Jump to content

Wikipedia talk:Did you know: Difference between revisions

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia
Content deleted Content added
Line 260: Line 260:
== RfC on proposal to require a second QPQ from "senior" DYK editors (those with 20+ DYK credits) when there's a backlog of unreviewed nominations ==
== RfC on proposal to require a second QPQ from "senior" DYK editors (those with 20+ DYK credits) when there's a backlog of unreviewed nominations ==
<!-- [[User:DoNotArchiveUntil]] 19:01, 8 September 2021 (UTC) -->{{User:ClueBot III/DoNotArchiveUntil|1631127678}}
<!-- [[User:DoNotArchiveUntil]] 19:01, 8 September 2021 (UTC) -->{{User:ClueBot III/DoNotArchiveUntil|1631127678}}
{{rfc|proj|rfcid=F66982D}}
'''''Did you know ...''' that articles nominated for [[WP:Did You Know]] often wait months for approval?''
'''''Did you know ...''' that articles nominated for [[WP:Did You Know]] often wait months for approval?''


Line 367: Line 366:


===Take 2 -- Everyone please vote again on revised proposal===
===Take 2 -- Everyone please vote again on revised proposal===
{{rfc|proj|rfcid=F66982D}}
The [[WP:Did you know#Review requirement|rules for DYK's review requirement]] currently provide that:
:If, at the time a nomination is promoted to the main page, its nominator has fewer than five [[Wikipedia:List of Wikipedians by number of DYKs|DYK credits]] (whether or not self-nominated) then the nomination is exempt from QPQ.
Should this be replaced with more comprehensive criteria as detailed below? [[User:EEng#s|<b style="color:red;">E</b>]][[User talk:EEng#s|<b style="color:blue;">Eng</b>]] 05:22, 9 August 2021 (UTC)


OK, you nitpicking rabble, try this:
OK, you nitpicking rabble, try this:

Revision as of 19:13, 10 August 2021


Did you know?
Introduction and rules
IntroductionWP:DYK
General discussionWT:DYK
GuidelinesWP:DYKCRIT
Reviewer instructionsWP:DYKRI
Nominations
Nominate an articleWP:DYKCNN
Awaiting approvalWP:DYKN
ApprovedWP:DYKNA
April 1 hooksWP:DYKAPRIL
Holding areaWP:SOHA
Preparation
Preps and queuesT:DYK/Q
Prepper instructionsWP:DYKPBI
Admin instructionsWP:DYKAI
Main Page errorsWP:ERRORS
History
StatisticsWP:DYKSTATS
Archived setsWP:DYKA
Just for fun
Monthly wrapsWP:DYKW
AwardsWP:DYKAWARDS
UserboxesWP:DYKUBX
Hall of FameWP:DYK/HoF
List of users ...
... by nominationsWP:DYKNC
... by promotionsWP:DYKPC
Administrative
Scripts and botsWP:DYKSB
On the Main Page
Main Page errorsWP:ERRORS
To ping the DYK admins{{DYK admins}}

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

EEng's alternative for addressing/preventing the unapproved backlog

The alternative is: Change the rules as follows:

  • (a) First five noms, no QPQ required;
  • (b) 6th to Nth nom, one QPQ required;
  • (c) After N noms, two QPQs required (assuming there are "enough" noms needing review at the time the new nom is made; some details needed here about how exactly that works, but it's not rocket science).

N might be in the range 10–20; I actually think 10 is about right because I'm guessing few people even get to 10, especially compared to the large number of no-QPQ-required noms from newbies. But we can get stats on that and adjust N accordingly. Elsewhere people have been wringing their hands about how the system will collapse if the unreviewed queue becomes completely empty (we should have such problems!) but I assure you we can deal with that situation. The above change is what's important -- do we want a permanent mechanism for avoiding an unreviewed backlog, or not?
EEng 13:46, 6 July 2021 (UTC)[reply]

