomination. But in the long run, we'll be pushing the limits and need to come up with a solution. — Maile (talk) 21:09, 9 November 2016 (UTC)
I did, and also visited the link you have here. Neither stated this clearly, nor included any specific numbers or examples. Maury Markowitz (talk) 21:24, 9 November 2016 (UTC)

Summary and implementation?[edit]

So it's one thing for there to be a lot of support, but it's another for someone to do it. What next? EEng 01:30, 7 November 2016 (UTC)

  • We do need to work the mechanics of this. And we need a bot to help, perhaps Shubinator's bot or something already in existence that just need extra code for this. — Maile (talk) 02:00, 7 November 2016 (UTC)
"Just..." EEng 02:20, 7 November 2016 (UTC)
The nomination page seems to have returned to normal. Has someone actually resolved the problem, or is this as the result of some faulty nomination being promoted and archived? Cwmhiraeth (talk) 09:56, 13 November 2016 (UTC)
Isn't that interesting? — Maile (talk) 12:29, 13 November 2016 (UTC)
Yes, but it's only a matter of time before the problem comes back, and there were other good reasons for doing this. Thus I hope the extensive paid and pampered staff in charge of doing things like this get right to it. EEng 05:37, 15 November 2016 (UTC)
EEng, the problem did not fade away gradually but came to a sudden conclusion. One moment there were a host of nominations not properly displayed and a few hours later, there were none. This happened, as far as I can tell, late on the 12th November or early on the 13th. I think it was due to a problem nomination which was promoted and archived at that time, and will likely not recur. I suspect, without good evidence, the Moses Bensinger nomination. Cwmhiraeth (talk) 13:36, 15 November 2016 (UTC)

──────────────────────────────────────────────────────────────────────────────────────────────────── It's been a problem in the past and (I repeat) there were other reasons for doing this. (Commenting mostly to keep the thread alive.) EEng 19:40, 20 November 2016 (UTC)

There is more than one theory about why this happened, and we don't know for sure. But at this point, it's not happening. This would certainly support the idea that exceeding Template limits Post expand include size was not the problem after all, or this would still be happening. We have recently seen how one background edit can affect DYK like the bottom card being removed from a house of cards. We don't know why this happened, and we don't know why it stopped. What I have proposed here about a separate page for approved nominations would be a large undertaking to implement and maintain, unless there was a bot involved. I think the above Supports are mostly because it would be easier for promoters if we had a separate page for approved nominations. I yield to the majority, however this turns out. But we still need to get it implemented if we go with it. — Maile (talk) 13:15, 21 November 2016 (UTC)
  • @BlueMoonset, Cwmhiraeth, and EEng: I just noticed right above the special holding area, we are having this transclusion problem again. And it's getting worse. It magically clears up for a few days, and then clogs up again. One of the great mysteries of the universe. — Maile (talk) 22:27, 27 November 2016 (UTC)
  • Maile, the number of active nominations (and therefore templates) has been steadily increasing, so it's natural that we'd run into the transclusion problem. It had been happening with 160 or more active nominations; now it's happening with 250/260 or more. That's quite a difference. If we had four or five prep/queue sets built at any one time, we wouldn't be having transclusion problems at the moment, though if the number of noms continues to build, we would regardless. BlueMoonset (talk) 06:41, 28 November 2016 (UTC)
  • Let's not let the flame flicker and die. EEng 04:51, 19 December 2016 (UTC)

All hail Wugapodes![edit]

I saw this and whipped together a script that managed to do the first part of the reorder: put any hooks that have been accepted onto a different page (and remove all the accepted/closed ones from the nom page). You can see the output on User:Wugapodes/Did you know/Approved hooks and User:Wugapodes/Did you know. If people like this and think this is something I should continue working on, I can make it so that the holding queues are on the other page like suggested. Let me know if this is helpful or not. Wugapodes [thɔk] [ˈkan.ˌʧɻɪbz] 07:12, 19 December 2016 (UTC)

YES! YES! YES! Everyone, please review what Wugapodes has done so far. (Wugapodes, by "holding queues" do you mean the special holding areas for e.g. holidays?) Also, let's all remember that this was a package of ideas about changing the sequence of events in review, approval, and promotion, especially with regard to when nom pages get closed and so on. Let's make maximum use of this opportunity to implement as many good ideas for improving things while this sucker wonderful volunteer Wugapodes is willing to dig in and do the work. EEng 23:05, 19 December 2016 (UTC)
Ah, yes, I did mean the holding areas. Poor phrasing on my part. Be sure to let me know of any ways I can procrastinate writing help. Wugapodes [thɔk] [ˈkan.ˌʧɻɪbz] 23:20, 19 December 2016 (UTC)
I like what you've done. Please pursue this — Maile (talk) 00:54, 20 December 2016 (UTC)
It looks good. One thing I noticed, at the moment the nomination José Antonio Raón y Gutiérrez for 23 October appears on your approved list. The template is splattered with ticks but the nomination has not in fact been approved. Cwmhiraeth (talk) 12:22, 20 December 2016 (UTC)
Awesome work! But yes, we do have to decide how to deal with "challenged" nominations, where the approval is superseded by a later comment. Also, perhaps approved nominations should remain visible until they reach the main page, to encourage discussions to occur there and not on the main DYK talk page if they get pulled from the queue. Antony–22 (talkcontribs) 01:18, 21 December 2016 (UTC)
Wugapodes, great idea. For your algorithm: it has always been the case that the final icon rules; that's how Shubinator's DYKHousekeepingBot builds the List of DYK Hooks by Date table on the DYK queue and nomination pages. So if the final icon of the six allowed is one of the two ticks, then the nomination goes on the approved page; if red arrow, question mark, slash, or X, it goes on the regular nominations page. I queried Shubinator a couple of days ago about updating his bot to combine the contents of the current noms page and a new Approved noms page, and he hopes to have something ready to test by the end of the week. We'd need to decide on a name/location for the approved page: I would suggest an /Approved page directly below the current nominations page (Template talk:Did you know/Approved). I don't believe we want to use the word "hooks" in the page name because each entry is an approved nomination, not an approved hook. Finally, because Special occasion hooks are supposed to be approved, they should be kept on the Approved page but in their own section where the new moving bot should probably not be allowed to make modifications. We may want a stub of a Special occasions section on the regular nominations page, also where new the bot should not go, with much the explanation that is there now, along with a link to the approved special occasion nominations. BlueMoonset (talk) 02:08, 21 December 2016 (UTC)

──────────────────────────────────────────────────────────────────────────────────────────────────── Let me suggest the following:

  • Once the green tick appears, move the nom to the Approved page.
  • If the green tick is later overridden, don't move it back off the Approved page -- too confusing and doesn't happen so much that it matters. Thus the Approved page is really the "got approved at some point even if maybe it's not currently approved". This way more eyes get on a "troubled" nom, and that's a good thing.
  • I thought about having a delay of X hours, after the green tick appears, before moving the nom to the Approved page, to give a little time in case the original tick is going to be overridden, but again I don't think it's worth the complexity (and sometimes we're trying to rush something through the process, so we don't want a delay).

Other points:

  • Keep the nom page open until the hook is swapped off the main page. In fact, all the ancillary stuff that currently happens as the hook set is swapped onto the main page (closing nom page, handing out credits to user talk pages) can be delayed together to the swap-off.
  • I'd like to make a plea for not importing, to the Approved page, the date structure of the main nominations page. Please, just add newly-ticked noms to the end of the page, so that those doing QA can simply watch for new stuff at the end. Please, please. This obsession with maintaining some kind of priority structure based on "date of creation or date expansion began" is completely stupid. (Having special-occasion holding areas is fine, and of course prepbuilders are free to jump around the Approved page in selecting hooks.)

EEng 04:20, 21 December 2016 (UTC)

@BlueMoonset and Antony-22: My interpretation of the proposal was like EEng's, even noms approved but later challenged would be on the approved page, and I think that's a good thing per EEng. Though if we'd like to discuss which is better, It's an easy change. @EEng: I'm not sure what you mean with your first other point, are you proposing a change to the way hooks get promoted to queues or is this something I can change in the script? I agree with your second point, and was how I wanted to set it up but I decided to not rock the boat too much. If others like that idea I would be glad to make that change. Wugapodes [thɔk] [ˈkan.ˌʧɻɪbz] 04:58, 21 December 2016 (UTC)
The current procedure is that the prep builder selects a hook from a green-ticked nom page and adds it to a prep set; at the same time he/she closes the nom page (by changing some parameter in the enveloping template, and subst'ing it). Unfortunately, this means that if there's later trouble with the hook, there natural venue for discussing it (the nom page) is no longer available -- this is the main reason you see so much "pull" discussion at Talk:DYK. Also, in the current procedure, as the bot swaps a Queue of hooks onto the main page, that's the moment that the bot goes to the talk pages of the various involved editors for each article, to post congratulations.
My idea is this: when a nom is selected to donate a hook to a prep set, the nom is no longer closed; instead the prep builder simply posts a comment at the bottom of the nom, "To Prep 4 (without image)" or whatever. After the prep set becomes a queue set, and then is eventually swapped onto the main page, no credits are given as they are now. Instead, 24 hours (or whatever) later, as the hook set is swapped off the main page, at that time the bot passes out credits to editors (as it does now, just 24 hours later than it used to), plus (a new job for the bot) the nom page is finally closed. This way, the nom page remains open "from cradle to grave" for discussion of problems, no matter how late they arise. Also (hate to say it) if the hook is modified during its main page appearance, the credits that appear in various places will quote the final hook as of the moment it's swapped out, not the original (presumably inferior) hook that was swapped in at the beginning of the 24 hours. EEng 05:15, 21 December 2016 (UTC)
EEng, can I suggest that we take this in stages? There's an immediate issue: the nominations page is too big. There's going to need to be a lot of coordination between the new Wugapodes bot and the Shubinator bots. DYKHousekeepingBot is going to need to keep track of how many nominations there are between the pages, and be able to count both kinds (approved and not approved) on both pages. It's also going to have to figure out which nominations out there aren't yet transcluded, which involves checking both pages. Adding the rest into the initial separation stage is going to cause all kinds of delays in the separation. Let's concentrate on getting the pages separated before redesigning the whole process.
As for keeping the nominations open after promotion, I think this is going to cause more problems than it solves. Assuming we do keep the nominations open until they've been promoted to the main page and left it—remind me how we make sure that an open nomination isn't in one of the preps or queues or on the main page so it doesn't get promoted multiple times?—DYKUpdateBot (which puts the notifications on article and user talk pages) will have to do the close. The notification of promotion is now less friendly: instead of being told that the article you nominated is now on the main page and you can go see it there, you get notified after it has left the main page, so you probably missed it. I think you're being optimistic about the number of people who will see formerly approved hooks on the approved page; reviewers generally won't go there because the hooks are supposed to be approved. Special occasion hooks that run into trouble rarely find reviewers after the fixes have been made because they're in an area where only approved hooks should be; I think we'll be looking at the same thing here. Finally, the set builders select from both green- and gray-ticked nominations. It's the tick that counts, not so much its color. (With the gray AGF ticks, more care should be taken.) BlueMoonset (talk) 05:51, 21 December 2016 (UTC)
  • Yes, for "green tick" I should have said "green or gray".
  • I don't care too much when the credits are handed out -- do them during swap-in if you want -- but I feel strongly about keeping the nom page open until the final swap-out. Right now Talk:DYK is far too cluttered with discussions that should be going on back on the nom page, where all the relevant background already is. [Later-added point]: I do see the value of having involved editors notified at the moment of swap-in to the main page, since they can monitor for vandalism etc. Of course, that assumes they log in and find the notice, but we can but try.
  • I'm not being optimistic about the # of people will be reviewing the Approved page, because I think there will be few such people -- people like TRM and Fram, our resident eagle-eyes (with eagle beaks and claws, of course). Right now serious post-tick QA doesn't start until the hook is in a prep set, by which time it's already a hassle to pull it back; this new Approved page, in addition to making prep-building easier and solving (we hope) the technical transclusion-limit problem, provides the perfect place for that final QA to take place. I think we'll find that most of the attention now directed at Talk:DYK (which should really be a place for policy and process discussions, not individual hooks) will switch to the Approved page. Obviously for any of this to work the nom needs to stay open until final swap-out.
EEng 06:14, 21 December 2016 (UTC)
  • @BlueMoonset: EEng suggested writing a comment, but I think a template like they use for categories at AFD might be useful. It could keep track of the whole history, including who promoted it to prep and who removed it from a queue and when. Something like:
Wugapodes promoted this to prep 3 ~~~~
So it's obvious but not intrusive. I think the suggestion is a good one because I agree that discussions of a nom, even after being promoted, should take place there just so the history is easier to see. Wugapodes [thɔk] [ˈkan.ˌʧɻɪbz] 06:54, 21 December 2016 (UTC)
(All hail!, indeed!) @Yoninah and Cwmhiraeth: The two of you do the major lifting in promoting to prep, so your input here would be good feedback. — Maile (talk) 13:55, 21 December 2016 (UTC)
When hooks are reviewed they are often clunky, ungrammatical or otherwise not ideal. As a promoter, I consider the wording of the hook and if I think only trivial alterations are needed, I will move it to prep verbatim, and make alterations when it is in the prep set. Afterwards, others may also think it needs rephrasing, so what appears on the main page may be far removed from that on the template. I think it would be useful for this history to be available from the nomination template. Cwmhiraeth (talk) 14:27, 21 December 2016 (UTC)

──────────────────────────────────────────────────────────────────────────────────────────────────── make alterations when it is in the prep set -- We've all done this, and we've all gotten in hot water at one time or another for it; it's the wrong place to be doing that. I submit that if you think a hook on the Approved page needs tinkering with, you should post something to the nom like, "Isn't there a grammar problem with ALT1? I think this ALT1a corrects it...", then propose your ALT1a and move on to find a hook elsewhere, giving time for those watching the nom page to evaluate your suggestion. In other words, either take the hook as approved, or suggest a change, but don't edit it on the fly or in prep -- the people who know the article, the topic, and the nom's history best aren't watching there. Because this is all happening under in the "Approved page fishbowl", our sharpest eyes will now be on these final adjustments. EEng 16:14, 21 December 2016 (UTC)

It would be nice to be distinguish the noms on the approved page by their status: approved, challenged, in prep, in queue, on Main Page. This could be either by having separate sections for each, or having an index or some other visual difference between them. Antony–22 (talkcontribs) 16:45, 21 December 2016 (UTC)
Whether a nom is approved-unchallenged or approved-challenged or gone-to-prep will be apparent by the ticks and comments at the end of the nom. (Perhaps we could have a "gone to prep" tick, but again I emphasize that it should be specified exactly which hook went, w/ or w/o image, which prep set, who moved it, etc.) ("Gone-to-prep" includes three sub-stages, really: in Prep, moved to Q, on main page; and swapped-off-off-main page closes the nom and removes it from Approved.) Weparate sections would make it a little easier on prep-builders, but at the cost of a lot of complexity, and we can always add that later if experience suggests it would be worth it. EEng 17:13, 21 December 2016 (UTC)
  • Question @BlueMoonset, EEng, Wugapodes, Cwmhiraeth, and Antony-22: Depending ... if the hooks are left on the new "approved" page after promotion, then I think we need to do a little rewriting on the how-to of promoting to prep, etc. Specifically this part:
In the DYK nomination template
1) Replace the line {{DYKsubpage with {{subst:DYKsubpage
2) Replace |passed= with |passed=yes
3) Check in Preview mode - if it was done correctly, everything will be against a pale blue background. There should be no stray characters (like }} ) at the top or bottom.
4) Edit summary should indicate which prep area you are moving the hook to.
5) Save
So, will a bot actually close out the template once a hook has retired successfully from its main page appearance? — Maile (talk) 23:06, 21 December 2016 (UTC)
I'm not sure what you're saying. My proposal is that the promoter do step (4) when taking a hook to put in prep. (As I said before, maybe we'll invent a new tick/template for "gone to prep".) Then, as the hook comes off the main page, the bot does the other stuff. EEng 23:12, 21 December 2016 (UTC)
Step 1 and 2 right now actually close out the nomination template, and that's done manually by the promoter. So, you can't have a step 4 without 1, 2, 3. But...yes...that was the question. The template will stay open until the hook has had a successful run on the main page. Then a bot closes the template. Yes? — Maile (talk) 23:21, 21 December 2016 (UTC)
Um, ok... I guess. I still don't see what the confusion was, but I think we're somehow saying the same thing. EEng 23:56, 21 December 2016 (UTC)
Maile, at the point that this goes live, we're going to have to have already prepared numerous revisions to the various instructions, not merely to that one section, assuming all these changes are made to the process. We'll probably also want to put warnings on the Prep pages that there are changes and promoters should familiarize themselves with these before building sets. You never know when someone who's been away for a few months will come back and use the process they know, unaware of the changes since they left. BlueMoonset (talk) 00:57, 22 December 2016 (UTC)
I guess that was my main concern, that promoters wouldn't know the new routine. And I agree that the new process should be posted in numerous places. Even if promoters are currently active, and checking this talk page ... if they haven't participated in this wall of text on how it will be changed, they might be completely unaware. — Maile (talk) 01:18, 22 December 2016 (UTC)

──────────────────────────────────────────────────────────────────────────────────────────────────── You leave for a couple hours and suddenly there's a bunch of new stuff.