  • Oppose backlog drive proposal. Drives are bandaid solutions; they're fun barnstar printers, but you end up in the same place quickly (see WP:GASTATS right after a bolded month) but Support EEng's solution, even though everyone hates it for some reason ("EEng's [...] even though everyone hates it for some reason" is a fruitful game of madlibs). I might prefer a higher credit count than he does, though -- ten makes for a pretty narrow single-QPQ band. "20 credits, 2 reviews" has a good pattern to it and allows for an adjustment period, as well as solving the "we have multiple bottlenecks, short-term reductions of the DYKN backlog are in practice mostly increments of the DYKNA backlog" problem (by resulting in a more moderate but long-term decrease). Vaticidalprophet 08:12, 6 July 2021 (UTC)[reply]
    Hey, watch out who you're calling fruitful! EEng 13:29, 7 July 2021 (UTC)[reply]
  • I like EEng's proposal, too. I like that it requires more qpqs from frequent nominators, and I like that it means more reviews are being done by experienced people. I like that it's a permanent improvement. —valereee (talk) 12:44, 6 July 2021 (UTC)[reply]
    • It's a good one. Re. what number n is, I don't think the important part is how many people have n DYK credits as an absolute, I think it's how many nominations are made by those people as a percentage. Hell, I just barely scrape 20+ and I had multiple nominations open for over a week until a couple hours ago; plenty of people far higher in that range are far more prolific. My wildly unscientific impression is that the 20+ range makes up a much more disproportionate share of the backlog compared to 10-19. (I also find that when I build preps, I promote a lot of the same names.) Vaticidalprophet 14:43, 6 July 2021 (UTC)[reply]
  • Support EEng's one. Woohooo (you wanted a rationale for my support? Can't have one. Nur...Ok, don't pout, you can have a perfunctory one: per Valereee ) Belle (talk) 00:52, 7 July 2021 (UTC)[reply]
  • Support 2 reviews at 20 DYKs to fix our pipeline problem. If you wanted to be more extreme, 3 reviews at 50 DYKs might be something to think about --Guerillero Parlez Moi 01:11, 7 July 2021 (UTC)[reply]
  • Oppose as per my views at Wikipedia talk:Did you know/Archive 179#Vote on formal proposal for a DYK backlog drive which I don't want to just regurgitate here. I prefer an honour system (do more reviews if you have time) rather than mandatory increasing of workload. But if people do support this, I will (reluctantly) comply with it. Joseph2302 (talk) 10:16, 7 July 2021 (UTC)[reply]
    It's funny, I had a similar conversation the other day with the IRS: I told them I preferred the honor system and would pay them if I had time. They said they would take the idea under advisement. EEng 12:01, 7 July 2021 (UTC)[reply]
  • Comment Based on some quick investigation and discussion at User_talk:BlueMoonset#Distribution_of_editors_according_to_#_of_DYKs, it seems like 20 might be a good "trigger point" (at which the 2-review requirement kicks in) to start with; once the backlog has been eaten down, a higher trigger point would be enough to keep it down for the long term. But these numbers can be determined more carefully once the idea has been approved in concept. EEng 12:01, 7 July 2021 (UTC)[reply]
  • Support, two reviews at 20 DYKs seems good to me. Cwmhiraeth (talk) 12:48, 7 July 2021 (UTC)[reply]
  • Support EEng's solution (a lifetime first, I think). N could be 25 - as said above, a high % of noms are by those (like me) who have 100+ DYKs, so using a much higher number than 25 would probably have a good effect. Johnbod (talk) 12:56, 7 July 2021 (UTC)[reply]
  • General support although we might want to clarify when one QPQ is enough (say, when there are fewer than 25 open nominations). Perhaps anyone eligible for listing on Wikipedia:List of Wikipedians by number of DYKs should do extra QPQ reviews (although I'd prefer a number larger than 27 as the cutoff out of laziness). —Kusma (talk) 13:04, 7 July 2021 (UTC)[reply]
    Yes, since if this works there will be times there is no (significant) backlog, there will need to be some trigger that suspends the double-review requirement until the backlog grows again. Now, you might imagine people will game that -- delay making a nom for a while to avoid having to do a double review. But we aren't cynical enough to imagine any of our esteemed fellow editors would do such a thing, do we? EEng 13:29, 7 July 2021 (UTC)[reply]
    If we're going to use Wikipedia:List of Wikipedians by number of DYKs as a measure, we'd need to update it reasonably frequently. It looks to be very out-of-date (it says I have 113 DYKs, when I just passed 200). Might be worth trying to integrate it with the QPQ check tool that gets the up-to-date information from it. Otherwise, we'd be missing people who reach the threshold. Joseph2302 (talk) 13:08, 7 July 2021 (UTC)[reply]
    Obviously we'll find some way to have reliable counts. No doubt our tireless script/bot wizards can cook something up -- the same machinery that gives you your little talk-page congratulation can increment a tally. EEng 13:19, 7 July 2021 (UTC)[reply]
    For some insane reason WBDYK is manually updated, so something that switches it to auto-updating sounds overdue -- it's not like we don't have the machinery for it (this has been around for ages, as you can tell by the fact it's maintained(?) by someone who's been indeffed for the better part of a decade). Would recommend doing something about WP:DYKSTATS, while we're at it, as it hasn't been updated since April. Vaticidalprophet 13:32, 7 July 2021 (UTC)[reply]
    Or we can just use the ever-so-popular honour system; the only people who don't know their personal DYK score have less than N because they haven't yet been fully initiated into the cult....starts chanting Temple of Doom style:QPQ did you know, QPQ did you know, QPQ did you know... Belle (talk) 13:43, 7 July 2021 (UTC)[reply]
    I thought about that, and we might use that at first, but I think in the end something automated is appropriate (even if just to have reliable statistics as a general principle). We've always used a manual/honor system for the "first 5", but that only requires counting on the fingers of one hand. I personally have only a vague idea how many DYK credits I have -- I'm sure it's at least 20, but beyond that I really don't know. EEng 13:59, 7 July 2021 (UTC)[reply]
    30 according to the tool. Vaticidalprophet 14:00, 7 July 2021 (UTC)[reply]
    So EEng can still count them on one hand [pokes out tongue] Belle (talk)
    Just you wait! EEng 15:19, 7 July 2021 (UTC)[reply]
    Silly me. I didn't even know there was such a tool. Parties wanting to know more about Dr. Young's Ideal Rectal Dilators ("forcibly withdrawn after officials clamped down on them") and other such things might want to check out User:EEng#dyk. EEng 15:19, 7 July 2021 (UTC)[reply]
    DYK credits were done manually back in the Dark Ages, so the automated list is incomplete (compare [1] and User:Kusma/DYK). —Kusma (talk) 13:51, 7 July 2021 (UTC)[reply]
    Can anyone think how we'd go too far wrong if we trawled each user's contribution history for page creations of the form Template:Did you know nominations/xxxxxxx? EEng 13:59, 7 July 2021 (UTC)[reply]
    You'd miss stuff like this successful nom from 2006 (that was before the introduction of QPQ). —Kusma (talk) 14:02, 7 July 2021 (UTC)[reply]
    (edit conflict) Those ages seem pretty dark indeed. If someone who's <20 by the autostats is >20 by pre-2007 manual stats, I'm happy to count them as <20 until they get over the hump -- they don't have the recent DYK experience-slash-backlog-flooding that's being picked up on here. Vaticidalprophet 14:00, 7 July 2021 (UTC)[reply]
    If someone codes this somewhere, please anticipate a possible namespace change for the DYK nom system. —Kusma (talk) 16:19, 7 July 2021 (UTC)[reply]
  • Support. I've struck out my own proposal and offering general support for this and N=20. Desertarun (talk) 14:03, 7 July 2021 (UTC)[reply]
    Wanna give a shout-out here to Desertarun for his work on the backlog proposal. Since the double-QPQ idea is really for preventing a backlog, and would take a long time to eat down the huge existing backlog formed over many years, there may still be a place for a one-time drive after all. EEng 15:10, 7 July 2021 (UTC)[reply]
    I think fixing the structural in built problem with the QPQ system is the most important thing we can do for DYK. Whatever achieves that goal is ok with me. I could set up a more casual backlog drive if this goes through. Desertarun (talk) 15:40, 7 July 2021 (UTC)[reply]
    I have some old cattle prods in the closet. EEng 15:42, 7 July 2021 (UTC)[reply]
  • Oppose as written. Only because monitoring DYK history of other editors is too big of a headache. See my modified proposal of EEng's proposal below for a simpler but similar solution that I think will be easier to implement. Best.4meter4 (talk) 17:07, 7 July 2021 (UTC)[reply]
    As noted elsewhere there's a simple tool that gives the info needed. I predict there will be relatively few people above the key number (relative to all DYKers, including the many who make just one nom, or a few noms, then disappear), and they'll know who they are. We don't need some kind of rigid gatekeeping or enforcement. EEng 04:45, 8 July 2021 (UTC)[reply]
  • Oppose, As someone that has been a low key contributor of hooks every few months over the past 10 years this is now penalizing me for contributing, since I fall into the +400 range of hooks am I going to be told I need to review 3 4 or 5 nominations for the single one I contribute?--20:52, 7 July 2021 (UTC) — Preceding unsigned comment added by Kevmin (talkcontribs)
    No, just two, and only when there's a backlog (which it looks like won't be often, after this one's cleared). EEng 22:21, 7 July 2021 (UTC)[reply]
  • Support: I'd prefer to run a backlog drive, but I think this is a great second-choice. We already have the tools to determine how many DYKs someone has, it just involves counting. Some expressed concerns that longterm editors have several unused DYKs from years ago, negating this effort, but I don't think that's every longterm DYK nominator, and I'm OK with rewarding editors who reviewed a lot in the past. I'm a little worried about conveying the "trigger" of 2 QPQ to those who don't frequent this talk page: the current 60/120 trigger for preps and queues happen every week or two. If the triggers for 2QPQ are similar, it can cause much greater confusion. I hope the triggers will be much wider to avoid this. Z1720 (talk) 02:06, 8 July 2021 (UTC)[reply]
    We lack hard statistics, but I believe the cycle time for this new scheme will be more like 9 months 1QPQ, 3 months 2QPQ, repeat (after the current backlog is eaten down, which will take quite some time). 02:34, 8 July 2021 (UTC)
    My impression was that the triggers would be the number of articles waiting for approval. The above comment gives me the impression that it will be set times for the triggers to take affect (Like January-Sept: 1QPQ, Oct-Dec: 2QPQ). I prefer article triggers over set times every year. Z1720 (talk) 15:12, 8 July 2021 (UTC)[reply]
    Of course it will be based on the # of noms awaiting review; a fixed schedule would make no sense. I was simply envisioning how rapid (or not rapid) the cycling back and forth might be -- very slow, on the order of months, because it appears the backlog grows very slowly. EEng 15:26, 8 July 2021 (UTC)[reply]
    Sorry, that was my misinterpretation of your comment, EEng. I hope the triggers are wide enough to cause the 9 month/3 month cycle you describe above. Z1720 (talk) 16:31, 8 July 2021 (UTC)[reply]
Let's escalate!
  • Anyone else? EEng 19:41, 11 July 2021 (UTC)[reply]
  • <sound of crickets> EEng 17:02, 18 July 2021 (UTC)[reply]
  • *steps on crickets* Support 2 for 20 per Belle. (And also support further escalations such as 3 for 50+, 4 for 100+, and so on. I support further escalation in general.) Waiving it during low-backlog periods also sounds reasonable. Levivich 20:40, 18 July 2021 (UTC)[reply]
  • Oppose because I'm an ass. Actually, Support. In seriousness, this seems like it has the potential to get more editors at low output levels, while not letting it get dominated by a few regulars. Personally, I find most of the DYK hooks boring. But with a wider selection of editors (and a faster rate of review by the regulars, who are going to be better able to do that well).
    And if it just crashes and burns and ruins everything, we can blame Eeng for it, let some grumpy admin block him and then march on ANI with torches and pitchforks and demand an unblock and a grovelling apology. ᛗᛁᛟᛚᚾᛁᚱPants Tell me all about it. 05:07, 19 July 2021 (UTC)[reply]
  • Strong oppose per WP:NOTBURO. 1 QPQ is fine. IronGargoyle (talk) 03:41, 22 July 2021 (UTC)[reply]
  • Oppose We don't have a problem as bad as GA (Who don't even have required QPQs by the way). Most of our backlog (which is fairly minor) comes from reviews that end up with walls of text abandoned by the original reviewer that scares people off having a look. I think this proposal is a sledgehammer to crack a nut, there's no need for it. The C of E God Save the Queen! (talk) 08:50, 22 July 2021 (UTC)[reply]
    We don't have a backlog as bad as GA because we have QPQ. So we already have the hammer, and the proposal is to finally crack the nut by wielding the hammer just a bit harder. Other techniques -- yelling at the nut, pleading with the nut, and praying that the nut will somehow crack open on its own -- haven't worked. EEng 12:32, 22 July 2021 (UTC)[reply]
  • Support. Even though it affects me and I don't particularly enjoy doing QPQs. Edwardx (talk) 13:24, 6 August 2021 (UTC)[reply]

Getting specific

  • Comment: since we do major changes like this after an RfC, can we have one with a firm proposal that uses actual numbers, such as 20 or 25 DYK credits for the point where the second QPQ can be required, and equally as important, the number of nominated but unapproved DYK noms at which the requirement for the second QPQ turns on, and the lower number when it turns off again. (We don't want a single number for both, and we should probably discuss what makes sense before an RfC begins.) Also, is the determination as to the number of QPQs due based on when the nomination was made, when the initial full review is made, or when the final approval happens? We should decide this as well so it's clearly stated up front. I should note that I absolutely Oppose the idea that we "escalate" beyond two QPQs. If going to two QPQs doesn't solve the backlog, then the basic assumptions behind the proposal are flawed and some other approach should be tried, not doubling down on something that isn't working. BlueMoonset (talk) 16:10, 19 July 2021 (UTC)[reply]
  • I absolutely Oppose the idea that we "escalate" beyond two QPQs – You don't know Levivich well enough to know when he's kidding. I do. He was kidding.
  • I don't think we can tie ourselves to a hard, fixed number of credits as the point where the double-review requirement kicks in -- there are too many imponderables and we may need to adjust according to experience. We can say for certain that it should never be less than 10, and almost certainly wouldn't be higher than 20. You see, 20 might be too high -- not enough double reviews getting done -- so it might need to be as low as 10. But if you don't need to be lower than 20, it might be better that it not be, so that editors have as much experience as possible before becoming subject to the double requirement.
    So in summary, I think we should say we expect it will start at 15, but in time might be adjusted to anywhere between 10 and 20 by discussion on this page.
  • Setting the high and low trigger points (number of unreviewed nominations) in advance is similarly difficult. Among the considerations:
    • Some proportion of "unreviewed" noms are in fact being reviewed, just the review isn't complete, issues have arisen that require the nominator's attention, etc.
    • Even after you exclude those, it's desirable to not go below having a few dozen noms ("virgin" noms -- no review started) so that the pump is always primed.
    • We don't want to cycle between "doubles required" and "doubles not required" too quickly, because that causes confusion. (That's a consideration for the 10-20 question as well.)
I want to gather some statistics before going on, but I've gotta go right now.
EEng 17:05, 19 July 2021 (UTC)[reply]
  • Comment. By specific I think more exact wording for the proposal is requested. As we have 6 support votes for N=20, that would be the base number, so your proposal should start with something like below and then give the background to the RFC, its need etc. I don't think it will add anything to the RFC if we're complicating it with too much detail about changes to N we might want to make in the future.
  • (a) First five noms, no QPQ required;
  • (b) 6th to 19 noms, one QPQ required;
  • (c) 20 or more noms, two QPQs required;
Desertarun (talk) 19:03, 19 July 2021 (UTC)[reply]
  • Comment: I think if we don't tie ourselves to a hard minimum number of credits—remember, we want reviewers to have gained enough experienced before we add a second review to their dish—we're going to have problems. The idea that someone with five reviews under their belt has enough experience to graduate to a doubled reviewing requirement is highly problematic: I count myself a good reviewer but I was not yet one after five QPQ reviews. You can count on my opposition if the number isn't at least 20 credits (approximately 15 DYK reviews). Desertarun, a couple of points on your breakdown: first, let's make it 21 or more: first 5 credits are free, nominations for the next 15 credits (through 20 total) require one QPQ, with two QPQs thereafter. Second, please note that it's a nominator's number of credits (nomination and creation/expansion both) that governs, not just the number of their nominations: right now, the first five DYK credits are free but your next nomination after that, whether you were the nominator on all five or not, requires a QPQ. (PS: Levivich may have been kidding, but he wasn't the first to suggest going beyond two reviews per nom.) BlueMoonset (talk) 01:49, 20 July 2021 (UTC)[reply]
  • As there haven't been any comments for a while I've marked the proposed text as final. Desertarun (talk) 18:44, 1 August 2021 (UTC)[reply]
    Well, actually, I was waiting for the Olympics rush to die down before bringing up a last point. I think we're agreed (passively, at least) that we're not setting criteria right now for when we go into "backlog mode" (where the double-QPQ requirement holds) and when we come out of it, because there are too many imponderables. I'm just wondering whether we should remain silent about that in the RfC text, or say something about it. I don't want opposition because that bit's undefine. Other than that I believe we're ready to go.
    If we are ready to go, then I'm going to ask BlueMoonset to post the RfC because (a) I forget how to do it and (b) BMS is the godfather of DYK and the name carries weight. EEng 19:03, 1 August 2021 (UTC)[reply]
    I think the only text we could add would be along the lines of "the DYK people will decide when there is a backlog". This would be self evident anyway, as a management issue of the DYK process. I wouldn't add this to the text myself nor be resistant to it being added. It just feels a bit redundant, unless I'm missing something. I think it would be a good idea for BMS to send the RfC. Desertarun (talk) 19:36, 1 August 2021 (UTC)[reply]
    Alternatively you could ask one of the admins in this thread. Or maybe all of them on a more the merrier basis. Desertarun (talk) 09:07, 3 August 2021 (UTC)[reply]