  • @EEng: I'm not too concerned about edits that happen in prep. Like the guide says, we can't stop and ask the nom about every change, plus it's CC licensed so might as well take advantage of that. If it's a big change, send it back (or don't promote), but for small grammatical errors and rewordings that don't change the meaning, I think the way it's done is fine.
  • @Antony-22: I actually like that idea. I think having, minimally, an "approved", "promoted", and "pulled" section would be useful for promoters and reviewers. I'll think about how to add something like that in.
  • @Maile66 and BlueMoonset: I'm thinking about how to actually implement a lot of this and I think a lot of changes can be done on the back end of templates and bots to make procedure changes minimal. Essentially all that happens when the DYKsubpage template is subst'd is that it includes the archival template. If we just edit that template to not do that any more, we can keep it open but still keep the actual procedure similar.

If you need me, I'll be in template space trying to whip up some examples/proposals. Wugapodes [thɔk] [ˈkan.ˌʧɻɪbz] 03:07, 22 December 2016 (UTC)

You know, actually, maybe the changes shouldn't be minimal? Perhaps this is a good time to actually streamline the process, simplify, and make it easier for editors. Wugapodes [thɔk] [ˈkan.ˌʧɻɪbz] 03:20, 22 December 2016 (UTC)
I'm not too concerned about edits that happen in prep -- Unfortunately experience shows that judgment, by persons not previously involved in a particular nom, about what constitutes a big change vs. a small rewording has been (to put it charitably) unreliable. It makes little sense for a detailed review and discussion process to be followed by a silent tinkering visible only in the edit history of the Prep template, made by someone who's been looking at the nom and the article for 90 seconds. And yes, we can ask the nominator and reviewer about every little change (in the sense of posting a suggested change to the nom change and waiting for comment) -- part of the reason it's healthy to have, to the extent possible at any given time, a large pool of approved hooks awaiting promotion is exactly so there's no hurry to promote any given hook, and thus comment can be invited on even apparently minor changes. This is the main page, after all, and there's many a slip twixt the cup and the lip. The guide says what it says currently because the current process forced us to make many such on-the-fly changes, and we were just closing our eyes and hoping for the best. One salutary outcome of the current effort should be to put an end to that by keeping the nominator and reviewer in the loop until the very end. EEng 04:03, 22 December 2016 (UTC)
  • @EEng: Right now it takes some searching to figure out which puck is the most recent one; it's a minor annoyance right now when I'm scanning through the noms to find a hook to review. Even having a prominent banner at the top of each nom would work, if people don't want them divided into sections, though if a bot's controlling the page sections wouldn't be hard to maintain. Antony–22 (talkcontribs) 03:28, 22 December 2016 (UTC)
Well, if we rechristen the |passed= parameter as |status=, then it could take on values like some_hook_approved, hook_to_prep, appeared (i.e. has finished its main-page appearance, though maybe this is just closed), and these could manifest as the banners you're envisioning. Back on the giant concatenated nominations page, it could be the change to some_hook_approved that signals the bot to move the nom to the Approved page, instead of the mysterious scan-for-bottom-tick system used now. (We can still have the ticks, for humans.)
I really think having a bunch of sections is gilding the lily. Just add newly approved noms at the end, and prep builders should look for hooks, in general, near the top. Plus the special-occasion hold areas, of course. EEng 04:03, 22 December 2016 (UTC)
Oh, wait, I misread Antony's post as talking about promoting hooks, when he was actually talking about finding noms to review. Since the reviewed noms (|status=some_hook_approved) will move to the Approved page, what is now the Nominations page will have only noms that haven't been approved yet, so finding a "virgin" should be easy. EEng 04:20, 22 December 2016 (UTC)
I was in fact talking about the new approved hooks page, but I was making an analogy with finding a hook to review. Just like people scan the noms page right now to find one to review, people may also want to scan the approved hooks to find one to promote or double-check, so it's important to be able to tell its status at a glance. Either displaying a prominent banner or having the bot keep them in separate sections would work for this. Antony–22 (talkcontribs) 06:14, 22 December 2016 (UTC)

New template[edit]

For those interested in the idea I floated about an afd-like template, I made one that handles promotion and pulling. You'll want to see {{DYK moved}}:

{{subst:DYK moved|alt=1|toPrep=3}}

ALT 1 promoted to prep 3 – Wugapodes [thɔk] [ˈkan.ˌʧɻɪbz] 04:51, 22 December 2016 (UTC)

{{subst:DYK moved|fromPrep=3}}

the nomination was pulled from prep 3 – Wugapodes [thɔk] [ˈkan.ˌʧɻɪbz] 04:51, 22 December 2016 (UTC)

It has pretty intuative syntax (looking only add human-readable parts, "DYK moved alt 1 toPrep 3" and "DYK moved fromPrep 3"), it pulls left so it stands out and starts a new conversational block, it produces a standard output that is easy for bots to look for and parse. It could also be incorporated into the current {{DYKsubpage}} template in addition to the "status" parameter EEng mentioned above to automate the process perhaps. That will be my next goal. Wugapodes [thɔk] [ˈkan.ˌʧɻɪbz] 04:51, 22 December 2016 (UTC)

Please pace yourself. We need you for the long haul. The benefits of all this could be far-reaching. EEng 04:57, 22 December 2016 (UTC)
Don't worry, it's all been rather simple. Believe it or not it's a nice break from my other work...not sure what that says about me. I tend to follow an iterative design model: prototype, seek feedback, and then scrap or adjust. I think it's easier for people to discuss changes when there's something to work off of which is more what I'm trying to provide than finished products, perhaps that wasn't clear. Wugapodes [thɔk] [ˈkan.ˌʧɻɪbz] 05:31, 22 December 2016 (UTC)
@Wugapodes: Some comments/questions
(1) The blanks would be filled out by humans, for the bot to sense and do its thing. Right?
(2) If "alt=" is left blank, does it assume it's the non-alt hook?
— Maile (talk) 17:43, 22 December 2016 (UTC)
I think it would be better if the value was always explicit i.e. use 0 for "ALT0" i.e. the original hook. EEng 18:20, 22 December 2016 (UTC)
(1) That was my thought, yes. If you look at {{DYKsubpage/sandbox}} and Template:DYKsubpage/testcases you can see a mock-up of how it would work. Instead of subst'ing the dyksubpage template like previously, the promoter would fill in alt= and prep= which would automagically add this comment line. When it leaves the mainpage, a bot would subst the template closing it.
(2) For promotions it assumes alt 0, though that can also be explicit or treated as an error to leave it blank, for pulled hooks it just says what prep it was pulled from (if this gains traction, we'd want to add pulled from queue and mainpage also) but the specific hook that was pulled can be specified as well. Wugapodes [thɔk] [ˈkan.ˌʧɻɪbz] 21:51, 22 December 2016 (UTC)
I agree with User:BlueMoonset that we should take this in stages. As I've said previously, modifying DYKUpdateBot to close out nominations when taking sets off the Main Page isn't on the table right now. Let's focus on addressing the immediate issue of the nominations page, and then we'll have plenty of time to fine-tune other parts of the process. Shubinator (talk) 19:25, 23 December 2016 (UTC)
But it is on the table. We're discussing it now. Do you have any other reasons for not discussing it other than that you say we're not discussing it? EEng 22:02, 23 December 2016 (UTC)
EEng, what is being discussed here is going to require extensive modifications to DYKUpdateBot. Shubinator, the bot owner, has effectively said that this isn't something he can accomplish quickly: it's going to require discussion, specifications, and time to write and test, not to mention the necessity of getting the existing bot and Wugapode's proposed new bot not to get in each other's way. Right now, we have an immediate problem that needs solving: the nominations page is too big and dozens of nominations are not being transcluded, making it extremely difficult for people to work on them. To separate them, another of Shubinator's bots, DYKHousekeepingBot, needs to be modified to deal with two separate pages rather than one: a new Approvals page on top of the current Nominations page. Once that's done, we can safely separate the pages, the transclusion problem goes away, and the design of the new process can take center stage. It's a matter of priorities and time available to work them. BlueMoonset (talk) 00:13, 24 December 2016 (UTC)
I don't care if the implementation is done in phases, but the design (rough, at least) should all be up front, especially if, as you say, it will require changes here there and everywhere, to avoid re-redoing stuff over and over. EEng 04:33, 24 December 2016 (UTC)

New page setup[edit]

"Conscience does make cowards of us all"

The native hue of resolution
Is sicklied o'er with the pale cast of thought;
And enterprises of great pitch and moment,
With this regard, their currents turn awry,
And lose the name of action.
Hamlet III.1

[FBDB]Can the poetry, you ruttish dismal-dreaming joithead! EEng 19:03, 24 December 2016 (UTC) Click here to refresh the Shakespearean Insult Generator

@Shubinator, BlueMoonset, EEng, Maile66, and Antony-22: It seems like we've all had wonderfully productive discussion, but what I think we need next is to decide what we are actually going to do and when. I think we all agree the first thing that needs to happen to get the nominations back in order is splitting the pages. I can do that literally whenever. What seems to be the biggest holds on actually doing it are (in order of my perceived importance):

  • How such a change to the nomination page would affect with Shubinator's bots
  • How a bot moving noms from one page to the other would interact with Shubinator's
  • How the approved hooks page would look/function
  • What changes to the DYK process would need to be made to document these changes

Given this, I have a few questions. The first is for Shubinator: what do you need from me to most efficiently modify the DYK bots so that they can work with the most minimal change of splitting the pages? The rest are general things to discuss, in order of imminent necessity:

  1. How should the approved hook page be organized: like in the example (retain date sections), like EEng suggests (just add them to the bottom as we find them), like Antony-22 suggests (approved section, contested section, pulled section), or some other idea?
  2. Are the more superfluous suggestions made so far worth discussion after the immediate problem is solved?
  3. If so, how and where should that discussion take place? Like we have been already in this section, in a new section on this talk page, on a subpage to craft a proposal RFC, or some other option?

Hopefully we can get the immediate problem solved asap, while also improving the project. The discussion has been great so far, and I hope for more, but let's not "lose the name of action". Wugapodes [thɔk] [ˈkan.ˌʧɻɪbz] 01:06, 24 December 2016 (UTC)

Wugapodes
  • In some ways, I wouldn't mind if this went to a separate page RFC. But what usually happens then, is out-of-sight-out-of-mind, and participation drops. On the other hand, this particular thread has been open since November 1, and it's down to the few of us to figure out the mechanics.
  • I think it is probably visually essential to retain the date sections, so that we can focus on getting the older nominations promoted. Otherwise, I think older nominations would fall through the cracks. IMO, when BlueMoonset started the regular updated sections on older nominations needing DYK reviewers, it was an improvement in bringing eyes to nominations that had been forgotten. — Maile (talk) 01:32, 24 December 2016 (UTC)
  • Wugapodes, I didn't see that there would be any significant changes to the Nominations page (Template talk:Did you know) itself insofar as the existing bots were concerned. There would be fewer hooks transcluded there with a bunch having been moved to another page, but varying numbers of transclusions are normal. (I suppose it might depend on whether the page changes while Shubinator's bot is reading it, but wouldn't that cause a collision error now if it could have happened?) There will inevitably be some textual changes to the instructions when the page splits, and then as the other proposed changes come on line, but it seems to me that the documentation modifications will take far less time than the design, specifications, coding, and testing. Please see below about the page split. BlueMoonset (talk) 02:46, 24 December 2016 (UTC)

I had posted to Shubinator's talk page a while ago, while this conversation had been moribund for a while, asking what he would need in order to revise DYKHousekeepingBot to check through two pages, of nominations and of approved nominations, and still be able to build the List of DYK Hooks by Date table that appears on the Queues and Nominations pages and lets us know how many hooks are out there and how many of them are approved. Losing this functionality (and knowledge of how many are approved) by separating the pages—the bot wouldn't see any on the Approved page—seemed to me to be a very bad idea. Per our discussion there, I've just created the bones of an Approved page, and seeded it with four hooks from December 6 plus the Special occasion hooks for next year.

My assumption was initially that the Approved section would have subsections by dates, but when I created a page today for Shubinator to test with—Template talk:Did you know/Approved (which I figured we could then populate for real once the bot was working)—I noted that the discussion seemed to be veering away from dates, so it might be best to just combined everything on the Approved page into one line on the List of DYK Hooks page. (I didn't populate any dates, just the main section as a whole.) Since Maile seems now to be heading back toward dates, maybe Shubinator should allow for date and non-date headers in the bot code revision, combining the dates from both pages, and adding lines for each additional category on the Approved page (but only one line for the entire Special occasions section). BlueMoonset (talk) 02:46, 24 December 2016 (UTC)

I noted that the discussion seemed to be veering away from dates to continue quoting Hamlet, "ay, there's the rub". I honestly have no feelings about how the approved page should be set up as all (except Antony-22's) would be trivial to implement (the sections-by-status would require some changes to the process to be viable, so perhaps we should stick a pin in that one). So trivial, in fact, I'll program it to do both and when we come to a decision on that, have it output the one we agree on. I'll get to work on that, incorporate the structure you have at Template talk:Did you know/Approved already, and should have something ready soon. Thanks for your response, it really helped clear things up. Wugapodes [thɔk] [ˈkan.ˌʧɻɪbz] 04:23, 24 December 2016 (UTC)
So, perhaps a stupid question, but how will the special holding areas work? Will they be nominated on the nom page and then moved to a section on approved when approved? If so, is there some consistent formatting that the bot can look for to know to move it to a holding queue? If not, then this may be a minor snag. Two solutions could be to have humans do it (for now or forever), or to modify the way dates are requested to make it bot-readable. Wugapodes [thɔk] [ˈkan.ˌʧɻɪbz] 04:53, 24 December 2016 (UTC)
After the bot moves it to the normal part of the Approved page, someone can then pick it up manually and move to the right special holding area; I really don't like locking too much structure into bot code. Since we're adding all the bells and whistles it might be nice to have a template parm |special_occassion= |special_occassion_requested=; it needn't be more than yes/no, and if it's yes that raises one of those famous colored banners to warn everyone it's not on the normal assembly line. EEng 06:26, 24 December 2016 (UTC)
Special occasion should be a manual move: for one thing, it isn't an automatic grant. A human reviewer needs to take a look at it, and if they agree that this is indeed sufficiently special, and meets the criteria (it shouldn't be more than six weeks in the future, for example), then they can move it by hand to the Special occasion section and set up a new date section there if necessary. I would imagine that some of the time the reviewer will approve and move even before the bot takes action. BlueMoonset (talk) 06:58, 24 December 2016 (UTC)
The date separations is a valueless complication for no purpose. If newly-approved stuff is added at the bottom, and prep builders work from the top, it will tend to be FIFO, which is good enough. EEng 04:30, 24 December 2016 (UTC)
Re dates, after having just read BlueMoonset post beneath mine, it occurred to me that we need Shubinaor's feedback on what might be the best avenue to take on that. I'll go with the majority opinion on this, as long as it is workable for Shubinator. — Maile (talk) 12:15, 24 December 2016 (UTC)
DYKHousekeepingBot can go either way on date sections. Right now it supports date sections, if we remove them the bot will need minor tweaking, definitely doable. Independent of the bots, when I was building DYK sets, the date sections were useful in reducing edit conflicts. Shubinator (talk) 23:43, 24 December 2016 (UTC)
Unless I'm misunderstanding something only the bot will be editing the Accepted page anyway (adding noms that have just transitioned to some_hook_approved, and removing them at the end of their life) -- the exception being the rare time a hook is moved from the main Approved list to a special-occasion area. So I don't understand the edit-conflict argument.
One of the things I find really annoying about the date sections we now have on the giant nominations page is that new noms are popping up here and there all the time within the current 7-day window, so there's nowhere to watch to just see new noms as they arrive. If the date sections are just "date moved to Approved page" that's fine, but please don't arrange them by date nominated, because that means new stuff will always be appearing all over the Approved page, and those doing QA will have no way to find new additions systematically. EEng 02:59, 25 December 2016 (UTC)

I apologise in advance if this is a really bad idea, but what about a page that was a sortable table with a link to the nomination page, status, nomination date, etc. No transclusions, the bot could just add to the list, update entries, and remove items when they come off the main page. Length would be a non-issue. Wouldn't that help us a lot with working through the nominations in varying stages? An editor could sort by status if looking for pulled / needing review, or by date, or even by nominator / approver / promoter, if that was somehow useful. Include if there is an approved image, so we could look for older approved hooks with images, etc. EdChem (talk) 08:59, 24 December 2016 (UTC)