RFC post

Since a lot of people seem very set on 20/21 as the boundary, let's go with that. So is this right as you see it, BlueMoonset? Feel free to fix (and we'll discuss what goes in the placeholder in (c) separately) ...EEng 06:47, 20 July 2021 (UTC)[reply]

I've started work on version 1. Desertarun (talk) 07:19, 21 July 2021 (UTC)[reply]
I heartlessly eviscerated it [2]. EEng 12:49, 21 July 2021 (UTC)[reply]
I put version numbers there expecting you'd move mine to version 2, and over time they'd merge to an agreed post, I wasn't expecting you to delete it. I'm not going to be putting my name to this, it is your proposal but getting the wording right will be difficult and right now neither yours nor mine is that good. Desertarun (talk) 14:08, 21 July 2021 (UTC)[reply]
I didn't delete it. I thought I could improve what was there and edited it, and in the end it came out a lot different. If you think you can improve it, do so. (I included a diff above for easy access to the prior version.) But remember that too much background is undesirable because many people just stop reading. There's no need for parallel versions. EEng 16:31, 21 July 2021 (UTC)[reply]
Your version is better than mine and I agree parallel versions aren't really worth it. I'm thinking the opening line should state the purpose of what we want to achieve for the community, this would be reduced waiting times. I don't think that falls into background info, it is a selling point, or the hook, for the proposal. Desertarun (talk) 18:43, 21 July 2021 (UTC)[reply]
@Vaticidalprophet: Your first line is blindingly clever! I'd like to use it but have no idea if we can. Thank you for making me laugh! Desertarun (talk) 21:36, 21 July 2021 (UTC)[reply]
<coughs politely> [3] EEng 21:47, 21 July 2021 (UTC)[reply]
Laughing again! ;-) Desertarun (talk) 21:51, 21 July 2021 (UTC)[reply]
You[Confused editor?] were so blinded by the cleverness that you confused Vprophet for me. I don't know which of us is more insulted. EEng 21:55, 21 July 2021 (UTC)[reply]
Isn't the important part that they talk about you at all? Vaticidalprophet 22:25, 21 July 2021 (UTC)[reply]
There is only one thing in the world worse than being talked about, and that is not being talked about. EEng 01:57, 22 July 2021 (UTC)[reply]
  • I think we should be using the word voted, because the vote we had here, by the people that know the issues, was important. We're looking for the RFC to rubber stamp our vote. Desertarun (talk) 07:44, 22 July 2021 (UTC)[reply]
    Because it's my own scheme I want to make sure this point is clear in case anyone feels it misrepresents: per Desertarun's suggestion I've changed DYK regulars broadly support to DYK regulars have !voted to support. EEng 20:56, 22 July 2021 (UTC)[reply]

RfC text

Did you know ... that articles nominated for WP:Did You Know often wait months for approval?

Editors nominating an article for Did You Know generally must review someone else's nomination; this is called the QPQ (quid pro quo). But because new DYKers are allowed to make up to five nominations exempt from this requirement, there are always slightly more nominations coming in than reviews being performed, and over the years a large backlog has accumulated. To reduce this backlog, so that nominations can be reviewed more promptly, DYK regulars have !voted to change the rules for DYK's review requirement. Those rules currently provide

If, at the time a nomination is promoted to the main page, its nominator has fewer than five DYK credits (whether or not self-nominated) then the nomination is exempt from QPQ.

The proposal is to substitute this text:

At the time a nomination is promoted to the main page ...
  • (a) if the nominator has fewer than 5 credits (whether or not self-nominated), no QPQ is required;
  • (b) if the nominator has at least 5 credits but fewer than 20 credits, the nominator must do one QPQ;
  • (c) if the nominator has 20 or more credits, the nominator must do two QPQs (if there's a backlog) or one QPQ (if not).

"Credits" and the "first five free"

  • Popping in a bit, I've barely been able to edit these past few days...I'm unconvinced having a cyclical system is a good idea. I recognize EEng started talking about it, and it didn't net active opposition, but it strikes me as primarily adding an additional layer of confusion. I already suspect there will be people coming in confused that suddenly they have to do two QPQs, is it really a good thing to spring "oh, and by the way the rules will change on you repeatedly" on them? I think we at least want to trial this at first, to see if ending up with too few open noms is even a risk with this method. The 20-credits range still includes plenty of people who don't follow DYK with the intensity this seems to demand. Vaticidalprophet 22:25, 19 July 2021 (UTC)[reply]
    If by "cyclic system" you mean having times when the double-QPQ requirement aaplies, and times when it doesn't, I'm afraid there's no way around that; we can't insist on a nominator doing an extra review when there aren't surplus noms waiting for review. What we can do, probably, is arrange for the transitions from double-requirement to no-double-requirement, then back to double-requirement, etc etc etc, be infrequent and far apart (on the order of many months). We do this by making upper and lower trigger points far enough apart, though I don't think that means making them as far apart as one might think. More on this later but first I'd like to see where people are with the (a) (b) (c) I posted above. EEng 06:59, 20 July 2021 (UTC)[reply]
    I support the a/b/c idea, but not for anyone but the nominator. People get added as creators, sometimes without their knowledge, for simply doing the heavy lifting to get an iffy nom into reasonable shape. —valereee (talk) 16:21, 20 July 2021 (UTC)[reply]
    For years I thought that for the purposes of "first five free", only nominations count -- your first five nominations are free, any nominations after that require a QPQ. Recently BlueMoonset pointed out that a close reading of the rules suggests that any "credit" counts, where a credit is being either the nominator or a creator/expander/GA-ifier; in other words, any time the bot congratulates you on your talk page, that's a credit that counts against your free five. Maybe that makes sense or maybe it doesn't, but apparently that's the way it already is. EEng 21:06, 20 July 2021 (UTC)[reply]

Wikipedia:Did you know#Eligibility #5 Review requirement, "If, at the time a nomination is promoted to the main page, its nominator has fewer than five DYK credits (whether or not self-nominated) then the nomination is exempt from QPQ." only refers to the nominator. It is not anytime the bot congratulates you. QPQ only applies to nominators. — Maile (talk) 21:41, 20 July 2021 (UTC)[reply]

Oh, wait a minute. Now I think I see the confusion "whether or not self-nominated" ... maybe that needs to be clarified. — Maile (talk) 22:26, 20 July 2021 (UTC)[reply]