If, as I think you're suggesting, this would be an auxiliary structure that summarizes and indexes the content of the other page(s), then that really is something we can add later without impact on the design we're developing. EEng 09:07, 24 December 2016 (UTC)
EdChem I'm not sure if you're suggesting the table as the separate page we're talking about, or converting the nominations page to wiki links instead of transclusions. For one thing, wiki links also add to a page's size and bog it down. But the big issue is the visuals. Promoters should be able to scroll a page to eyeball potential hooks to complete a set. If we make it a system where they have to click on each link to see what it is, we discourage the incentive to promote hooks. Did I misunderstand what you are suggesting? — Maile (talk) 12:41, 24 December 2016 (UTC)
I'm still concerned that when someone's scanning down the Approved page looking for a hook to promote, it's more difficult if the already-promoted hooks are mixed in. I agree that they should stay in date order within the sections, just without subsection headings. I know having the bot juggle noms between sections is a bit more work, but it also makes things a bit easier for promoters down the line. Antony–22 (talkcontribs) 18:48, 24 December 2016 (UTC)
The colored banners you proposed should make things easy enough. Maintaining the structure you're proposing will require constant bot intervention. It's technical rococo. Simple is better. EEng 19:12, 24 December 2016 (UTC)
@EEng and Maile66: I didn't really have a concrete proposal, just an idea that was a thought-bubble. On reflection:
  1. I would keep the nom page with the transclusions for the nominations yet to have an icon added (ticked, needs work, no, etc.).
  2. I was originally thinking of the issue of which order is best, which a sortable table would allow each editor to tailor to their own needs. However, Maile's point on transclusions for the in progress / pulled / etc page is well made, so I guess what I am suggesting would be an adjunct.
  3. My table page suggestion could summarise both pages and allow anyone to see every current nomnination in one place, by oldest, or status, or whatever else in the table. Statuses could be something like "nomination" (for on the current page), "new reviewer needed", "pulled", "GTG", "GTG (AGF)", "Waiting for action" (for when the nominator has been asked with a ? or / icon, or has acted and waiting for further input from the reviewer), "Promoted" or "In prep / queue". Having a table with no transclusions but only links means we don't run into the issue we presently have with too many templates.
  4. Obviously any decision is contingent on consensus on the way forward, refinements / alterations / rejections / etc of suggestions made, and agreement from the bot operators on practicality and reasonableness, etc.
EdChem (talk) 22:10, 24 December 2016 (UTC)
@EdChem: have you seen Wikipedia:Good article nominations/Report? It seems rather analogous to what you're suggesting in that it augments the GA process by summarizing the nominations and directing attention to the older ones in each category (and overall). Wugapodes [thɔk] [ˈkan.ˌʧɻɪbz] 23:22, 24 December 2016 (UTC)
Wugapodes, I see that is one option. With nominations spread across two or more pages, a single page summary would (I think) be useful, and it needs to leave out the transclusions to be viable. I'm just throwing in an idea that seemed to me to address the concern over ordering and to be useful more generally, without advocating any single approach as the way forward. EdChem (talk) 02:53, 27 December 2016 (UTC)

Wugapodes, DYKUpdateBot doesn't read the nominations page or the prep areas, so we're good on that front. DYKHousekeepingBot will need to be modified to 1) reflect the noms on the Approved page in the table generated at Wikipedia:Did you know/DYK hook count, and 2) not count nominations transcluded on the Approved page as orphaned nominations. DYKHousekeepingBot only reads the noms pages, so it shouldn't conflict with any nomination shuffling. #2 is a one-line change. As BlueMoonset mentioned, we've been iterating on #1, take a look at the modified bot's output here: User:Shubinator/Sandbox/DYK hook count. Feel free to provide feedback! Shubinator (talk) 23:52, 24 December 2016 (UTC)

DYKMoverBot Prototype[edit]

Happy holidays everyone, I have completed a prototype of the nomination mover script. It is currently set up to out put two styles, with date sections and without date sections. I have a few more things to work out before I can submit a BRFA, but the formatting for each page is set enough for feedback. Take a look, decide which you like best, make that preference known to me somehow. A possible option is to quasi-A/B test it. Use one for a few days, use the other for a few days and determine which was best for your workflow. Wugapodes [thɔk] [ˈkan.ˌʧɻɪbz] 02:51, 27 December 2016 (UTC)

I have filed a Bot Request for Approval. Feel free to comment there. Wugapodes [thɔk] [ˈkan.ˌʧɻɪbz] 23:20, 28 December 2016 (UTC)

Time to move to two sets a day?[edit]

As BlueMoonset mentions above, there are around 150 approved hooks at the moment, and with new nominations being made at the rate of more than ten per day, this number is set to increase. Coupled with this, the nominations page is insufficiently large to transclude all the nominations, so all nominations made after January 1st are currently not seen in full on the page. This is not ideal!

At the moment we are on a 24 hour cycle, with a single set of eight hooks appearing on the main page each day. I propose that we increase the throughput, to a 12 hour cycle with seven hooks in each set. This would mean that 14 hooks reach the front page each day and we may then keep rate with the addition of new nominations or even reduce the backlog slightly. Cwmhiraeth (talk) 09:40, 8 January 2017 (UTC)

A bit concerned that this will mean extra work to check each one of them. How many hooks are pulled nowadays for an issue? Jo-Jo Eumerus (talk, contributions) 09:48, 8 January 2017 (UTC)
Things have improved marginally, mainly because each and every single approved hook is being re-checked before it hits the main page. The Rambling Man (talk) 10:05, 8 January 2017 (UTC)
I certainly cast my eye over the hooks and the articles before I promote prep sets to the queue. Schwede66 10:15, 8 January 2017 (UTC)
I'd tend to agree that we have a problem, but we moved to a 24 hour cycle recently, essentially because the project could not handle the throughput. I'd suggested an 18 hour cycle at some point, but apparently that's technically difficult...I think 12 hours might be worth giving a shot, especially if we reduced the size of each set to 6 or 7. Alternatively we could expand set size to 9, or ten, and keep it at a 24 hour cycle...Vanamonde (talk) 10:34, 8 January 2017 (UTC)
I completely agree with the suggestion for a 12 hour cycle with 7 hooks per run, even if only temporarily. We badly need to get through the backlog of approved hooks. Prioryman (talk) 10:51, 8 January 2017 (UTC)

Looks like it's good to go. 2 sets × 7 hooks per day it is. Of course, if the error rate creeps back to its old level, we'll need to re-think it. The Rambling Man (talk) 10:55, 8 January 2017 (UTC)

I'd agree. Can we be flexible, and only do two sets some days, as the throughput allows? Anyway thanks to all who keep the railway running. Johnbod (talk) 15:18, 8 January 2017 (UTC)
Since Queue 2, currently set to hit the main page on January 10 assuming one set per day, has two hooks specifically requested to run on January 10, can we please not make the switchover until that set has had a chance to run? The soonest we could change would be starting with Queue 1 (which will be promoted at 00:00 January 9), and we'd have to move the January 10 hooks if we changed then. Since those sets are already made and have eight hooks in them, let's save the change until Queue 3 and January 11. Sound good? Note to Johnbod: that kind of flexibility requires active and knowledgeable admins who know not only how to make the changes, but are keeping track of special occasion hooks so they don't run at the wrong time. It's generally better if frequency changes are done as infrequently as possible. BlueMoonset (talk) 16:13, 8 January 2017 (UTC)
Could another admin please show me where and how to change the frequency of uploads to the homepage? From my perspective, we can do that change now and if hooks and credits need to move to a different queue, that's easily done. Schwede66 18:39, 8 January 2017 (UTC)
Shubinator has to tell his bot to change its schedule, I think. — Maile (talk) 20:19, 8 January 2017 (UTC)
No, it isn't up to Shubinator; any admin can do it. BlueMoonset (talk) 22:02, 8 January 2017 (UTC)
Could Shubinator (or whoever admin will want to change it) show me how that is done, please? Schwede66 06:17, 9 January 2017 (UTC)
Schwede66 I was correct, in that it is the bot. Took some searching around to find this. Time Between Updates is protected so only admins can do this. But if you open the edit window on that, you'll find the secret instructions. — Maile (talk) 14:16, 9 January 2017 (UTC)
Schwede66, one thing I've never been sure about is what happens if you decrease the value on the User:DYKUpdateBot/Time Between Updates page, which the bot uses in conjunction with value of Template:Did you know/Next update/Time (which, counterintuitively, is the time of the most recent promotion to the main page) to determine when the next promotion is. If you're increasing the time between updates value, the next update will take place that number of seconds after the most recent one. If you're decreasing it, I'm not sure whether it will also update that number of seconds later (which could already be in the past, and could mean an immediate promotion). The safest thing I can suggest, absent a definitive statement from Shubinator, who wrote the bot, is to wait until after the DYKUpdateBot has completed its 00:00 January 10 run a few hours from now and promoted Queue 2 to the main page (it takes a few minutes for the job to finish), and only then reduce the Time Between Updates from 86400 to 43200 seconds, to begin the every 12 hours promotions. The subsequent promotion would thus be 12:00 January 10. BlueMoonset (talk) 17:46, 9 January 2017 (UTC)
I think that has worked smoothly. Have changed the update frequency and then purged the cache on the Queue page, and it now shows the 12-hour update cycle. Thanks, Maile66. Schwede66 00:18, 10 January 2017 (UTC)
For the record, your interpretation is correct - when the bot wakes up, it takes Next update/Time and adds Time Between Updates to figure out when the next update should be. If that's in the past, the bot will immediately update. Shubinator (talk) 05:26, 12 January 2017 (UTC)

I built Prep 5 as a 7-hook set. Hope that's okay. Yoninah (talk) 23:26, 8 January 2017 (UTC)

Is there some reason you want it to be 7 hooks instead of 8? — Maile (talk) 00:04, 9 January 2017 (UTC)
Maile, the above discussion seemed to converge on two sets of 7 hooks per day, so at some point we ought to make the switchover from 8 hooks to 7 hooks. It makes sense to me to leave the completed sets at 8, and start with 7 at some point, which Yoninah has just done with Prep 5. BlueMoonset (talk) 02:39, 9 January 2017 (UTC)
OK. I didn't read the whole wall up there. — Maile (talk) 02:43, 9 January 2017 (UTC)
There's an additional aspect to this that hasn't been mentioned above. Getting an article through the DYK process with the QPQ requirement and the extra referencing required is a significant amount of work. The 24 hour sets double the exposure the article gets and the "reward" for that work. Aren't there regular complaints that hooks are dull and articles of low quality? How about culling more from the queue? (Not mine I hope!) Philafrenzy (talk) 02:17, 9 January 2017 (UTC)
It doesn't seem to matter how many times complaints about dullness are made, they are usually refuted/ignored. Article quality is not an issue for DYK, once again calls to improve this seldom make any difference. Realistically, all we can do is to ensure that actual errors aren't sent to the main page. The Rambling Man (talk) 07:37, 9 January 2017 (UTC)
Also, remember that DYK? frequency varies over time. A few years ago it was three sets of hooks a day and noone complained about getting too little "reward". Regards SoWhy 08:30, 9 January 2017 (UTC)

But not forever[edit]

Please, the moment the approved reserve (hooks in Q + hooks in Prep + noms approved but not moved to prep) drops below 50, go back to one set (of 7 or 8) per day. We don't want to be rush-approving hooks like we used to in the old days. EEng 02:27, 9 January 2017 (UTC)

I would also like to note that moving hooks to change twice (or more) per day really lowers the incentive to create hooks, and possibly to improve new articles at all. The last DYK that I submitted when it was up for a full 24 hours got over 11,000 views that day, and when my hook was up earlier this week for 12 hours, it only got 1,500 views. I know this could be due to a variety of factors, but I don't know that that few views is really worth it to go through all the DYK hassle. Just a thought to consider, although I understand why the 12 hours thing is happening. Kmwebber (talk) 22:26, 12 January 2017 (UTC)Kmwebber

Lead too short[edit]

I am not an advocate of tagging articles by adding MOS-type tags such as "Lead too short". Adding such tags to DYK-nominated articles that have already been promoted to prep seems particularly undesirable; the article might proceed to the main page without the tag being actioned. The tag disfigures the article, and the DYK rules and supplementary rules include no requirement for these newly created or expanded articles to conform to MOS guidelines. What do others think of this practice? Cwmhiraeth (talk) 10:12, 10 January 2017 (UTC)

One of the aims of DYK is too attract (new) editors to new articles. Identifying problems and issues with such articles, so people know more easily what may need to be done, seems to be a good thing for DYK, not a bad thing. The fifth aim of DYK is "To encourage readers to edit articles that appear on DYK or start their own, thus facilitating the recruitment of new editors." Fram (talk) 10:42, 10 January 2017 (UTC)
This is covered by the supplementary rules under D6: "The article is likely to be rejected for ... the presence of dispute tags." One could argue that this is also covered by D7 (emphasis added by me): "There is a reasonable expectation that an article—even a short one—that is to appear on the front page should appear to be complete and not some sort of work in progress." If the tagging happens after promotion to prep but before it goes into the queue, I for one will ping nominators and the reviewer to deal with it. I've done this twice in the last 24 hours if I remember rightly (here and here). If the tag were to remain, I wouldn't promote the hook to queue, or remove it from the queue before it goes live. Maile66, how do you handle these things? Schwede66 11:20, 10 January 2017 (UTC)
Schwede66 "Lead too short" is not listed as a WP:DISPUTETAG template. — Maile (talk) 13:11, 10 January 2017 (UTC)
I wouldn't describe "Lead too short" as a dispute tag. For articles to actually appear on the main page carrying tags is deprecated. There have been occasions when the queue is empty, and a prep set gets moved to the queue shortly before the deadline without the kind of detailed inspection you are referring to. Cwmhiraeth (talk) 11:27, 10 January 2017 (UTC)
Sorry, got that wrong. Had another look and you are right. This leaves us with rule D7. Schwede66 18:11, 10 January 2017 (UTC)
Schwede66 That would seem to be in conflict with WP:DYKNOT. You asked that I would do. As an admin, you have to chart your own course on things like this. The best advice I can give you, is to try and take the high road, according to your own instincts. Hopefully, you take into consideration what it feels like to be the nominator. You will be pulled in opposite directions. No matter what you do, it's bitch-moan-blame from one direction or another. There are some who will say anything to get a reaction. Being a promoter to prep and being an admin can be a fast burn-out, because of this. I can think of one admin who did one promotion to queue and never another one. I can think of one fairly recent prep promoter who quit after a short time precisely because of the blame-laying that goes on with it. People quickly get tired of being the punching bag and move elsewhere. I refer you to the posts on this page Christmas Day. So make up your own mind how you want to handle things. And good luck. — Maile (talk) 19:15, 10 January 2017 (UTC)
Thanks for the heads up. I guess with 200+ DYK credits, I certainly know how these things can be like for a nominator. Schwede66 19:22, 10 January 2017 (UTC)
@Fram: how would leaving MOS tags attract new editors to articles featured in the DYK section? If anything a tag would only deter a new editor from editing the article as they wouldn't understand MOS terminology and would probably leave it to a more experienced editor to deal with. DYKs are meant to set a standard for quality, leaving tags in articles when they've been promoted wouldn't set a good example for anyone. ("quality", heh) JAGUAR  11:42, 10 January 2017 (UTC)
DYKs are meant to set a standard for quality – are you joking? EEng 13:16, 10 January 2017 (UTC)
Yes I was, did you not see the "("quality", heh)" in small print at the end? JAGUAR  13:21, 10 January 2017 (UTC)
Oops, no I didn't. EEng 13:51, 10 January 2017 (UTC)
The main point I am making is that if an article is going to be tagged with "Lead too short", this should be done on the nominations page so that the matter can be resolved before the article is promoted. Cwmhiraeth (talk) 12:01, 10 January 2017 (UTC)
I agree with that. But fact is, these things will be overlooked by reviewers from time to time. In one of the examples I gave, it was overlooked during the GA review. If it slips through the net there, there is little hope to expect that all DYK reviewers will pick up on it. Schwede66 12:11, 10 January 2017 (UTC)
Everyone can edit articles whenever they want, and a promotion doesn't immunize an article from getting tagged (whether this is "lead too short" or adding "dubious" or "citation needed" to specific statements, or even nominating it for AfD if this is done in good faith). You may try to get consensus at AN that DYK pages should get fully protected between promotion and disappearance from the main page, but I highly doubt you'll get agreement for such a change. Many people never look at the nominations, but may see an article linked from the main page, go to read it, and notice problems with it: are you claiming that they can't tag these problems at that time? Let's say that someone now goes to Moon Duchin and adds a section "Publications" with an "expand section" tag, would that be disruptive, problematic, whatever? No, that would be normal editing and the kind of improvement we hope to see from DYK. (By the way, I note that queue 3 is on the main page but still filled at Template:Did you know/Queue/3). Fram (talk) 12:54, 10 January 2017 (UTC)
I would return articles to the nomination area if, upon inspection before promotion, their leads are inadequate. It's a shame you didn't let me know about this discussion Cwmhiraeth, but I'm not surprised. For you information, I will continue to tag items you promote if they fail to meet any one of a vast number of issues that are continually sent on their way to the main page. It doesn't matter much anyway, because everyone by now knows that DYK is in no way interested in article quality. The Rambling Man (talk) 13:23, 10 January 2017 (UTC)
FYI, I did not choose to personalise the issue. Cwmhiraeth (talk) 13:30, 10 January 2017 (UTC)
Doesn't make much sense not to include the person doing the tagging though, does it? The Rambling Man (talk) 13:41, 10 January 2017 (UTC)