Not sure what you're saying in your second post, but as written: (1) yes QPQ applies only to nominators, but (2) one of your "free five" gets eaten up each time you get any credit (nominator, creator, expander, GA-ifier), not just when you nominate. I'll say again that for 10 years this had escaped my notice, but now that BlueMoonset has pointed it out, it does seem to be correct. Plus I can't recall BMS being wrong about anything ever, so resistance is futile.
But maybe we should resist after all. We're always fretting about inexperienced people doing reviews. Now, the way I thought (and Maile thought, and it appears valereee thought) things worked, by the time you're required to do a review you'll have had the experience of having made 5 nominations, and thus been on the receiving end of 5 reviews. (Let's call that the Maile Rule.) But under the Blue Moon Rule (shall we call it), you might have done nothing but stood by innocently while someone nominated five articles you created; then, someday, you make your first DYK nomination and BOOM!, you have to do a review.
Notice that if we switch from the Blue Moon rule to the Maile Rule, the backlog gets bigger, but maybe not by all that much -- 75% of noms currently awaiting approval are self-noms, though that's only a rough indicator. EEng 23:04, 20 July 2021 (UTC)[reply]
Resist! Resist! Power to the newbies! To tack on a QPQ for being the creator (but not nominator) of a previous DYK, is a bit like handing your credit card over to a sales clerk who, after running it through their scanner, sort of mumbles, "Oh, by the way, the store reserves the right to add other costs to your purchase." — Maile (talk) 23:24, 20 July 2021 (UTC)[reply]
Yeah, I had no idea that was the rule. Belle came in here just a couple days ago wondering how they'd ended up with a credit on their talk. I agree that it probably doesn't add much to the backlog, and no one who hasn't been reviewed five times should be required to do reviews. I guess someone could game it...just keep finding new stooges to nominate your articles...tell me, (username redacted), did EEng offer you anything to make that nom? —valereee (talk) 11:10, 21 July 2021 (UTC)[reply]
I've got probably hundreds of reviews that I could use for QPQ (unless they expire); anybody who wants to nom and avoid a QPQ come see me and we can work out our own "QPQ". [wink] This is a joke. Please don't. Please. Belle (talk) 16:44, 21 July 2021 (UTC)[reply]
Now, Belle, remember what they said at your parole hearing. EEng 17:13, 21 July 2021 (UTC)[reply]
The credits came in when we closed the loophole that allowed people to nominate for each other without QPQs required by either creator/expander or nominator; only self-nominations had to have QPQs after the initial self-nominated freebies. Somehow, the bulk of those who'd taken advantage of the loophole managed to do QPQs once it was the nominator's responsibility regardless of who worked on the article, and the reviewing imbalance lessened. I'm really not seeing a problem if a nominator has one or two credits from another's nominations—it may be how they become aware of DYK in the first place, getting that first "credit" post on their talk page. It used to be that it was the fifth self-nomination that had a QPQ; that's long since morphed into the first five as freebies rather than four, but if someone is actively involved in four rather than five, it's not the end of the earth. People are always willing to help out a first-time reviewer. BlueMoonset (talk) 20:44, 23 July 2021 (UTC)[reply]
I don't really follow, since it seems like the loophole would have been closed by simply making all noms count, not just self-noms. But now is not the time to tinker with that. {{|BlueMoonset}}, you've been very quiet while the RfC text is being developed. What think you? EEng 22:17, 23 July 2021 (UTC)[reply]
  • Another point: for some reason, the proposed text includes the phrase at the time you make a nom, which changes how things currently work. If someone has only four credits, and they nominate another four within a short period of time and before any of these new ones appear on the main page, this says that all four are freebies, which is hardly desirable. How DYK currently works is that one would be free, but the other three would need QPQs. (This is similar to someone with four credits nominating a four-article DYK: the rules are specific that the first of the four is free, but the rest will require QPQs.) It should be easy enough to reword so the wording expresses the intent (five free, fifteen at one QPQ each, and all subsequent at two QPQs each). BlueMoonset (talk) 03:54, 24 July 2021 (UTC)[reply]
    If someone has only four credits, and they nominate another four within a short period of time and before any of these new ones appear on the main page, this says that all four are freebies, which is hardly desirable. How DYK currently works is that one would be free, but the other three would need QPQs. I did pretty much exactly this (made a bunch of nominations in a short period with <5 credits), and it worked as the former, not the latter. I did the QPQs anyway because I was under the impression it was the latter, but in at least one case was told by the reviewer I didn't need to. I suspect there may be confusion on this point. Vaticidalprophet 04:12, 24 July 2021 (UTC)[reply]
    I suspect that confusion is due to nominations often not being seen, rather that an issue with the rule. The QPQchecker tool catches only when DYKs hit the front page (and thus a user's talkpage), so anyone using it won't see if a nominator has other open noms. I agree with BlueMoonset that it should be per main page rather than at the time of nom (and if the occasional reviewer/nominator misses this weird edge case and gets a 'freebie' then that's not a huge issue). CMD (talk) 04:26, 24 July 2021 (UTC)[reply]
    Jesus, it's all true, and I'm ashamed to say I missed this race condition. All of this complication arises from this idea of counting credits instead of just noms, but I'll repeat that now isn't the time to tinker with that (though I'm sorely tempted). Anyway, I fixed this in the RfC text. EEng 05:01, 24 July 2021 (UTC)[reply]
  • What is the purpose of the notes? In answering that question I think it is to say a) Don't take against this proposal because you think we're trying to introduce credits! and b) don't take against this proposal because you don't understand credits! and c) If you don't believe we have a credit system now here's the [link]. What we want from the RFC is permission to change but the credit system isn't changing and we don't need permission for it to stay the same. Maybe we could copyedit the explanation of what credits are and reword. I guess we could just leave it like that as well, because people often don't read notes. Desertarun (talk) 10:03, 24 July 2021 (UTC)[reply]
    I think I added the note when we discovered that so many people (including I) didn't understand how they worked. But I realize we don't need to mention it at all, since it's not changing. Ignorance is bliss. EEng 20:30, 25 July 2021 (UTC)[reply]
    Your latest additions are good, as you think best with the notes. Desertarun (talk) 20:33, 25 July 2021 (UTC)[reply]
    I'm not sure I can think of anything else to say, without input from others. Desertarun (talk) 20:37, 25 July 2021 (UTC)[reply]

Thinking about upper/lower trigger points

I've just counted what we have on the nominations page (not approved page).

  • 49 nominations less than a week old that have never been reviewed.
  • 71 nominations over a week old that are never reviewed.
  • 8 nominations under review in 0-7 days old
  • 10 nominations under review 8-14 days old
  • 17 nominations reviewed 15-28 days old
  • 25 nominations reviewed over 28 days old
  • 8 nominations in the process of withdrawing, commented upon but not reviewed or otherwise unclassified.
  • 188 nominations total backlog.
  • 120 or 63% are never reviewed

Based upon these stats I think we should go with 2 QPQ kicking in when there are 80 on the nominations page, and go back to 1 QPQ when we are below 40. Of those 40 we could expect 63% i.e. 25 to be new and unreviewed and 15 to be under review. Desertarun (talk) 21:59, 19 July 2021 (UTC)[reply]