──────────────────────────────────────────────────────────────────────────────────────────────────── If people are promoting articles to sets for the main page which are clearly sub-standard then they should expect them to be tagged as such. Cwmhiraeth makes it clear that her job is simply to promote hooks that have been ticked by the DYK to a queue yet diligently spends around an hour or so per day preparing each set. That's admirable but the problem is that as soon as those sets are prepared, they are reviewed (principally by me) and numerous issues are found. Before anyone sounds the klaxon, yes I know that DYK has no quality threshold at all, and anything goes, but Cwmhiraeth maintains that it's not her job to review these articles and hooks in any detail. Just as an aside, I would estimate my "complaint success" at around 80%, so clearly there's work to be done here still. And we're going to two sets per day so that'll be good!!! The Rambling Man (talk) 19:38, 10 January 2017 (UTC)

Well, while preparing a set, which usually takes me about ninety minutes, my main checks are that the hook facts are accurate and in the article, and are supported by inline references which I check. I then make any adjustments I feel are necessary to the wording of each hook. You then look at the set and make further adjustments to the hook (or ask someone to make them for you) and that's fine, you are doing a good job and it's just how the system should work. It's when you start tagging the article that I take issue with your actions.
A "lead too short" tag discovered by the admin moving the prep set into the queue involves a flurry of extra work, either to expand the lead, or to ping the nominator and ask him or someone else to sort it out, or to return the hook to the nominations page and replace it with another hook. There is nothing theoretically wrong with adding the tag at any time, but it does cause considerable inconvenience if it's done while the article is in prep, and if it is done when it is in the queue it would be even worse. Cwmhiraeth (talk) 20:03, 10 January 2017 (UTC)
Well that's unfortunate, because that's the point where I review the articles and usually find numerous errors, grammar issues, technical issues and that's when they get highlighted. It would be far better to make these checks yourself and boot inadequately prepared articles back to the nomination area. By the way, for the avoidance of doubt, I will continue to review articles and tag them as appropriate, as is my right as a Wikipedia editor. The Rambling Man (talk) 07:49, 11 January 2017 (UTC)
Bandiera della Pace.png
May there be peace! I reckon it would be good to accept that over a large number of articles, not everything that is possibly wrong will be picked up the the initial DYK process. I also reckon that overall, it's better that outstanding issues are identified before articles hit the homepage. Yes, it may be inconvenient, but it doesn't necessarily have to be fixed by the nominator or reviewer; there are lots of editors around who can lend a hand. And if an admin has to pull a hook out of a queue because nobody found the time to attend to it before promotion to the homepage, well that's not really such a big deal either. We are all here as volunteers and if we always remember that, it's easier to concentrate on what everybody can contribute to the project. Show each other as much respect as you would like to receive yourself. Be king in your choice of words. We wouldn't be here if we didn't have similar goals to one another. It's much better to have fun doing this than stress. May there be peace. Schwede66 08:14, 11 January 2017 (UTC)
Indeed, and all I'm interested in is quality on the main page, not some blind adherence to some arcane list of project "requirements", and I have plenty of fun while I'm ensuring such quality. Now I'm not sure of the purpose of this thread, it certainly hasn't achieved anything so I'd suggest it's closed and we can all get on with improving articles. The Rambling Man (talk) 08:20, 11 January 2017 (UTC)

Stubby stubs[edit]

Amy Satterthwaite
Amy Satterthwaite
Bisaldeo temple
Bisaldeo temple

The above item brings me to raise a concern that I have. I fully expect to hear about 'scope creep' in response, and I don't know whether it's been discussed before, but it concerns me nonetheless. Through DYKs on the homepage, we are drawing attention to articles, and sometimes a lot of attention. This is true for the DYKs themselves, which are covered by all sorts of rules, but in addition, we are wikilinking other articles, too. And some of them may get as much attention as the DYK itself, especially when it's the lead hook where an image may relate to a non-DYK hook. What I suggest is that the lead hook should not point to non-DYK articles that are stubby stubs. I'll give you a couple of examples.

The first one is currently in prep 1; the DYK link points to a nice and decent list article, but the hook photo goes with a supporting article that was in a dire state when it first hit the prep. Fortunately, I managed to attract the attention of members of the relevant Wikiproject and then wrote an article section myself, and the article is now looking quite respectable.

The second one is currently in queue 5. The non-DYK link points to an article of two short sentences. I put a note onto the article's talk page and posted links to this to the two relevant Wikiprojects (here and here), but so far to no avail.

What I suggest is that we don't draw attention from the homepage to stubby stubs through the lead hook. If linked articles are short stubs, we should not make that respective DYK the lead hook. Would appreciate your thoughts on the matter. Schwede66 12:08, 10 January 2017 (UTC)

I think you're drastically overestimating the number of people who actually click those links. As an example, this is what happened to Assyrian sculpture when it was a non-bolded link in the lead hook a couple of days ago. I'm sure the world has figured out by now that some Wikipedia articles are poor quality. ‑ Iridescent 12:16, 10 January 2017 (UTC)
"What's short, green and curly and produces films?" - Stubby Broccoli. Martinevans123 (talk) 12:24, 10 January 2017 (UTC)
Iridescent, we may well get similar page views with Bisalpur Dam. But when you stick a photo of a sporty woman on the homepage, the views will be way higher. Shall we have a look in a few days? Schwede66 12:39, 10 January 2017 (UTC)
The temple and cricket articles have now both run and page view stats are available. And they are as I expected. The dam gets far few views than the temple (and the temple scored a good size view; well done!). With cricket, the bold link got significantly fewer views than the auxiliary link. Schwede66 19:13, 13 January 2017 (UTC)
I agree with Schwede66, and applaud him for taking the time to expand Amy Satterthwaite before it hit the main page. Yoninah (talk) 13:40, 10 January 2017 (UTC)
...and thanks to @David Levy: for switching the Amy Satterthwaite image. Now her article is going to get even more hits! Yoninah (talk) 18:57, 10 January 2017 (UTC)
Have also just destubbed Henrietta Harley, Countess of Oxford and Countess Mortimer, which is currently in queue 4.
@Schwede66: at this rate, you can figure in some double hooks! Yoninah (talk) 18:57, 10 January 2017 (UTC)

A kind fellow editor has 'come to the rescue' and Bisalpur Dam is looking rather presentable. Thanks, Utcursch! Schwede66 04:40, 11 January 2017 (UTC)

I was inspired to expand Oecophylla smaragdina by seeing that tomorrow's featured picture linked to a stubby ant article with citation needed tags. Cwmhiraeth (talk) 07:40, 11 January 2017 (UTC)
This is all wonderful collaboration, but I don't think DYK editors should be expected to expand every stubby link. Perhaps we should consider linking only the featured article in each hook, as Gerda Arendt has mentioned is done on the German Wikipedia? Yoninah (talk) 10:48, 11 January 2017 (UTC)
No because many hooks contain technical or highly-specific terms which our general readers wouldn't not understand. The Rambling Man (talk) 10:51, 11 January 2017 (UTC)
I agree with the Rambling Man, but stress that our focus of attention should be the bolded article. Technical terms could sometimes be avoided. We don't need links to familiar composers, as suggested in a hook I rejected as boring. --Gerda Arendt (talk) 12:11, 11 January 2017 (UTC)
If there's one thing the above discussion makes clear, it's that this is one of umpteen situations on Wikipedia where a rule is not the solution, but judgement is required. We cannot eliminate all links, per TRM; but links can be a problem, per Schwede. Personally I think it might be a good idea to tweak the instructions, but these are in such a mess anyway that I'm not sure I can be bothered...Vanamonde (talk) 13:01, 11 January 2017 (UTC)

Linking only the target article in hooks[edit]

Discussion[edit]

Gerda Arendt's idea, mentioned above, seems to me a good one i.e. link only the bolded article in each hook, nothing else. [Later clarification: The proposal is to not have nonbold links by default, but not forbid nonbold links.] It's true, as T.R.M. said, that the hook may contain "technical or highly-specific terms which our general readers wouldn't not understand", but so what? Presumably they're linked from the article, so the reader can click on the article and go from there. The more I think about it the more obvious it seems that linking anything other than the target article dilutes the hook's purpose for no real reason other than imitation of the linking in articles -- but hooks aren't articles and don't have the same purpose. Thoughts? EEng 04:44, 12 January 2017 (UTC)

My feelings exactly. Any technical term that appears in the hook will of necessity appear in the article, so someone who wants to explore some side issue will just have to click on the hook. This will also circumvent the situation where the auxiliary link is more popular than the hook subject, as in this hook: * ... that Alan Hale, who discovered Comet Hale–Bopp (pictured), said that he "predicted" its appearance would trigger suicides—​​and it turned out he was right? where Alan Hale got 12,110 hits and the auxiliary subject, Heaven's Gate (religious group) (the last link), got 21,112 hits – and when the auxiliary link may be siphoning off views from the hook subject, as in this hook: ... that since it honors a notionally illegal cult, suburban Shanghai's Tianfei Palace is officially classified as a museum? where Tianfei Palace (Songjiang) got 10,030 hits and the auxiliary subject, Mazu (goddess), took another 4,151 hits. Yoninah (talk) 14:43, 12 January 2017 (UTC)
I looked into DYK clicks back in 2010, as summarised at User:EdChem/DYK clicks. Skip to the tables at the end if you don't want the methods used, but I found (from a small sample) that at least half of all additional views of articles on the front page in DYK went to non-bolded articles. At least cutting back on non-bolded links seems wise to me. EdChem (talk) 14:59, 12 January 2017 (UTC)
Good work, and interesting, but the way I see things it doesn't matter. If there are many clicks on "auxiliary" (let's call them) links, then they're siphoning off from the main link and we should therefore eliminate them; and if there are few clicks, then no one's using them anyway. Either way, get rid of them -- just have the one bolded link. EEng 19:43, 12 January 2017 (UTC)
Another example, my DYK for aluminium sulfacetate got 5020 views for the bolded link and 5490 for the only other link (5278 corrected for typical daily views). I didn't link "empyreumatic" to avoid syphoning clicks (and with only a Wiktionary link as a potential target), but didn't expect mordant to be such a popular click. From my DYK for e-baby, there were at least 5000 additional clicks for non-bolded links. Without the extra link, my DYK on section 127 of the Australian Constitution would likely have exceeded 5000 views. The predatory conference DYK had the clicks directed mostly at the bold link, while the aluminium triacetate DYK had a spike in views for all six linked articles. Just this quick survey of these of my recent DYK articles suggests that there is still significant interest in non-bolded links; the question is whether we view this as helpful to readers going to the articles they choose, or unhelpful syphoning of attention from the bolded article. That, to me, connects to another issue being discussed on this page, when the non-bolded links are of poor quality. I am concerned that we are inadvertently highlighting poor articles and wonder if we shouldn't avoid non-bolded links in cases where the target is poor quality. Going to bold-link-only would require an RfC, IMO, to ensure input from interested editors who work on main page content. EdChem (talk) 22:15, 12 January 2017 (UTC)
You've raised another great point, which I want to repeat: auxiliary links (as I'm calling them) often lead to poor-quality articles, right there from the main page. EEng 22:35, 12 January 2017 (UTC)
I have to agree with removing all unnecessary unbolded links (except for maybe April Fools Day), I've noticed this happen on a few of my hooks too. The C of E God Save the Queen! (talk) 22:43, 12 January 2017 (UTC)
There's something terribly wrong. We've got a sensible proposal for changing the DYK rules, and it's actually gaining traction. EEng 22:46, 12 January 2017 (UTC)
(ec) But we don't need a separate discussion to avoid linking "unnecessary ... links" that's WP:OVERLINK. Come on. The Rambling Man (talk) 22:50, 12 January 2017 (UTC)

──────────────────────────────────────────────────────────────────────────────────────────────────── I'm a fan of this idea. I think an alternate proposal might be to have some minimal quality standards for auxiliary links. That way it disincentivizes overlinking, but still allows for specialized terms or related topics with good coverage to be linked when appropriate. Just to play devil's advocate though, there may be some value in linking to less-than-stellar articles as it may encourage improvement of them. We may want to think about the impact the added traffic has on improvement of those auxiliary linked articles. Wugapodes [thɔk] [ˈkan.ˌʧɻɪbz] 22:55, 12 January 2017 (UTC)

Noooooooooooooooooo! The last thing we need is to turn every review of one nominated article into a review of 3 more linked articles too. My proposal is to eliminate, by default, all links in hooks other than the bolded link. [Later clarification: The proposal is to not have nonbold links by default, but not forbid nonbold links.] I suppose there might be some special reason to link something once in a while (though I can't think what that would be right now) so the idea isn't to make a rule forbidding such links, just saying that in general auxiliary links should be omitted -- like I just said -- by default. EEng 23:11, 12 January 2017 (UTC)
Yes, don't worry, I agree with you wholeheartedly. But this is presumably a discussion, and this will likely be brought up if wider comment is sought. I think that's actually how WP:ITN evaluates its content. They have far less of a backlog though which is probably why it works there and why it wouldn't work here. Wugapodes [thɔk] [ˈkan.ˌʧɻɪbz] 23:18, 12 January 2017 (UTC)
Wait... what is "how ITN evaluates its content". What does evaluation of content have to do with anything? We're talking about linking. EEng 23:48, 12 January 2017 (UTC)
Non-bold links at ITN (if I remember right) have minimum quality requirements for non-bold linked articles as well as bold-linked articles. Actually it's been a while since I read their guidelines so I might be wrong here. Wugapodes [thɔk] [ˈkan.ˌʧɻɪbz] 23:58, 12 January 2017 (UTC)
It just happened again. For the hook ... that in 1429, John Beaumont, 1st Viscount Beaumont, was knighted by a seven-year-old king?, the bolded article got 3,254 hits, and the cherry at the end of the article, the seven-year-old king, got 6,409 hits. Too much linking is definitely diluting the exposure of our featured articles. Yoninah (talk) 00:10, 13 January 2017 (UTC)
That's a great example, too. The novelty of a 7yo king is meant to pique interest in the Beaumont article, but by linking the kid-king it completely backfires by drawing all the clicks away. (You have to be careful in interpreting the data, as EdChem points out, to remember that the kid-king gets a lot of clicks normally anyway, but the point remains.) EEng 00:19, 13 January 2017 (UTC)

Could post this here or in the item above. Thalia Awards, linked from the Petra Hřebíčková hook, is unsuitable to be linked to from the homepage. The hook needs to go back for further discussion as the DYK article has an unreferenced section. Schwede66 02:38, 14 January 2017 (UTC)

Vote[edit]

  • Support proposal 100%. — Maile (talk) 00:20, 13 January 2017 (UTC)
  • Support proposal 130% (as proposer, who stole the idea from Gerda). EEng 01:17, 13 January 2017 (UTC)
Comment I just noticed that Rule 4 says "Articles for DYK (including other articles linked in the hook) must conform to the core policies of Verifiability, Living Person Biographies and Copyright". To be honest, I don't think I ever remembered to apply that rule to the non-bold-linked articles, and unless I'm very much mistaken no one else (or almost no one else) is either. If we were to actually start trying to do that -- check all the non-bold-linked articles -- I think the entire DYK process would finally come to a total, final, and (if we are lucky) irrecoverable collapse. Realistically I think we have to eliminate non-bold links. EEng 01:28, 13 January 2017 (UTC)
CommentComet I've seen it. That thing where somebody thinks all linked articles should yadda yadda yadda yadda. But that's like saying that all links in the Main Page Featured Article blurb would have to be FA standard. On second thought, I'll see your 130% and raise my support to 130.5%. — Maile (talk) 01:35, 13 January 2017 (UTC)
REMOVED - EEng, that tidbit in the rules was added in August 2016 without consensus, and by someone who doesn't even come up on a QPQ check as having had a DYK. Very limited editing history by that editor. So, I just removed it as no consensus. — Maile (talk) 01:52, 13 January 2017 (UTC)
Good. But the point still remains that every time we have a nonbolded link in a hook, we're linking from the main page to an article of completely unchecked quality, and that's a bad thing. EEng 11:03, 13 January 2017 (UTC)
  • Fun with statistics (caveat, these are meant to be quick and dirty summary of the last few months at DYK. There are many problems with drawing conclusions from these statistics. Take them with a grain of salt) I was bored, so I decided to do some statistics and hypothesis testing to try and quantify the problem. The average bold link gets about 2100 views while the average non-bold link gets about 260. This is considered a medium effect size with 32% of bold links getting fewer hits than the average non-bold link. nitty gritty stats Over the last 200 revisions to T:DYK and last ~1770 links in the template, unsurprisingly, bolded links get significantly (t(1770)=660.6, p < 0.00001) more hits than non-bolded links but this could be for a number of reasons. Hits on bolded links were more variable than non-bolded links with bold links having a standard deviation of 5350 hits while non-bold links have a standard deviation of 2000. The effect size is medium at 0.46, meaning that 68% of bolded links are above the mean for non-bolded links were they normally distributed which they most definitely aren't. Do with that what you will. Wugapodes better living through statistics 02:42, 13 January 2017 (UTC)
(I still support the proposal. I think the rationales are fundamentally sound. While the stats show it's not a systemic problem, they also show there are a lot of outliers, and I think reducing those would be a positive) Wugapodes [thɔk] [ˈkan.ˌʧɻɪbz] 02:45, 13 January 2017 (UTC)
Wugapodes, you're a great addition to the project but your statistics are enthusiastic nonsense. EEng 11:03, 13 January 2017 (UTC)
@EEng: Is that a critique of the write-up or the methods? If the former, I've updated it to explain the numbers in regular terms and the more important stuff first. If the latter, yeah, probably, I did it in a couple hours so there are lots of improvements that could be made, but I see them as a tool to answer a question, not an answer themselves. If there's a particular suggestion you have or question you want taken into account let me know. Wugapodes [thɔk] [ˈkan.ˌʧɻɪbz] 18:35, 13 January 2017 (UTC)
Wugapodes Good work with your statistical analysis. Just in case you get bored again, I can offer two more hypotheses that you could test. Firstly, if you restrict your analysis to lead hooks, my hypothesis is that the difference in page views between auxiliary links in lead hooks with bold hooks is less than what you found when you tested the sample of all hooks. Secondly, and more importantly, an image referring to an auxiliary link (as determined by the placement of pictured) will result in significantly less difference in page views between auxiliary hook and bold hook. Schwede66 18:57, 13 January 2017 (UTC)
Wugapodes, please pardon my bluntness, but misuse of statistics is a hot button of mine. (To quote myself, "Blindly plugging numbers into statistical formulas has caused a lot of problems, as the ex-managers of the Chernobyl and Fukushima nuclear plants would be able to tell you first-hand (if they weren't both dead, of course).") You seem to be exploring (though incoherently – there's that bluntness again) whether nonbold links get different numbers of clicks than bold links, and that's not the question. The question is whether the presence of nonbold links draws clicks away from bold links. (That, at least, is the statistical question; the still leaves questions statistics can't answer, such as whether that's a bad thing or not, what the ultimate purpose of hooks is, and so on.)
A few things that are obvious, and not obvious, can be seen in the view stats from one hook's appearance last year [1][2][3] – the first is the bold link (Paddy Murphy) and other two are the nonbold links. You don't need any fancy techniques to see that (a) it's necessary to correct for the base number of clicks articles get on a normal day anyway; and (b) the two nonbold links each got 2000 clicks they would not have got had they not been in a hook that day. What's not clear is (c) how many of those clicks would have gone to the bold link (which got 10,000 clicks) had the nonbold links been omitted; and (d) whether we care about that anyway. Answering (c) with any confidence would require an experiment (as would answering Schwede66's question about the effect of pictures -- I can't even understand what Swede's first question is asking). EEng 20:21, 13 January 2017 (UTC)
EEng For reasons's I'll elaborate on your talk page, I'm not convinced those are systematic biases in the data. However, I am doing more detailed work and will post those results here when I'm finished, for the rest I will respond on your talk page so as not to clutter this page with fancy statistics talk. Wugapodes [thɔk] [ˈkan.ˌʧɻɪbz] 22:05, 13 January 2017 (UTC)
Bias has nothing to do with it. You've computed a bunch of numbers with no clear objective. I'll be happy to discuss this elsewhere, as you suggest. EEng 22:58, 13 January 2017 (UTC)
  • Support As I said above having experienced this before, I fully support this and that seven year old king example above is further proof this change is needed. The C of E God Save the Queen! (talk) 10:39, 13 January 2017 (UTC)
  • Oppose people will just shrug shoulders at hooks they don't get and move on, linking helps stave that off. Also, driving readers to find other pages from the DYK section is a good thing. If everyone could stop being so precious about pageviews of their own nominations, and consider the purpose of the main page more objectively, the thousands of auxiliary hits generated through these other links should be recognised as positive, not negative. P.S. Having done considerable work on OTD lately, their hooks average about three to five times as many hits as the quirky section. They also have auxiliary terms linked. It's not doing them any harm. Perhaps it's just the content of DYK that's the issue, not the way the hooks are linked. The Rambling Man (talk) 11:19, 13 January 2017 (UTC)
  • Support The whole purpose of the hook is to draw the reader to the new article being featured, not to direct attention all over Wikipedia. Yoninah (talk) 12:59, 13 January 2017 (UTC)
  • Oppose I think it is unusual for me to agree with TRM, but this is, in my view, throwing the baby out with the bathwater. We need links in hooks, because we are dealing with a global audience, who frequently might not have the faintest idea what certain terms are. Links can make a hook hookier if they allow people to click on something and understand the hook. Now I get that there are problems with some links, but this just needs a quick check during the promotion process: in fact, its among the quickest checks we can ask people to perform. Vanamonde (talk) 13:15, 13 January 2017 (UTC)
  • Support - If the reader doesn't understand everything in the hook, he can click on the link to find out more. Cwmhiraeth (talk) 13:56, 13 January 2017 (UTC)
    No, he won't bother. He'll just go to OTD instead where the non-quirky factual hooks with auxiliary links get three or four times as many hits as the DYKs do. The Rambling Man (talk) 14:00, 13 January 2017 (UTC)
  • Cite your source for this assertion.--Kevmin § 14:05, 13 January 2017 (UTC)
  • Don't be silly. Pageviews of OTD hooks by far outweigh the general calamitous input at DYK three- or four-fold. Best thing there is there's no "ownership", no "desperate grab for pageviews", just a dedication to the purpose of the encyclopedia. DYK has a lot to learn. The Rambling Man (talk) 17:26, 13 January 2017 (UTC)
  • Support Anything used in the hook should be in the article itself already. The interest in what an unusual term/person is in a hook should lead to clicking the hook link. I have tried to avoid over-linking in any hook i write as a default, and this just clears up any problems that main page protection crusaders will have.--Kevmin § 14:05, 13 January 2017 (UTC)
  • Oppose We've always encouraged people to limit the number of links in a hook, and to try to arrange the hook so that the nominated article's hook is first. I'm happy with continuing that practice. Still, there are sometimes terms or phrases that are important to understanding the hook and should be linked there. To forbid other links makes it harder on the reader, who will have to search for it once they get to an article (and it may not be easily seen there), assuming they don't skip over the hook because it's too opaque. BlueMoonset (talk) 17:03, 13 January 2017 (UTC)
The proposal isn't to "forbid" other links, just that they not normally be linked in the same routine way we link stuff in articles -- to take a more conservative approach to linking. EEng 20:21, 13 January 2017 (UTC)
When I read it and posted my opposition, the proposal said there could only be bold links. If only bold links can be used, non-bold links are therefore forbidden. I see you've revised the proposal, but I'll maintain my oppose; I think encouraging fewer links without heavy-handed rules is the better way to proceed. BlueMoonset (talk) 05:59, 14 January 2017 (UTC)
Well, people started voting before I'd had a chance to really think through what I was proposing. EEng 06:16, 14 January 2017 (UTC)
  • Neutral but leaning more towards oppose than support. I don't mind auxiliary links; in fact, they can be useful. I'm not precious about where readers click; if it's not the bold hook but they are more interested in the auxiliary link, then so be it. What I mind is auxiliary links pointing to stubby stubs, especially when they are in the lead hook, and most certainly when the hook image relates to the auxiliary link. What I like about the proposal is that it will avoid this from happening. But there could be a much simpler way of addressing it, and that is "if it's a stub, it can't be linked to from the hook". Schwede66 19:02, 13 January 2017 (UTC)
  • Support the general idea but don't think should be a hard rule. We have too many rules and we already use our editorial judgement in preparing and approving hooks. Andrew D. (talk) 19:32, 13 January 2017 (UTC)
Definitely not meant to be a hard rule (though I didn't make that clear at first). I think the idea is that a hook starts with no nonbold links by default, and they're added very judiciously where they serve the hook's purpose (though I'm note sure I can explain what that means exactly, at this point). EEng 20:30, 13 January 2017 (UTC)
  • Yes, as I noted above, we already have WP:OVERLINK which should be applied in any case. I can see shedloads of cases where it's necessary to link terms in these hooks, and I can also see that our editorial base and readers will see this as a downgrade just to satisfy the precious pageview-centric DYK nominators. That's not what it's about, as ably demonstrated by OTD. They regularly get pageviews of 10k to 20k for the most mundane things, yet a "good" DYK might creep over 5k. It feels a little like re-arranging the deckchairs on the Titanic so I'm delighted to go along with whatever suggestion gains most traction, but the project should know that they're batting bottom of the pile, despite having a huge portion of the main page. The Rambling Man (talk) 19:41, 13 January 2017 (UTC)
Christ, give it a rest, will you? There's no way to tell why OTD gets more views – it may just have to do with the kind of material that tends to be listed there, nothing about some "quality" difference. Maybe people just like historical stuff more. This isn't to say DYK doesn't have serious quality problems – it does – but this isn't evidence for it.
You may now insert one of your standard lashings-out. EEng 20:30, 13 January 2017 (UTC)
You mean like "Christ, give it a rest, will you?" That OTD is maintained by one individual, and is free of ego and free of ownership and free of your kind of communication style yet still massively overtrumps DYK in terms of page views is all that needs to be recognised. You can all shuffle around and pretend that what you're doing is the core of DYK essence, to bring "new and improved articles" to the attention of our readers but as it turns out, they couldn't care less. On one day last month, my "talk page" got more hits that the average DYK. You need to give it a rest, you need to look harder at what you're doing, and if re-organising the deckchairs to try to manipulate figures to justify the existence of the project is your thing, go for it. It still doesn't explain why all this arcane effort goes into basically ZERO interest. The Rambling Man (talk) 20:54, 13 January 2017 (UTC)
You're making stuff up as usual. On a random recent day (Jan 9) OTD items (700, 2500, 4500, 6500, 7000) got about the same views as DYK (2200, 3000, 3500, 4000 , 5000, 6000, 6500, 12000), and if anything fewer. A fraction of OTD-listed articles get very high view counts (a recent appearance of my own pet article got 60,000+ views) but much if not most of that can be ascribed to mentions of the subject elsewhere on the net (these being anniversary items, after all), plus OTD items (unlike DYK items) are selected for their general interest in the first place. EEng 22:58, 13 January 2017 (UTC)
You're making stuff up as usual. careful now, that kind of talk could result in a block! And it's noted. In any case, just a day or so ago, OTD smashed DYK into the long grass, embarrassing the quirky project wholesale. And it did it without the personality overdrive (e.g. you) and without the ownership (e.g. Maile, Yoninah) and without the excuses (e.g. Cwmhaerith). It also did it without all the endless and arcane nomination templates, prep areas, queues etc. You really do need to think again if you genuinely believe you're appealing to new editors. This latest attempt to "optimise" DYK to "maximise" hits on a particular page is utter bullshit, perpetuated around a profoundly disturbing need for "pageviews" from the DYK projecteers. How odd. The Rambling Man (talk) 23:21, 13 January 2017 (UTC)
Let's take some real figures, 12 Jan, OTD had 13,775, 8,578, 5,798, 21,748, and 18,063 hits. Now I'll leave it to you to tell me how may hits the quirky project got. The Rambling Man (talk) 23:25, 13 January 2017 (UTC)
Well, for the "mo(u)rning session" DYK got 4,049, 1,204, 1,298, 1,942, 1,062, 643, 3,791. Stunning. Add them all up and get just about what one of the OTD topics (e.g. Occupation of the Ruhr) got. I'd say a three-to-five-fold difference is normal. DYK just doesn't interest our readers right now. So reducing the interest by limiting the links, how does that help our audience since it's been proven that other links generate interest in Wikipedia? Or is it all about the credits? The Rambling Man (talk) 23:32, 13 January 2017 (UTC)
(ec) The DYK stats you're giving are for 0000 to 1200 UTC, offpeak for the English wikipedia. For the other 12 hours the stats are 1200, 3000, 4500, 8000, 12000, 14000. And, again, those are for 12 hours (not OTD's 24). What are you trying to prove? You keep ignoring that DYK and OTD are, by design, quite different and wouldn't be expected to get similar numbers of views no matter what. That DYK consumes more editor time than does OTD has already been explained: DYK never runs anything twice and requires a review of the article, while OTD works from the same menu of existing items year after year. EEng 00:12, 14 January 2017 (UTC)
Is there any point to this? Or is it a case of "my project is better than your project, doo dah"? By the way, have you seen how many pages link to Occupation of the Rhur? Quite a few more than the average DYK. So it stands to reason you cannot claim that all views for articles on OTD are directly from being in OTD as it has other sources for views. Most DYKs don't have that luxury and gain most of their views from their spot on DYK. The C of E God Save the Queen! (talk) 23:56, 13 January 2017 (UTC)
Yes, of course there's a point. OTD makes little-to-no effort every day to knock up some main page hooks. The audience responds; three or four times more people go for OTD items than DYK items. OTD doesn't ban links on auxiliary topics. That's what this discussion is all about. The Rambling Man (talk) 00:06, 14 January 2017 (UTC)
Well, as both I and CofE have both tried to get you to understand... (oh, forget it). EEng 00:12, 14 January 2017 (UTC)
TRM, the largest of several flaws in your argument is that you're not correcting for the baseline level of traffic to each article. It's not surprising that brand-new articles have low baseline traffic. Opabinia regalis (talk) 00:59, 14 January 2017 (UTC)
Oh dear. I suppose you looked at all those OTD pageview graphs, right? The spike shows when it was on the main page. That's the relevance. Not the baseline. The spikes easily outdoes anything on DYK. Take Theodora for instance. The "baseline" hits you refer to averages out at 100. The day it featured it saw 13,775. So that's a hike of 13,675 over the "baseline level of traffic". Significant. Or try the Bulo Marer rescue attempt, baseline average was optimistically 40. The day it featured the pageviews rose to 18,063. So that's a spike of 18,023. So I'm not sure what you're on about. Perhaps your own argument is flawed? The Rambling Man (talk) 08:41, 14 January 2017 (UTC)
Opabinia regalis sorry, I forgot to ping you, looking forward to your response about my "largest of several flaws" (sic). The Rambling Man (talk) 22:44, 15 January 2017 (UTC)
Good, thanks! If you're going to claim to be Mr. Accuracy, you have to do the stats right. Next step, if you think this is a useful metric, is to look at a larger sample rather than a small handful of articles, making sure to correct for the amount of time an article is linked on the main page. Of course, what that really says is that to a first approximation nobody cares about anything linked on the main page - 6k or 13k is a drop in the bucket of 23 million views. (I believe I'm on the record in past discussions of main-page content in thinking we ought to just delete the main page and replace it with a search box.)
I wasn't (just ;) being snarky in my other comment, though - I'm genuinely surprised and perplexed that someone who has put so much time and energy into highlighting concerns about poor or inaccurate material on the main page would happily advocate for links to unvetted articles, even after seeing evidence that at least some readers are choosing to consume that content instead of clicking on the vetted, bolded link. Opabinia regalis (talk) 04:08, 16 January 2017 (UTC)
Opabinia regalis My stats were accurate, as demonstrated. To the links: somewhat ironically, DYK's auxiliary links may well be a better condition than the target links. All other sections of the main page (TFA, OTD, ITN, TFA, TFL) have auxiliary links to help the reader. They don't "vet" the auxiliary links. Why is DYK so "special"? Is it because the the precious need for pageviews for the DYK owners? There's no evidence anywhere to support that removing auxiliary links will help our readers enjoy the content of Wikipedia, nor will it drive more hits to the bold targets. But you already know that. The Rambling Man (talk) 12:28, 16 January 2017 (UTC)
  • leaning Oppose, we have relevant links to help readers to other relevant topics. This should be as true in a DYK hook as any article. Cas Liber (talk · contribs) 19:53, 13 January 2017 (UTC)
  • oppose. This is a wiki, the point of which is to provide links not just in indices or contents but within content, so within the article text and on every other user facing page, including the main page. It would be distinctly odd to make DYK the one place this is not practiced. The key thing is to use common sense. Not link every word or even every term that has and article, but link ones that are interesting and relevant to the article or hook.--JohnBlackburnewordsdeeds 22:05, 13 January 2017 (UTC)
    Indeed, despite DYK hooks not being "articles", I still think WP:OVERLINK should apply, i.e. don't link common sense terms, but as and when necessary link terms to avoid ambiguity and to assist our reader, not assist pageviews for our precious DYK nominators. The Rambling Man (talk) 22:48, 13 January 2017 (UTC)
  • Leaning oppose - maybe I'm missing something, but the whole premise of this idea seems off to me. Readers demonstrably do have an interest in the peripheral/auxiliary links (and/or don't always understand the significance of the bolding). Why would we make changes constructed to stop readers from consuming the content they want, so we can redirect them toward consumption of the content we want them to want? Opabinia regalis (talk) 22:56, 13 January 2017 (UTC)
    Because it undermines the general "I want hits" mentality of the members of the DYK project. It's not about increasing traffic into Wikipedia, it's not about interesting hooks that embellish the encyclopdia, it's about ring-fencing hits to make the nominator feel great about their singular contribution to Wikipedia. Which, of course, is bollocks. The Rambling Man (talk) 22:59, 13 January 2017 (UTC)
    You know, I would've guessed you'd be on the other side of this argument. ZOMG, nobody really ever checks the other linked articles, there might be typos in there! What if we accidentally link from the main page an article whose dashes are the wrong length?? ;) Opabinia regalis (talk) 00:59, 14 January 2017 (UTC)
    Ah yes, of course. The path less travelled, and all that. Martinevans123 (talk) 23:48, 13 January 2017 (UTC)
    I'm sure you're enjoying your own little session, but are you actually actively contributing to this discussion or using it as another excuse to demonstrate how to use piped links? The Rambling Man (talk) 23:51, 13 January 2017 (UTC)
    Ah yes, shucks, got me there. Maybe we need a percentage click-thru comparison of how DYK piped links compare with ITN and OTD piped links? Martinevans123 (talk)
    Again you're so clever you're leaving the audience cold, like DYK. I'm just talking about clicks on OTD targets vs clicks on DYK targets. The difference is stark and the irony is that OTD targets take a few moments to select, while DYK targets are subject to a litany of arcane regulations and then usually promoted in a crap state, to then go on to get maybe 1000 page views. The Rambling Man (talk) 00:06, 14 January 2017 (UTC)
    Cold? I imagine they're dead by now. Who wants a mere 1,000 hits anyway, eh? Martinevans123 (talk) 00:16, 14 January 2017 (UTC)
  • Oppose per The Rambling Man. More clicks are a good thing. A rising tide lifts all boats. If a hook piques the interest of readers in articles about seven-year-old kings, we have succeeded in interesting a reader. Maybe they'll also feel like digging deeper into the subject and click on the main link as well, who knows? It's not a zero-sum game. —David Eppstein (talk) 01:25, 14 January 2017 (UTC)
  • Support It might lead to more interesting hooks, because hook-creators can't rely on other links to provide interest. For example, because I had done quite a bit of work on the Dorchester, Dorset article, when it later became a GA and an item in DYK, I took an interest in what the hook was going to be, and I was disappointed to discover that it was "...that the Tolpuddle Martyrs were tried in the Shire Hall at Dorchester?", because it seemed to me to be a hook that made Dorchester almost incidental (and sure enough, the page views show most readers weren't interested in Dorchester - 3,677 for the Tolpuddle Martyrs but only 814 for Dorchester, both from a similar baseline). Now, if the Tolpuddle Martyrs couldn't be linked, maybe a different hook could have been created, one which might have been more interesting with regards to the town of Dorchester. PaleCloudedWhite (talk) 01:40, 14 January 2017 (UTC)
  • Oppose. The point of DYKs is not to stroke creator's egos, but to enlighten readers. Everyone wins if a DYK gets lots of hits for an article that isn't bolded. Espresso Addict (talk) 11:56, 14 January 2017 (UTC)
  • Oppose Hooks are better understood if there's links to relevant specific terms. I'll note that only 3 of my 50-60 hooks ever have not had another link in them. Joseph2302 (talk) 12:47, 14 January 2017 (UTC)
  • Supportish I agree with Joseph's point and others on that side, but I am more concerned if it is actually true that the non-bold link articles have to be improved first, then this whole thing grinds to a halt on peripheral links - maybe the thing to do is warn noms of this, and add a requirement to noms and reviewers that they have to work on and insist on improvements to peripheral links too (if the standard is that promoters have to not promote these things). My hesitancy is having nomed and reviewed in the past, I don't recall anyone ever mentioning I had to make sure the other linked articles were good too. -- Alanscottwalker (talk) 13:11, 14 January 2017 (UTC)