I've mentioned some numbers above, but do we really need to have specific numbers in the RFC proposal? I don't think so. That's more of DYK management issue which should be discussed here, we don't want to be tied down by the RFC in case we need to change things. So we need some kind of generality statement for c) If you have 21 or more credits at the time you make a nom, two QPQs are required when we're in backlog. Desertarun (talk) 21:13, 22 July 2021 (UTC)[reply]
How long would we be in backlog initially? I know we were talking about a one off backlog drive like originally planned by me, which would be a good idea. Is that going to be mentioned? Desertarun (talk) 21:23, 22 July 2021 (UTC)[reply]
Striking, its just complicating matters, we should do that independantly of the RFC. Desertarun (talk) 21:33, 22 July 2021 (UTC)[reply]
So first, I've been meaning to thank you for gathering the stats. When I said above I want to gather some statistics, those were very much the stats I had in mind. Imagine my pleasure on discovering you'd saved my the trouble!
On the main question of trigger points, I had the same idea you do. It's impossible to predict how quickly the backlog will draw down once the new rule takes effect, but pretty sure it will be somewhere between really slowly and just plain slowly. And (as your stats confirm show) a LOT of unapproved noms are not "virgins" but noms with significant activity which has stalled for some reason, and this complicates even the question of what counts as "unreviewed backlog" (or is it "unapproved backlog"?). So what I'd prefer, if we could sell it, is to leave the high/low trigger points undefined for now, see what we learn as the drawdown progresses, and define the trigger points later. EEng 04:35, 23 July 2021 (UTC)[reply]
I'm going to put that in the text. We should also mention we won't be in backlog for that long. I added some rough and ready wording. Desertarun (talk) 07:27, 23 July 2021 (UTC)[reply]

Alternative wording to hopefully get the RfC going again

Here's what it currently says about review requirements:

5. Review requirement (QPQ) – For every nomination you make you must review one other nomination (unrelated to you)‍—‌this is called quid pro quo or QPQ. The review must address all five criteria listed here.
Exception: If, at the time a nomination is promoted to the main page, its nominator has fewer than five DYK credits (whether or not self-nominated) then the nomination is exempt from QPQ.

  1. You can do your QPQ review before or after you make your nomination, but for your nomination to be approved you will need to provide a link, at your nomination, to your completed QPQ review. For help in learning the reviewing process, see the reviewers' guide.
  2. Qualifying QPQs need to be full reviews, and not simply a "check mark." Ideally, a QPQ should be submitted within a week of a nomination. After one week, and a reminder to the nominator, a nomination may be closed as "incomplete." QPQs do not expire and may be used at any time for a future DYK nomination.

Here's potential alternative wording (changes shown with struck out text or red font):

5. Review requirement (QPQ) – For every nomination you make you must review one other nomination (unrelated to you)‍—‌this is called quid pro quo or QPQ. The review must address all five criteria listed here. Exception: There are two exceptions to the QPQ requirements:

  • If, at the time a nomination is promoted to the main page, its nominator has fewer than five DYK credits (whether or not self-nominated), then the nomination is exempt from QPQ.
  • If, at the time a nomination is made, there is a backlog and its nominator has 20 or more credits, the nominator must do two QPQs.
  1. You can do your QPQ review before or after you make your nomination, but for your nomination to be approved you will need to provide a link, at your nomination, to your completed QPQ review. For help in learning the reviewing process, see the reviewers' guide.
  2. Qualifying QPQs need to be full reviews, and not simply a "check mark." Ideally, a QPQ should be submitted within a week of a nomination. After one week, and a reminder to the nominator, a nomination may be closed as "incomplete." QPQs do not expire and may be used at any time for a future DYK nomination.

I've tried to capture the spirit of the previous discussion, reflect the main issue brought up in opposition (including by me), without adding any more complexity to this (e.g. by also defining how we determine a backlog). Thoughts? Schwede66 03:55, 9 August 2021 (UTC)[reply]

  • Sorry, Schwede66, I didn't see this until after posting my continuation of the RfC just now. I'm glad to see we're both pulling in the same direction. I'd be so proud if you'd be the first support. EEng 05:28, 9 August 2021 (UTC)[reply]

Oldest nominations needing DYK reviewers

The previous list having been nearly exhausted, this new list below includes 34 nominations that need reviewing in the Older nominations section of the Nominations page, covering everything through July 15. We currently have a total of 258 nominations, of which 114 have been approved, a gap of 144, down 21 in the past nine days. Thanks to everyone who reviews these.

Over three months old:

Over one month 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) 00:29, 3 August 2021 (UTC)[reply]

RfC on proposal to require a second QPQ from "senior" DYK editors (those with 20+ DYK credits) when there's a backlog of unreviewed nominations

Did you know ... that articles nominated for WP:Did You Know often wait months for approval?

Editors nominating an article for Did You Know generally must review someone else's nomination; this is called the QPQ (quid pro quo). But because new DYKers are allowed to make up to five nominations exempt from this requirement, there are always slightly more nominations coming in than reviews being performed, and over the years a large backlog has accumulated.

To reduce this backlog, so that nominations can be reviewed more promptly, DYK regulars have !voted to change the rules for DYK's review requirement. Those rules currently provide

If, at the time a nomination is promoted to the main page, its nominator has fewer than 5 DYK credits (whether or not self-nominated) then the nomination is exempt from QPQ.

The proposal is to substitute this text:

Noble but imperfect first proposal

Previous discussion here

At the time a nomination is promoted to the main page ...
  • (a) if the nominator has fewer than 5 credits (whether or not self-nominated), no QPQ is required;
  • (b) if the nominator has at least 5 credits but fewer than 20 credits, the nominator must do one QPQ;
  • (c) if the nominator has 20 or more credits, the nominator must do two QPQs (if there's a backlog) or one QPQ (if not).

Note: The above text is the result of extensive discussion above. 18:43, 4 August 2021 (UTC)