Hold it, people![edit]

As the proposer, I'd like to interrupt for a minute. I made an offhand suggestion, and suddenly everyone's voting. Could we perhaps have a discussion about good approaches to linking? It might help if people find examples of hooks they would have linked differently. Even though I unthinkingly made it sound that way at first, I never meant for there to be an absolute prohibition on hooks, just a conservative approach recognizing that links in hooks might serve a somewhat different purpose than do links in articles.

Please keep the discussion free of tiresome rumbling about how terrible DYK is, how nothing will every pull it from the crapper, etc. Just focus on the question of what is the best approach to linking in hooks.

(signed)--> Your Pal EEng 14:10, 14 January 2017 (UTC)

Now then... let's look at

... that in 1429, John Beaumont, 1st Viscount Beaumont, was knighted by a seven-year-old king?

I think seven-year-old king should not have been linked. It's in no way necessary to understand the hook, and (as observed above) competes with it. I'm not sure about knighted -- as some have argued, maybe we should link things that at least some readers might not understand. EEng 14:10, 14 January 2017 (UTC)

... not just because it's a playful or quirky Easter egg link that might amuse some people, or annoy and enrage others? Martinevans123 (talk) 16:40, 14 January 2017 (UTC)
... It would have encouraged me to click on the DYK link, not discouraged me. It's straightforward, apply WP:OVERLINK as any other "decision" will be summarily ignored because every hook's linking will become subject to debate over whether it's overlinked or not... The Rambling Man (talk) 16:43, 14 January 2017 (UTC)
It all depends on what you want readers to look at. If the whole point of DYK is to highlight new content, I would have thought that is more likely achieved by only linking the article with that new content. If DYK doesn't have that purpose, then, er, why the restrictions on which articles qualify for inclusion? PaleCloudedWhite (talk) 16:56, 14 January 2017 (UTC)
Agree. That sounds like 100% common sense to me. Martinevans123 (talk) 17:08, 14 January 2017 (UTC)
In all of this discussion, it seems like we're not considering what's actually best for the reader, just for the pageview hungry DYK owners. The Rambling Man (talk) 17:14, 14 January 2017 (UTC)
Well, what is best for readers? Above I posed the question of what DYK is for. Is it legitimate to direct readers to new content? If it is, DYK should be organised to maximise focus on that new content. If it isn't, why have any restrictions on which articles qualify for inclusion in it? If any article could qualify, at least then we might have more interesting hooks. PaleCloudedWhite (talk) 17:22, 14 January 2017 (UTC)
That's part of it, and that's done by use of the bold target linking. Auxiliary links help the reader should additional context or subject matter expertise be required. WP:OVERLINK should be all that's needed. The Rambling Man (talk) 17:33, 14 January 2017 (UTC)
Is DYK primarily meant to encourage readers, by showing them material they had no idea existed? Or is it to encourage editors to create articles that would be interesting to the general readership and expand articles that are currently languishing as stubs? Or perhaps both? How does one find the right balance when these two objectives might conflict? Martinevans123 (talk) 17:36, 14 January 2017 (UTC)
The needs of the reader should far outweigh the needs of the editors. Besides the vast majority of DYKs aren't the first articles created by editors, indeed there's a regular cadre of editors nominating the same old stuff to DYK (Hawaii politics, Indian politics, Paralympic athletes, obscure insects), this isn't benefiting the editing community at all. The Rambling Man (talk) 17:42, 14 January 2017 (UTC)
Without editors to edit there'd be nothing for readers to read, so the needs of editors count too. DYK's mission is to highlight new and expanded content, so the newness of the editor doesn't matter. I agree that the long strings of articles on insects, Pennsylvania waterways, boat races, etc. overdo it. EEng 17:48, 14 January 2017 (UTC)
As hilarious as I find most of your comments and baiting, please double check how many "boat race" DYKs have run on the main page (out of the possible total, naturally). Otherwise, take your goading elsewhere. The Rambling Man (talk) 20:21, 14 January 2017 (UTC)
Huh? Anyway, just to reiterate, DYK's mission is to highlight new and expanded content, so whether the editor is new doesn't matter. If the thought of having their work read by interested readers motivates the creation of new articles, that's fine. People who don't write stuff people want to read might not understand that, of course. EEng 21:17, 14 January 2017 (UTC)
Huh? I think what you just said there was "stuff that's already been said" and nothing much more. The point here is that we don't you to bait people and we don't need new rules specific to DYK relating to linking. WP:OVERLINK does the job just fine. In the meantime, have a great Saturday! The Rambling Man (talk) 21:24, 14 January 2017 (UTC)
Thanks for reiterating what reiterate means. EEng 21:37, 14 January 2017 (UTC)
You can say that again. Martinevans123 (talk) 21:50, 14 January 2017 (UTC)
That. EEng 22:53, 14 January 2017 (UTC)
New and expanded content is only highlighted for the editors who submit here. Regardless, we shouldn't be overriding MoS for clickbait. There was a discussion some years ago about how to write hooks to lead with the bolded link, which is good practice (maybe someone can find it?). Like FAC, DYK should follow regular linking policies. Victoriaearle (tk) 17:56, 14 January 2017 (UTC)
MOS is for articles (though of course big chunks of it should be applied elsewhere e.g. technical stuff like dates and units of measure). Hooks aren't articles and have a different purpose. Recognizing that wouldn't be "overriding MOS". EEng 18:28, 14 January 2017 (UTC)
While hooks aren't articles, they are complete sentences and there's no reason at all (at least none given here) for all but the quirky hook to comply with MOS and comply with OVERLINK. That way you remove the possibility of yet another tinkering "supplementary rule" in this already arcane and over-the-top system which actively discourages new editors and new articles being nominated through its unnecessary complexity. The Rambling Man (talk) 20:21, 14 January 2017 (UTC)
The reason's been given: to channel traffic to the bookhook. You not accepting the reason doesn't make it not a reason. EEng 21:22, 14 January 2017 (UTC)
The reason isn't sound. There's not reason for our readers to be channeled to the "book", they should be allowed to be intelligently directed to items of interest. Stop tinkering with the deckchairs, this is a waste of time, and as noted above, this isn't zero sum problem; you have precisely zero evidence that just removing all the other links will suddenly result in a leap in DYK hook hits. The Rambling Man (talk) 21:25, 14 January 2017 (UTC)
No leap is predicted, nor has anyone but you mentioned zero-sum. Clever how you picked up on my typing book for hook. Devastating! 21:37, 14 January 2017 (UTC)
If Jimbo starts selling books, I'd certainly buy one. Martinevans123 (talk) 21:51, 14 January 2017 (UTC)
Well it shows how much you pay attention, book, claim of no mention by anyone else, claim of no leap. I begin to wonder if you're having some parallel discussion elsewhere. You're certainly working very hard to undermine anything you've already said in this debate. And honestly, if "no leap is predicted", what's the actual point of this entire discussion? Another case of a solution looking for a problem? More "me me me"? The Rambling Man (talk) 22:18, 14 January 2017 (UTC)
Why is everything about you? EEng 22:53, 14 January 2017 (UTC)
It's not, you made several errors there, it's all about you. The Rambling Man (talk) 22:56, 14 January 2017 (UTC)
Not overriding MOS sounds legitimate, except that if that means hooks are written in such a way that readers aren't that interested in looking at the new/expanded article, then, as I asked above, what is the point of the restrictions on which articles qualify for inclusion here? PaleCloudedWhite (talk) 18:01, 14 January 2017 (UTC)

The rumbling having apparently subsided...[edit]

...let's try again. How about this?

  • When adding links to hooks, consider whether a given link helps the reader by explaining something unusual understand the hook, or merely distracts from the bolded link.

Thoughts? EEng 22:53, 14 January 2017 (UTC)

Let's try again, hooks should comply with WP:OVERLINK. In fact, your suggestion would definitely mean we'd link "seven-year old king" because that's "unusual". You're wasting time here. The Rambling Man (talk) 22:57, 14 January 2017 (UTC)
I asked for "thoughts". EEng 23:12, 14 January 2017 (UTC)
Yes, hooks should comply with WP:OVERLINK. The Rambling Man (talk) 08:04, 15 January 2017 (UTC)
  • Until now, I've been under the impression that we add links to technical terms, to geocentric features that readers in other parts of the world might not be familiar with (like names of U.S. states), and to names of people or awards. As an aside, I'm wondering if editors involved in this discussion feel that readers actually take the time to click on more than one hook in a link? Do they click on "seven-year-old king", and then go back and click on the bolded article? I doubt it. Yoninah (talk) 00:21, 15 January 2017 (UTC)
I think editors take a variety of approaches to linking in hooks, perhaps unconsciously. I'm beginning to think that the desirable outcome of this discussion would be something like I proposed above in this subthread, simply reminding editors that links in hooks may serve a somewhat different purpose than they do in articles, leaving them to judge for themselves what to do with that, and not legislating some rigid approach.
There's really no way of knowing how many of those clicking on a nonbold link are "lost" i.e. never come back to click the bold link. Certainly it's some proportion of cases, but what proportion seems to be the subject of vigorous speculation, as is the question of whether we should care. EEng 00:32, 15 January 2017 (UTC)
And there it is. There's no such thing as a "lost" click. Anything that encourages an editor to click into Wikipedia from the main page is a good thing. Reducing the number of links to improve the scores of DYK owners is going to be detrimental to that. The Rambling Man (talk) 08:04, 15 January 2017 (UTC)
Sometimes less is more. It makes more sense to concentrate the click traffic on the main article, where people can then either get a better and fuller explanation within the article or click through to the additional parts of the hook through the main article. The C of E God Save the Queen! (talk) 08:38, 15 January 2017 (UTC)
Not at all, how does it "make sense to concentrate the click traffic"? If a reader happens to discover there's an article about a seven-year-old king but happens to be entirely disinterested by the hook, we can still be hopeful that he'll click on the non-bold link and still enjoy the encyclopedia. Removing links does not serve our reader. The Rambling Man (talk) 08:47, 15 January 2017 (UTC)
  • @The Rambling Man and EEng: For Chrissake will you both stop? You've both made your point. In this particular case I agree with TRM, but by keeping on at each other you are a) not convincing each other, b) not convincing anyone else, and c) probably driving people away from this discussion by burying it in long threads. Please. Let others weigh in. Vanamonde (talk) 09:19, 15 January 2017 (UTC)
    Agreed and if we could reduce the snide personalisation too that'd be great. The Rambling Man (talk) 09:33, 15 January 2017 (UTC)

I wonder whether we might offer advice to avoid wordings that put the "hooky" parts in non-bolded links. Take the previously mentioned example:

An argument can certainly be made that, as a reader unfamiliar with DYK, I would click on seven-year-old king looking for information on this knighting. But, would I have found it? This fact is much more significant in Beaumont's life than in Henry VI's, so we are arguably not serving the reader with such an Easter Egg link to an article which potentially does not include the fact. One response would be to mandate that any link in a hook go to an article which includes the fact, and so if the fact is not worth including in the article then the article is not worth linking, in that clicking a link in a DYK hook presumably seeks information on the fact. I say "presumably" as my data from 2010 (linked above) includes a hook where [[most valuable player|MVP]] got a large number of clicks, presumably seeking the meaning of the abbreviation. There are also considerable clicks on the image, and perhaps the image should also link directly to the article?

In any case, another possibility is to recommend avoiding what amounts to seductive details in non-bolded links. I would be very interested to find out (though we can't do the experiment) how the clicks would have been distributed for the hook re-worded as:

Would readers intrigued by the King being seven have clicked on the link to Henry VI or to Beaumont? What about these wordings:

My point is that the same information and links can be provided without drawing the likely-only click to an article unlikely to provide information on the fact in the hook. Readers will follow their interests, and that is fine, but we avoid Easter Egg links based on the principle of least surprise, and it is worth wondering what a reader might have expected in the hook as it ran when clicking on "seven-year-old king". EdChem (talk) 10:18, 15 January 2017 (UTC)

It seems he was five, not seven. Just sayin' Martinevans123 (talk) 10:23, 15 January 2017 (UTC)
Ok, Martinevans123, and yes, under the principle of least surprise, you have a point... but my comments had nothing to do with reviewing and we all know that going down that road will derail this discussion – again – from the topic. EdChem (talk) 12:31, 15 January 2017 (UTC)
It now seems that it's not that straightforward. Sources differ. I'm sure we're not going down that road, haha. Martinevans123 (talk) 12:50, 15 January 2017 (UTC)
  • I think EdChem has it right: One response would be to mandate that any link in a hook go to an article which includes the fact, and so if the fact is not worth including in the article then the article is not worth linking, in that clicking a link in a DYK hook presumably seeks information on the fact... That's a great way to see it. EEng 00:10, 16 January 2017 (UTC)

Template:Did you know nominations/Pat Loika[edit]

Originally part of the above discussion under the "rumbling subsided" heading, but I have separated it as a discussion of an error in an upcoming hook. No objection to moving to the bottom of the page. EdChem (talk) 22:38, 15 January 2017 (UTC)

I think the following hook, now in Prep3, is seriously overlinked, and Pat Loika will lose clicks to the other links: Cwmhiraeth (talk) 10:29, 15 January 2017 (UTC)
Ignoring whether clicks will be lost, I'd note that the hook is playing off a common meme and doing it awkwardly. Suggest alternative:
  • ... that according to BuzzFeed, Pat Loika is to the comic book industry what Oprah is to talk shows?
Neither BuzzFeed nor comic book need links, either. EdChem (talk) 12:31, 15 January 2017 (UTC)
I just looked at this and the hook, if still worded as any of the above needs rewriting. The article says: "Buzzfeed described his efforts in interviewing comic book creators as being "the closest thing comics has to an Oprah"." Victoriaearle (tk) 13:48, 15 January 2017 (UTC)
I'd just like to say that I've never heard of Buzzfeed so a link would be great. For instance, if Buzzfeed was a non-notable blog with a hundred viewers a month, why would I give two shits what they say about anything? Unlinking Buzzfeed makes it matter-of-fact or worse, no pedigree. We tend to have unlinked claims from people who aren't worth listening to, or at least aren't notable enough for a Wikipedia article. Unlinking this all down to just Pat Loika is a savage mistake. The Rambling Man (talk) 22:32, 15 January 2017 (UTC)
(edit conflict) I agree with Victoriaearle, this hook can easily be read as portraying Loika as a writer / illustrator of comic books. Struck my proposal. I think the hook should be changed to a direct quote, and in light of The Rambling Man's comment (which I edit conflicted with), a version with a BuzzFeed link, and in light of the above discussion, version with Loika at the start:
@RightCowLeftCoast, Vanamonde93, Yoninah, BlueMoonset, and Cwmhiraeth: As the editors involved as nominator, reviewers, commenters, and promoters, your input is requested. The hook is presently in Queue 3 so can only be adjusted by an admin if consensus is that it should be changed. I have separated this thread under a new heading level 2 heading as I think it is not really related to my above posts about the form of hooks, and is now an issue of accuracy. It is my opinion that the existing hook needs replacing. EdChem (talk) 22:38, 15 January 2017 (UTC)

And now it starts to crystallise. EdChem says Neither BuzzFeed nor comic book need links, either and I would agree with the latter, "comic book" is common term so, per OVERLINK, it can be unlinked, but "Buzzfeed" is something I'm unaware of so a link would be very helpful, per OVERLINK, not only to allow me to understand the organisation behind the name, but also to ensure that it's not just Joe Bloggs (who wouldn't have an article) making a claim about something. So once again it seems that OVERLINK is good guidance, link the terms that may need to be linked for clarity, context or explanation, don't link terms that don't enhance the readers' experience. And stop focusing on getting the all-time top-score on DYK. It's not about y'all, it's about the READER!! Time to terminate this interminable debate. The Rambling Man (talk) 22:38, 15 January 2017 (UTC)

Oh, and just to add some more fuel, we shouldn't be linking within quotations. That's another MOS thing. So Oprah shouldn't be linked in any of these examples. Mind you, apparently MOS doesn't apply to the main page according to some of our users... The Rambling Man (talk) 22:46, 15 January 2017 (UTC)
(edit conflict) I am glad we agree on "comic book." As you will see from the post above, I have proposed options with and without a link for BuzzFeed. My initial opinion was that BuzzFeed does not require a link, but I am certainly open to the view that the hook is better with one. TRM, in case you haven't noticed, I have not !voted to prefer no non-bolded links, and have argued for the reader's perspective. I have no issue with linking BuzzFeed if that is the general view. I agree that a reader might want to click on a BuzzFeed link to find out who BuzzFeed is... but readers may also be looking for information on Loika and be surprised to find that the link they followed goes to an article does not address the hook fact. TRM, I truly believe that your post are lack some good faith at times: Warning... exaggeration approaching... the DYK community is not made up entirely of page-view-driven fanatics who care nothing of readers or accurate content or the encyclopaedia and live solely for the gratification of seeing a hook posted at WP:DYKSTATS... and astonishingly, some of us even find your disinclination to recognise anything positive about DYK as problematic / irritating and this does IMO lead to your valid points about hooks and criticisms being obscured or lost. On the latter point, what do you think about the accuracy issue in the Loika hook in queue 3? I try to make sure that I take seriously the issues you raise that need to be addressed, but your approach is not making it any easier for me or others. I am glad you keep watch on the preps / queues, and your comments / changes to articles I have nominated are almost always things I look at and think "Yes, that is better" or "Yes, addressing that point strengthens the article" (whether it is something required under DYK rules or not). I value those contributions highly. Can you see a way to continue making them without needlessly ruffling so many feathers? Please? EdChem (talk) 23:23, 15 January 2017 (UTC)
Added after ec: On linking "Oprah" to Oprah Winfrey, I generally wouldn't include a link in a quotation, per MOS. In this case, however, some readers may not know who she is (esp. outside the Western English-speaking world), just as some readers may not know who BuzzFeed is, and there is no reason to believe that we could be misinterpreting who is meant. We could end the hook as to an" Oprah? or as to an Oprah" Winfrey? or as to an Oprah" (Winfrey)? or as to" Oprah Winfrey? or as to" Oprah? or without a link as to an Oprah"? The MoS provides guidance and is generally worth following, but editorial judgement still applies and we have the authority to come to a consensus decision on whether or not to provide a link in this case, and in what form to provide it. It's not a matter of "MOS doesn't apply on the main page", the question is "what form of link (if any) is most appropriate in the context for the sake of readers, in line with the integrity of the source material, and following those considerations, in a way that offers an elegant presentation?" I am somewhat surprised, TRM, that the issue of whether the hook accurately reflects the source is not a more pressing issue from your perspective than whether or how a link to the Oprah article is provided. EdChem (talk) 23:35, 15 January 2017 (UTC)
I am somewhat surprised, TRM, that the issue of whether the hook accurately reflects the source is not a more pressing issue from your perspective than whether or how a link to the Oprah article is provided. where did I say that? The Rambling Man (talk) 05:56, 16 January 2017 (UTC)
You didn't say it explicitly, but you have commented on linking issues but not accuracy in a thread where accuracy is raised, which is usually an area which you are quick to comment. I have directly asked for your view above, and of course you are free to comment or not. The hook goes onto the main page in under 3 hours. EdChem (talk) 09:19, 16 January 2017 (UTC)
The point was that this was lumped in with the "overlinked" debate. The Rambling Man (talk) 19:21, 16 January 2017 (UTC)
Actually, I separated it from the overlink discussion when I edit conflicted with your post quoting me (for the record, I did this in the edit before I saw the ec). After that, you commented in the separated section about Oprah. In any case, I am glad that a substituted hook has been used. EdChem (talk) 20:08, 16 January 2017 (UTC)
There seems to be a proclivity to re-section things here, there are one or two editors who are particularly guilty of doing so in order to focus on their own specific interests. I don't see this kind of behaviour anywhere else on Wikipedia. So sorry for missing the complete change of tack on the thread. The Rambling Man (talk) 22:27, 17 January 2017 (UTC)
I would be fine with changing it to any of the versions proposed above. Vanamonde (talk) 07:00, 16 January 2017 (UTC)
  • @Vanamonde93: This hook is in Queue3 and going live in 2+ hours. It needs an admin to change the hook. Cwmhiraeth (talk) 09:20, 16 January 2017 (UTC)
  • Okay, I have swapped in the second of the hooks proposed above. Further discussion is, however, welcome. Vanamonde (talk) 09:25, 16 January 2017 (UTC)

Hook wording[edit]

One of the hooks in prep 3 doesn't read right to me, but I don't want to tweak it myself for two reasons: I'm not a native English speaker (and it may thus be ok as is), and if I tweak hooks, I would then not be able to promote it to the queue (or would that be ok?). Anyway, here it goes:

... that 180-million-year-old dykes in the Okavango Dyke Swarm formed during the opening of the Indian Ocean?

To me, that appears to imply that the dykes were 180 million years old when they formed. Obviously, they formed 180 million years ago. If others share the concern, would somebody please tweak this? Schwede66 23:57, 10 January 2017 (UTC)

No, it's fine. Compare, "Those old men were born in 1930" -- they weren't old when they were born. I guess we'd be tempting fate to try to follow up the last hook here. EEng 00:18, 11 January 2017 (UTC)
  • ... that the Okavango Dyke Swarm formed 180 million years ago during the opening of the Indian Ocean?
Based on the quote in this thread, this is an equivalent hook. Also, "180-million-year-old dykes" is an unfortunate phrasing, which prompted a wry smile from me given the alternative meanings of the term "dyke." Thoughts? EdChem (talk) 00:20, 11 January 2017 (UTC)
I'm guessing you didn't follow the link in my post. EEng 00:25, 11 January 2017 (UTC)
  • Well I am happy with either of those hooks above, and I don't think either is wrong. I prefer "formed" instead of intruded as that is ambiguous in a different way. I would have been the magma that intruded to make the dyke. Would you like "A swarm of dykes has intruded Botswana?" instead. Graeme Bartlett (talk) 05:50, 11 January 2017 (UTC)
The possibilities are endless, but even I dare not go in that direction again. EEng 06:26, 11 January 2017 (UTC)
Yes check.svg Done adjusted hook per EdChem's suggestion. Yoninah (talk) 10:54, 11 January 2017 (UTC)
  • How about, "... a swarm of DYKs has intruded onto the Main Page?" EEng 13:40, 11 January 2017 (UTC)

Temporarily disable the reviewers template[edit]

I would suggest that we temporarily or permanently disable the reviewers template which is linked in the DYK toolbox. Given the current transclusion problems on the nomination page, reducing the number of misc templates such as that one will give a small amount of relief, and its not used often as it is.--Kevmin § 15:53, 12 January 2017 (UTC)

Oldest nominations needing DYK reviewers[edit]

The previous list was archived earlier today, so here's a list of the 38 oldest nominations that need reviewing, which includes all those through December 11. Right now the nominations page shows 303, of which 127 have been approved, but that doesn't include the 55 nominations, 11 of which have already been approved, that can't transclude because we have too many transcluded templates to show them all. Thanks to everyone who reviews these, especially the 12 that are over six weeks old and urgently need a reviewer's attention.

Over two months old:

Over six weeks old:

Other old nominations:

Please remember to cross off entries as you finish reviewing them (unless you're asking for further review), even if the review was not an approval. Many thanks! BlueMoonset (talk) 20:34, 12 January 2017 (UTC)

Potentially silly question[edit]

Can anyone enlighten me as to why this is not full protected? Or protected at all? — Preceding unsigned comment added by Vanamonde93 (talkcontribs)

EEng Englightening Vanamonde
Because the signal for the revolution to start is that the DYK updates start coming at 1-minute intervals. Thanks to you the revolution is off. Thanks a lot. EEng 10:21, 13 January 2017 (UTC)
I am guessing, that that page is used to control which prep is used. I can't say whether it's a sensitive page. Jo-Jo Eumerus (talk, contributions) 10:32, 13 January 2017 (UTC)
Oops, I got it mixed up with the time between updates. So much for the revolution and enlightenment. EEng 10:47, 13 January 2017 (UTC)
EEng, no worries: I still chuckled. @Jo-Jo Eumerus: Yes, I believe it controls which prep is displayed at the top, or something like that. If an IP with malicious intent finds it, I don't imagine it will be more than a bit of a pain to fix, but the fact remains that none but administrators have any reason to edit it, so we might as well be safe, I think...Vanamonde (talk) 10:55, 13 January 2017 (UTC)
Actually, Vanamonde, I've edited NextPrep 170 times over the years and I'm not an administrator. The fact remains that administrators forget—and not infrequently—to update this page when moving the next prep to the next unfilled queue, so it's been very useful to have it adjustable by non-admins so the displaying of preps is in the right order. Unlike the "next queue" page, which is used by DYKUpdateBot, this doesn't control any automated process, so if there is a malicious IP about, hopefully the admin promoting the next prep will take a look before doing so and realize that something's not quite right. We've been quite good about synchronizing the prep and queue numbers since we increased the number of preps to match the number of queues, so if the next prep number doesn't match the next empty queue number, it should be a red flag. BlueMoonset (talk) 16:22, 13 January 2017 (UTC)
I see, thank you. It still does seem to me, though, that semi-protection might be a good idea: but if it's not controlling the automated process, then there isn't a very strong policy reason for that, so I'll just let it be. Cheers, Vanamonde (talk) 16:38, 13 January 2017 (UTC)
No objection to protecting against IPs and non-confirmed users to avoid potential disruption, but no need to protect against the rest of us. BlueMoonset (talk) 16:50, 13 January 2017 (UTC)
  • Aight. Indef semi applied. Going offline for many hours now, if somebody sees this and thinks of a good reason to unprotect it, don't bother to wait for me to come back. Vanamonde (talk) 17:34, 13 January 2017 (UTC)
  • Good! The admin instructions for promotion to the queue have quite a number of steps and it's easy to miss something, like resetting the next hook counter. Thanks, BlueMoonset, for keeping an eye on it. Schwede66 19:36, 13 January 2017 (UTC)

Question about special occasion hooks[edit]

I was thinking that Template:Did you know nominations/Cerro de las Campanas can be a special occasion hook as Raymie pointed out on the nomination page. Its 150th anniversary comes up in May 15, 2017. I reviewed the nomination and it is okay. How would we be able to save this hook for May 15? Thank you in advance. epicgenius (talk) 02:54, 14 January 2017 (UTC)

epicgenius, I'm afraid we wouldn't be able to save it that far in the future. As it says in the special occasion section, Articles nominated for a special occasion should be nominated ... between five days and six weeks before the occasion, to give reviewers time to check the nomination. So six weeks is the outside: hooks nominated today will need to run by the end of February. Sometimes this ends up stretching to six weeks after approval, or a few extra days beyond that, but four months from now is simply not possible. BlueMoonset (talk) 05:38, 14 January 2017 (UTC)

Don't cry for me, Argentina[edit]

Remembering a recent discussion here about living persons suspected of crimes, is this hook (currently in Prep3) permissible, mentioning as it does a suspect in an alleged incident for which nobody has been tried or convicted? Cwmhiraeth (talk) 07:29, 14 January 2017 (UTC)

  • Certainly this is the sort of thing we have to be careful about, though I'm not sure what the bottom line is in this case. However, let me suggest that a better hook might be based on the unexpected fact that, "His lawyer Fernanda Herrera, who is also a cumbia singer..." EEng 13:45, 14 January 2017 (UTC)
  • ITN routinely has items about fresh crimes such as the current 2017 Istanbul nightclub attack. And note that we have several similar items in the queue – Corruption in Angola, Ecuador and Italy. My view is that the case in question has so many bizarre features that we should stick with it: nuns on the run; claims of insanity; secret vaults and buried treasure; the Pope and the number of the beast. I suggest that it'd be best for the hook to focus on these farcial features while omitting the names of any of the people involved. Andrew D. (talk) 15:00, 14 January 2017 (UTC)
Is farcial when people fart? That would be at home in this article too. EEng 16:55, 14 January 2017 (UTC)
  • Hook has been moved to Prep 5 to give more time for consideration here. BlueMoonset (talk) 08:08, 15 January 2017 (UTC)
I suggest this alternative hook: Cwmhiraeth (talk) 10:20, 15 January 2017 (UTC)
  • ... that a man was detained while allegedly trying to hide currency worth millions of dollars inside a convent?
  • "Man" sounds too vague, in my opinion. How about "politician" or "argentine politician" (to provide some geographical context)? TomasBat 21:18, 15 January 2017 (UTC)
  • ... that a politician was detained while allegedly trying to hide currency worth millions of dollars inside a convent?
  • ... that an argentine politician was detained while allegedly trying to hide currency worth millions of dollars inside a convent?
  • I've just moved the hook out again to the newly emptied Prep 3 to give us even more time, but also because a hook needed to be moved out of Prep 5 to make room for a special occasion hook still awaiting promotion. (I don't have time right now to check that special occasion hook and its three articles, so if someone could do that and promote it to Prep 5, that would be great. Since Cwmhiraeth approved it, someone else will be needed to promote it.) BlueMoonset (talk) 17:14, 15 January 2017 (UTC)
  • Yes check.svg Done Triple hook promoted to Prep 5. Yoninah (talk) 17:49, 15 January 2017 (UTC)
  • I have changed the Argentine politician hook in Prep3. Cwmhiraeth (talk) 13:49, 17 January 2017 (UTC)

Template:Did you know nominations/Pioneer Cabin Tree[edit]

An unintended consequence of WP:ITN and WP:DYK. Perverse result that could not have been within the intendment of the rule. 7&6=thirteen () 12:59, 15 January 2017 (UTC)

7&6=thirteen, this appeared as a non-bolded link under Recent Deaths. It's fine - articles that were a non-bold link under Recent Deaths are still eligible for DYK. Move forward with the nomination. — Maile (talk) 13:58, 15 January 2017 (UTC)
Thank you. Cheers. 7&6=thirteen () 14:05, 15 January 2017 (UTC)

Queue 2 - DRDO Smart anti-airfield missile[edit]

DRDO Smart Anti-Airfield Weapon is India's first fully indigenous anti-airfield weapon?