  • Support Some longtime DYK editors feel aggrieved at this additional burden, but the feeling is that these are the editors in the best position, because of their experience, to do an occasional extra review. And while it's hard to predict more than roughly, it seems like the requirement for extra reviews will be in force only about 1/4 of the time (because it takes a long time to build a backlog, and likely not too long to wear it back down). EEng 18:43, 4 August 2021 (UTC)[reply]
  • Support. We can easily reduce the average time it takes to get a DYK on the main page by a week or two. This would benefit the community greatly. It would be worth getting the current backlog down quickly, so the regulars aren't at 2 QPQ for long. Desertarun (talk) 18:52, 4 August 2021 (UTC)[reply]
  • Support no issue at all with this. The Rambling Man (Keep wearing the mask...) 18:56, 4 August 2021 (UTC)[reply]
  • comment - how many unreviewed noms is a "backlog", exactly? Best Wishes, Lee Vilenski (talkcontribs) 19:16, 4 August 2021 (UTC)[reply]
    After a good deal of discussion (#Thinking_about_upper/lower_trigger_points) we decided it's not possible right now to set the "trigger points" that tell us when we go into/out of backlog mode because there are too many unknowns about how fast the backlog will get eaten down, how many unreviewed noms we want to leave to "prime the pump" and so on. As with the switches between running one set vs two sets per day, with experience we should be able to come up with those trigger points, but initially we'll have to wing it. EEng 22:16, 4 August 2021 (UTC)[reply]
  • Support, seems perfectly reasonable, and a logical extension of giving editors a free pass for their first 5. Vanamonde (Talk) 19:19, 4 August 2021 (UTC)[reply]
  • Comment - I think that the text should say "At the time a nomination is made", since the QPQs are part of the nomination process. A definition of a backlog needs to be made. Flibirigit (talk) 19:23, 4 August 2021 (UTC)[reply]
    Flibirigit, the reason it doesn't say "At the time a nomination is made" is that if someone with four credits makes another four nominations in quick succession, they only have one more freebie but this would give them three beyond that (or potentially more if they're particularly industrious). This is specifically addressed for multi-article nominations, where someone with four credits making a four-article nomination gets one of them free and must supply nominations for the other three. This was already discussed in the proposal stage above. BlueMoonset (talk) 22:18, 4 August 2021 (UTC)[reply]
  • Support the spirit of this proposal. I would prefer that the backlog numbers be defined, with numbers far enough apart that the changeover from backlog to non-backlog happens infrequently (unlike the 120-60 rule for 24h/12h hook sets, where for a while a changeover happened every two weeks). Z1720 (talk) 19:34, 4 August 2021 (UTC)[reply]
    That's one of the reasons we didn't set these triggers until we saw what actually happens once the rule is in place: we were afraid we might set them so things switch back and forth too quickly. Once the trigger points are properly set, I think the back-and-forth will much slower than we have with the 12h vs 24h changes. The backlog grows quite slowly, so once the current backlog is eaten down sufficiently (whatever that means, and however long that takes) the double-review requirement goes into abeyance, and it will be probably be many, many months -- perhaps a year -- before it will need to be reinstated temporarily. EEng 22:42, 4 August 2021 (UTC)[reply]
  • Support with many thanks to the gracious editors who, up to this point, have already been doing extra reviews to help offset the newcomers' freebies. DanCherek (talk) 19:35, 4 August 2021 (UTC)[reply]
  • Comment This just shifts the backlog from the unapproved nom page to the approved nom page, correct? It sounds to me like getting through airport security faster so that you spend more time waiting at the gate. Does it really help to get them reviewed faster? – Muboshgu (talk) 19:37, 4 August 2021 (UTC)[reply]
    • If my understanding is correct, once the backlog is shifted to approved noms this will trigger the more frequent update schedule. This would require more work from prep builders, of course... --Paul_012 (talk) 19:53, 4 August 2021 (UTC)[reply]
      The backlog presumably is gradually reduced in both places—a somewhat quicker flow to the Approved page, and therefore a somewhat quicker increase to 120 approved and the shift to two sets per day—and then back to a normal pace. The "more work from prep builders" is a variable thing; all approved nominations will eventually need to be prepped, so it's a matter of when, not if. At some point, the approved will drop below 60, and the prep building will have to slow. BlueMoonset (talk) 22:18, 4 August 2021 (UTC)[reply]
    • We have to stop 2 preps per day because the approved noms are getting low - but there are still hundreds of unapproved noms. Desertarun (talk) 07:41, 5 August 2021 (UTC)[reply]
  • Comment: This is a good idea in principle, but I'm wondering about the practicalities. In addition to the above comments regarding the definition of a backlog (which will probably need different values for when one is triggered and removed), is there a tool for counting DYK credits? Manually checking whether a user has more than 20 seems like a rather non-trivial task. --Paul_012 (talk) 19:53, 4 August 2021 (UTC)[reply]
    Indeed there is such a tool: [4]. EEng 22:26, 4 August 2021 (UTC)[reply]
    Seems like it would need modification to show up to twenty. Is Betacommand still maintaining the tool and contactable somewhere off Enwiki? --Paul_012 (talk) 08:14, 5 August 2021 (UTC)[reply]
  • Comment - Good point above by Muboshgu. Also, refer to the list of older nominations regularly created by BlueMoonset. That list is effective in bringing the older un-reviewed to the forefront. Those are often lagging behind for various reasons - language, subject matter, details reviewers are not familiar with, etc. etc. Lots of reasons they aren't the easy pick of the crop. Is there something in this proposal that accomplishes the same thing? Otherwise, we're back to the beginning of that situation. How do you move the older ones that most reviewers are hesitant to tackle? — Maile (talk) 19:55, 4 August 2021 (UTC)[reply]
Further to the list from BlueMoonset. If you look at the oldest nominations, you will frequently see that BlueMoonset goes over those oldest nominations and leaves comments at the bottom of the nom template as to its status, or what is left to be done. A very helpful practice. — Maile (talk) 21:05, 4 August 2021 (UTC)[reply]
Comment to clarify this notice "Note: After 120 or more approved nominations, we rotate to two sets a day and when we drop below 60, we rotate to one set a day." right above the hook table list. Yes, we rotate. But the number of sets is manually changed by an admin here: User:DYKUpdateBot/Time Between Updates. It's automatic in the fact that the trigger numbers have already been decided. But it still takes a human admin to make the changes. — Maile (talk) 20:57, 4 August 2021 (UTC)[reply]
  • Oppose as a solution in search of a problem. No where in any of the DYK structure is a "will appear within x time from from initial nomination" stated. As a long time 10+ years contributor of noms this is effectivly penalizing me for contributing, and for something that is not a problem at all, and already handled regularly by BlueMoonset. Additionally as noted this will create confusion for anyone noming in a "backlog" timeframe, and then noming in a "not backlog" time frame. Also what objective definition of "backlog is being used, given that all THIS proposal does is punt the wait time to approved que rather then unapproved que, contrary to what Desertarun states.--Kevmin § 20:30, 4 August 2021 (UTC)[reply]
    • all THIS proposal does is punt the wait time to approved que rather then unapproved que – Not at all. The pipeline from the approved Q to the main page shifts into high gear when predetermined high-water mark of approved hooks is reached, and easily eats down any temporary surfeit.
    • already handled regularly by BlueMoonset – No it isn't. Despite BMS's efforts we have a large backlog that's been growing for years.
    • penalizing me – You had your free 5 and have found DYK a source of pleasure and satisfaction. Time to give little extra back. The double-requirement will only apply a small fraction of the time anyway.
    EEng 22:36, 4 August 2021 (UTC)[reply]
    Yah, NO, I give back every time I nominate an article and do a correct QPQ, there is no onus on me to have to do more then that give that this is a volunteer project. Additionally All that will happen is that the noms que will be shifted to the approved que, forcing a more frequent manual waffle between 24/12 main page rotation. Again for something that is not at any point call a problem in the DYK rules sets. you say this will not happen often, but that is irrelevant to this is a solution seeking a problem.--Kevmin § 00:33, 5 August 2021 (UTC)[reply]
    Yah, YAH:
    • a more frequent manual waffle between 24/12 main page rotation – Nonsense. The switches between 12 and 24 are required because the steady-state arrival rate of approved hooks currently matches, apparently, something near the midpoint of those numbers. For all we know, during times when the double-review requirement is in force the increased rate of approved hooks will more closely fit with 12-hour page rotations, and fewer "waffles" will be needed. You're not thinking about this straight.