I'm sure it's not intentional but there's far too much repetition in this hook. The Rambling Man (talk) 17:07, 15 January 2017 (UTC)

So fix it. --Kevmin § 21:05, 15 January 2017 (UTC)
Are you sure you know what you're doing here? I can't fix a queue, it needs an admin to do that. But once again you seem to be unclear on how things work. Never mind, perhaps you'll leave this here for someone who can help out. The Rambling Man (talk) 22:51, 15 January 2017 (UTC)
Why do you make everything a personal attack on someone?--Kevmin § 22:56, 15 January 2017 (UTC)
Why did you bother responding? By the way, a statement of lack of competence about a particular user isn't a personal attack. It's a statement of fact. The Rambling Man (talk) 23:01, 15 January 2017 (UTC)
No, it's not a fact, it's an assertion on your part about the person that you are communicating with, and your tone makes it personal. Something that your do regularly in a manner and tone indicating scorn and disdain for others, and fully ignoring wp:CIVIL.--Kevmin § 23:09, 15 January 2017 (UTC)
Get your facts straight in future, and we can avoid this unnecessary communication. The Rambling Man (talk) 07:07, 16 January 2017 (UTC)

Please make a specific wording suggestion and I'll swap it over. Schwede66 23:31, 15 January 2017 (UTC)

Perhaps:
This avoids the repetition noted by The Rambling Man though it adds the fact of the testing year (which has about 5 sources supporting it in the article). It is an easter egg link, but not a surprising one. @Schwede66: Thoughts? This queue goes live in 7 minutes, if I am reading correctly, so there is some urgency. Thanks. EdChem (talk) 23:53, 15 January 2017 (UTC)
Yes check.svg Done Yes, you interpreted the time right, but we can still edit it when it's on the homepage already. Schwede66 00:13, 16 January 2017 (UTC)
Thanks, Schwede66, I realised that the main page can still be edited by an admin. However, if the edit is done in the queue then the bot will report the updated hook to the article talk(s) and to user talk(s), rather than needing manual updates. Further, I thought we could do without another WP:ERRORS report.  :) I'm glad you got to it quickly. EdChem (talk) 00:30, 16 January 2017 (UTC)
Thanks both, a vast improvement. The Rambling Man (talk) 10:30, 16 January 2017 (UTC)

New DYK page trial, volunteers needed[edit]

We're really close to fixing the transclusion problem, but need your help! A bot's being developed that will move approved hooks from the nom page to a new approved page, and the bot needs to be trialed. If you would like to volunteer to try the bot-updated approved page for a week, please let me know here or at the Bot Request for Approval. If you're a review checker or a prep builder, it would be especially helpful for you to try it out and give feedback on functionality or report errors. All you need to do is add User:Wugapodes/DYKTest and User:Wugapodes/DYKTest/Approved to your watch list and use that for a week rather than the usual nom page. Super simple. Thanks. Wugapodes [thɔk] [ˈkan.ˌʧɻɪbz] 05:35, 16 January 2017 (UTC)

@Wugapodes: watchlisted, many thanks. Quick question: where would you like feedback to be posted? Vanamonde (talk) 07:05, 16 January 2017 (UTC)
Feedback is probably best given at Wikipedia:Bots/Requests for approval/WugBot 2. If it's urgent, posting on my talk page as well would be appreciated (and probably get a faster response). Wugapodes [thɔk] [ˈkan.ˌʧɻɪbz] 08:55, 16 January 2017 (UTC)
I will be happy to trial it. Cwmhiraeth (talk) 09:26, 16 January 2017 (UTC)
Will watch, --Gerda Arendt (talk) 14:29, 16 January 2017 (UTC)
  • This sounds sensible and I'm watching too. I suppose that when this is done, there will be less need to repeat lists of outstanding reviews here because the nominations on the unapproved page will all be waiting attention and is already in date order. Andrew D. (talk) 14:36, 16 January 2017 (UTC)

Queue 3 - Petrodollar recycling[edit]

that a prominent example of petrodollar recycling was the 90% purchase of New York's Chrysler Building (pictured) by the Abu Dhabi Investment Council?

This hook is written like a school child saying "something I did at the weekend was...", it would be better switched around to " that the 90% purchase of New York's Chrysler Building (pictured) by the Abu Dhabi Investment Council was a prominent example of petrodollar recycling?" The Rambling Man (talk) 07:09, 16 January 2017 (UTC)

There's some value to having the highlighted link appear first, but your rewording wouldn't bother me.
On the other hand, I'm quite a bit more bothered by the last-minute image swap. As i just posted at User talk:David Levy#Please self-revert your image swap of Chrysler Building in DYK queue:

David, I'm sure you meant well, but the previous image has a much better view of the iconic crown, a much better contrast between building & sky, and showcases a nationally recognized photographer. The substitute image has a badly foreshortened perspective and a badly monochrome cast, which combine to make this landmark building almost unrecognizable at thumbnail size.

The previous image was painstakingly selected out of hundreds available, and approved by the GA reviewer, the DYK nomination reviewer, the DYK prep reviewer, the DYK queue promoter, and the US Library of Congress. Overriding everyone else's judgment unilaterally, at the 11th hour on an admin-only page, hardly seems a good example of either WP collegiality or photographic discernment. I hope you'll please reconsider and self-revert.

I'd happily accept the reworded hook if an admin could revert back to the previously promoted image. Thanks very much for your consideration. —Patrug (talk) 07:57, 16 January 2017 (UTC)
I'm happy to adopt both suggestions. I shall try and go through the queue, check everything out, and then change hook wording around and revert to the previous image (unless David is still online and can give his reason for the image change, but to me, it does look rather inferior in comparison). Schwede66 08:28, 16 January 2017 (UTC)

Queue 3 - Lok Sewak Sangh[edit]

It's minor, but in Queue 3, I think we are missing two hyphens (300 km and 190 mi) in the Lok Sewak Sangh hook. EricEnfermero (Talk) 08:40, 16 January 2017 (UTC)

It's worse than that, unless I'm reading it wrong, the source says "more than 300 miles". The Rambling Man (talk) 08:43, 16 January 2017 (UTC)
Have moved the hook to prep5; that should give enough time to look into the distance issue. And no, hyphens don't belong there. Schwede66 09:05, 16 January 2017 (UTC)
I think the hyphens do belong there, per the article and per "adj=on", i.e. 300-mile (480 km)... The Rambling Man (talk) 09:09, 16 January 2017 (UTC)
I'd try to avoid these hyphens, which would make reading even more complicated than it already is. We have in prep
* ... that the Lok Sewak Sangh organized a 300 km (190 mi) protest march in 1955 to demand inclusion of Bengali-speaking areas of Bihar into West Bengal?
My suggestion:
* ... that the Lok Sewak Sangh organized in 1955 a protest march of more than 300 km (190 mi) to demand inclusion of Bengali-speaking areas of Bihar into West Bengal? - I dont want to change the prep because I was the reviewer. --Gerda Arendt (talk) 09:24, 16 January 2017 (UTC)
  • FYI hyphens are never used with symbols such as km and mi, only with the full-word forms. EEng 09:34, 16 January 2017 (UTC)
Perhaps a better wording:
  • ... that a 1955 protest march organized by the Lok Sewak Sangh to demand inclusion of Bengali-speaking areas of Bihar into West Bengal travelled more than [DISTANCE]?
I've put it as "[DISTANCE]" in recognition of TRM's point that the source needs re-checking. EdChem (talk) 10:56, 16 January 2017 (UTC)
Well, I took another look at the source and it unquestionably says the distance was more than "300 miles", so I've corrected the hook now it's back in a Prep area. Wouldn't object to any rewording, they're all much of a muchness. The Rambling Man (talk) 10:59, 16 January 2017 (UTC)
I concur, definitely says more than 300 miles. Striking above and providing alternatives:
  • ... that a 1955 protest march organized by the Lok Sewak Sangh to demand inclusion of Bengali-speaking areas of Bihar into West Bengal travelled more than 300 mi (480 km) over 16 days?
  • ... that in a 1955 protest, nearly 1000 protesters from the Lok Sewak Sangh marched more than 300 mi (480 km) over 16 days to demand inclusion of Bengali-speaking areas of Bihar into West Bengal?
I think the 16 days is interesting too... @The Rambling Man, Gerda Arendt, Schwede66, Soman, and Cwmhiraeth: as commenters in this thread for comment or with involvement in the nomination, for comment. EdChem (talk) 11:09, 16 January 2017 (UTC)
Works fine for me, a touch-wordy for a hooky-hook, but interesting-enough. The Rambling Man (talk) 11:15, 16 January 2017 (UTC)

@Schwede66, Vanamonde93, and Casliber: This is in Queue 5 and goes onto the main page in 18 minutes. The error over the distance is fixed but there is still a rogue hyphen and one of the better wordings above would be nice. Paging any admin around? Thanks. EdChem (talk) 11:43, 17 January 2017 (UTC)

It looks fine as it is, frankly. The hyphen is correct, and changing to have (for example) "protest... protesters" would be a retrograde step. BencherliteTalk 12:34, 17 January 2017 (UTC)
  • My apologies, I was working...I have made the change. However, upon reading the article, the hook did not seem entirely grammatical to me. I have therefore, admittedly unilaterally, put the following on the main page. Further discussion would, I think, be welcome. Vanamonde (talk) 15:01, 17 January 2017 (UTC)
  • ... that in a 1955 protest, nearly 1000 protesters from the Lok Sewak Sangh marched more than 300 mi (480 km) over 16 days to demand inclusion of Bengali-speaking areas of Bihar into West Bengal?
  • ... that in a 1955 protest, nearly 1000 protesters from the Lok Sewak Sangh marched more than 300 mi (480 km) over 16 days to demand the incorporation of Bengali-speaking areas of Bihar into West Bengal?
  • Dammit. The unstruck version is the one I uploaded. I cannot, it appears, copy-paste accurately sometimes. Vanamonde (talk) 15:03, 17 January 2017 (UTC)

Prep 6[edit]

Why is there an Easter egg link to the magazine? Isn't the point of the hook to hook the reader to the publisher's article?

— Preceding unsigned comment added by Yoninah (talkcontribs)

The hook is fully compliant with current DYK norms. I'm not sure what you're trying to say. You can change it if you like, right? The Rambling Man (talk) 22:41, 16 January 2017 (UTC)
Yes. Thanks for your support. Yoninah (talk) 22:58, 16 January 2017 (UTC)
Since the magazine is one likely to be unfamiliar to many readers, a description (provided it is supported by an in-line citation in the article) is appropriate for a hook. If I clicked on a link "most widely read magazines", I would expect to go to an article on a magazine, which is what Die Gartenlaube is, is it not? EdChem (talk) 00:23, 17 January 2017 (UTC)
I notice that Yoninah has edited the prep to remove the link altogether, with edit summary "restore hookiness." Yes, this leaves only a bolded link, but I am concerned that a reader may want to know which magazine it was directly, and so a convenience link should be provided. Does anyone else have a view? EdChem (talk) 00:30, 17 January 2017 (UTC)
If you want to know what magazine it is, click on the bolded link! We just had this whole discussion above. We really need to be asking ourselves what is the point of the hook? Adding a link to the magazine takes away all the hookiness. Yoninah (talk) 00:35, 17 January 2017 (UTC)
This is a very good example of where a link would be very helpful to our reader so it should stay linked. By no means does it take away any of the hookiness of the hook. That assertion is simply nonsense. The Rambling Man (talk) 06:41, 17 January 2017 (UTC)
This is probably going to be an unpopular opinion, but I am not very concerned about easter eggs in hooks. Hookiness requires a little "funny", and easter eggs are one way of doing that. Jo-Jo Eumerus (talk, contributions) 09:38, 17 January 2017 (UTC)
Agreed. I for one don't want to see the no-fun police taking over DYK. The C of E God Save the Queen! (talk) 09:48, 17 January 2017 (UTC)
If you want to get really pissed about it, look at Prep 6: "... that an Australian cricketer took unpaid leave from his teaching job to play in the 2012 Champions League Twenty20?" WTF? When did this project suddenly become Guess Who? The Rambling Man (talk) 22:50, 16 January 2017 (UTC)
I agree with TRM, and have edited the hook to include the name of the cricketer. EdChem (talk) 00:30, 17 January 2017 (UTC)
Hi guys, I am the proposer for the cricketer hook mentioned above. I originally had Ian Moran's name bolded but I thought it would improve the hookiness if the name wasn't mentioned. After all, no one has even heard of Moran. I hadn't even heard of him before I wrote the article and I'm fan of the Sydney Sixers, the team he played for. Anyway, I am happy for the change to occur if everyone feels that I went to far with this easter egg. Cheers – Ianblair23 (talk) 02:29, 17 January 2017 (UTC)
No worries, it's just bad timing with the project at odds with itself over whether or not to easter egg links, whether or not to use auxiliary links. It just touched a nerve I suppose. Nice article by the way. The Rambling Man (talk) 19:22, 17 January 2017 (UTC)

QPQ question[edit]

If one encounters a situation where the person who nominated an article is exempt from the QPQ requirements, but the person who did the work creating/expanding/getting the article to GA is not exempt because they have 11 DYK credits, should one insist on a QPQ being done? For the record this is about Template:Did you know nominations/Why (Taeyeon song). ~ ONUnicorn(Talk|Contribs)problem solving 18:24, 17 January 2017 (UTC)

QPQ is the nominator's responsibility. It's got nothing to do with others who have been involved in creating or expanding an article. Schwede66 18:30, 17 January 2017 (UTC)
So someone with a dozen or so DYKs can get a relative newbie to nominate their article and get out of having to do a QPQ? (Not saying that's what happened here - but it defeats the spirit of the rule.) ~ ONUnicorn(Talk|Contribs)problem solving 19:15, 17 January 2017 (UTC)
Yes. As long as we get our pound of flesh from someone. Most nominators are the creator/expander, so the question doesn't come up that often, but QPQ is the nominator's responsibility. EEng 19:39, 17 January 2017 (UTC)
But EEng#s, we're not getting our pound of flesh ... ~ ONUnicorn(Talk|Contribs)problem solving 20:29, 17 January 2017 (UTC)
Yes we are, just every new editor is given five free pounds as a welcome gift, which he can use as he will. Look, it's just the way it works. Look, if you're worried about the accumulation of unreviewed stuff then revive my old proposal that everyone with 15+ credits (or whatever) needs to do two reviews for every new nomination (unless there's nothing awaiting review at the moment). That would dry up the backlog real fast. The beauty of that is that those with lots of noms under their belt are experienced reviewers, and so it's less of a burden for them. EEng 20:40, 17 January 2017 (UTC)
ONUnicorn, if you can show that an experienced DYK participant recruited a new nominator in order to avoid the QPQ requirement, I certainly think we can require the DYK regular to do a QPQ for trying to make an end run around the requirements. In fact, I've done that in the past (it was really quite blatant, as I recall, with the request made on the nominator's talk page by the creator, and this was when non-self-nominations were not subject to the QPQ requirement; we've since closed that loophole). However, if a newcomer finds an unsubmitted GA or new article and preps it for DYK themselves, then more power to them. If this happens, there's nothing wrong with asking the creator to volunteer to review another nomination, just as new nominators can start reviewing prior to completing their five freebies (and some do); the keyword here is "volunteer". BlueMoonset (talk) 01:07, 18 January 2017 (UTC)