    • I give back every time I nominate an article and do a correct QPQ – I guess I should have said "Time to give little extra back, because more is needed and people like you and me are in a good position to provide it". Luke 12:48: To whom much has been given, from him much will be asked.
    • there is no onus on me to have to do more – Absolutely. If this rule is adopted and you don't like it, you don't have to participate anymore.
    • Again for something that is not at any point call a problem in the DYK rules sets. you say this will not happen often, but that is irrelevant to this is a solution seeking a problem. – Sorry, can't parse.
    Honestly, I'm baffled by the intensity of your hostility. You act like the Bolsheviks are coming for your tiara. EEng 07:05, 5 August 2021 (UTC)[reply]
    Do me as an atheist a favor and leave the bible versus out of the conversation please. You do not know that the waffle will be less though, as your wording indicates. I have over 400 nominations due to being active for over 10 years. That doesn't mean I have one ounce of review ability out side the subject areas I nominate in, which are already sparse to begin with. I can't do anything with biography or location nominations since I don't know a fig about them. Forcing the very small percentage of contributors the have over 20DYK credits is not actually going to fix your faux problem. When will you start doing regular reviews? Again, this is a solution seeking a problem.--Kevmin § 15:50, 5 August 2021 (UTC)[reply]
    • Do me as an atheist a favor and leave the bible versus out – I'm an atheist myself, but not an antitheist, and I recognize that even fairytales contain wisdom.
    • When will you start doing regular reviews? – I've done far more full reviews than I've made nominations, and have made some 5000 contributions to other reviews and 2800 posts to this very page, so trying to paint me as not pulling my weight won't get you very far.
    • You do not know that the waffle will be less though, as your wording indicates. – Exactly: I was pointing out that the effect was unknown. Meanwhile you were asserting, based on no evidence at all, that the proposal will "[force] a more frequent manual waffle between 24/12 main page rotation".
      But since you've poked the bear I've now looked into the matter, and can state with certainty that this proposal will reduce the frequency of 12/24 waffles. Turns out the system spends most of its time in 12 mode, with occasional shifts into 24 mode when hooks become scarce. But when we're in 2-review-required mode, new hooks will become available at a faster rate, so there will be even fewer shifts into 24 mode. So you're totally and complete wrong. WRONG WRONG WRONG WRONG WRONG WRONG WRONG WRONG WRONG WRONG!
    EEng 21:34, 5 August 2021 (UTC)[reply]
  • Support in principle. I agree it would be good for DYKs to run faster. The automatic rotation from 24 to 12 hour sets prevents the backlog from just moving from one queue to the other - these processes are complementary. I do think we need to have enough unreviewed noms so an editor can find one to review in a subject area that they are confident they can do a good review. This would be a factor to consider when picking the thresholds, which intentionally are left TBD here. Another "tweak" I suggest is perhaps, when the backlog threshold is reached, requiring two QPQs from the main list or just one from the Bluemoonset old list - as a way to incentivize working on the old noms (which sometimes is like walking through a minefield). MB 21:50, 4 August 2021 (UTC)[reply]
    I thought about making the second review be from the "problem child" list -- after all, it's our most experienced reviewers who will be making these second reviews -- but it seemed a bit burdensome. For now I think the second-review requirement, and BMS's list of stalled noms, should remain complimentary. EEng 22:23, 4 August 2021 (UTC)[reply]
    According to this list, there at least 440 editors with 25+ DYKs (not all of whom are still active). Maybe those with 100+ (around 125) are "our most experienced reviewers". I wouldn't support all second QPQs having to be from the stalled list. MB 01:14, 5 August 2021 (UTC)[reply]
  • Strong Oppose per WP:NOTBURO. There is not a substantial problem here that needs fixing. IronGargoyle (talk) 01:10, 5 August 2021 (UTC)[reply]
  • Strong support per my prior comments during the formation of this RfC. This is a useful long-term solution to a resistant backlog and one of the best ideas to solve that issue WT:DYK has managed to come up with. I still think there's a couple more moving parts than ideal, but worst comes to worst we say the experiment failed, you know? 20+ credits is a pretty small minority of the project, even if it's an outsized proportion who watch this page (see the stat above showing only 440 of tens of millions of registered editors have 25+ credits); this does not actually put a particular amount of burden on people, certainly less than most alterntive proposals. It's the best possible mix of cost and outcome. Vaticidalprophet 01:20, 5 August 2021 (UTC)[reply]
  • Comment. I'm not happy about the prospect of making a nomination during a non-backlog time, making my single non-backlog QPQ before creating the nomination (as I always do and as I think everyone should be required to do), getting the nomination approved, and then suddenly weeks later in a time when I haven't planned for any availability getting a notification that my nomination is in trouble because now it's a backlog time and I need two. I think there should be some sort of mechanism so that the number of QPQs needed depends on when you did the QPQ not on when someone else gets around to promoting it. —David Eppstein (talk) 01:35, 5 August 2021 (UTC)[reply]
    • Shifting to oppose as worded, not because I have any in-principle opposition to a double QPQ when needed to cut the backlog, but because I don't think this is the right way to do it. It shifts the required time of QPQ even later in the process, from review time to promotion time, the opposite direction from what should happen. Really what needs to happen is stricter enforcement of it being done at nomination time. So this is just going to lead to people dragging out their required QPQs longer, or not doing them at all for nominations that have other problems that would have previously had QPQs, and these delays will cut into any improvement in process from more QPQs getting done. —David Eppstein (talk) 06:50, 5 August 2021 (UTC)[reply]
  • Oppose - As worded, "'At the time a nomination is promoted to the main page". Definitely not a good idea to leave the decision on the edge of promotion. I think we should stick with the policy of the review determines if the QPQ has been done. End of story on that. The burden of deciding that should not be on the promoter. — Maile (talk) 03:04, 5 August 2021 (UTC)[reply]
    There should be a way to finesse this so that the requirement remains, as it is currently, that the nominator has to do a QPQ review if they have five or more credits—they might only have three or four at the time the nomination is made, but have other promotions to the main page and are up to five before the nomination gets to the main page. We've had newer nominators have to come up with a QPQ after approval because of when their several nominations were promoted. The wording needs to deal with that, but a change was not intended in that regard. What would you suggest, Maile? BlueMoonset (talk) 03:31, 5 August 2021 (UTC)[reply]
    Uh ... "as it is currently"? I realize you have more experience and have more knowledge in your DYK knapsack than the rest of us. But this aspect is news to me, and probably others. It would be one more burdensome check point for a promoter. I don't think we should load promoters down with so much nitpicking stuff that they throw in the towel and abandon DYK. Most promoters are doing good to get the basics before promoting. It is not their job to re-review every nitpicking aspect of a review, just to do a promotion. The world as we know it will not come to an end, if this is not a part of the promoter's job. We should thank our lucky stars we still have participants willing to do the promotions. Enough is enough. — Maile (talk) 10:40, 5 August 2021 (UTC)[reply]
    Also, if a nominator is encouraged enough to do multiple nominations, the odds will even out in the long run for their QPQs. We should not be tacking it on as an added requirement because of recent circumstances. This is a little bit like driving down a long highway, only to be stopped at the end by a toll booth taker saying, "Oh, by the way. Hope you enjoyed the drive. But we're not going to let you off this road unless you pony-up a toll that was not applicable when you first entered the road." — Maile (talk) 11:08, 5 August 2021 (UTC)[reply]
    Maile, DYK nominators should be aware that after five credits they have to do a QPQ for all subsequent nominations. If they've received four, and then make four new nominations in short succession, they should expect to have to do their first QPQs on three of them. They can count, can't they? It's not like they didn't know from the outset that they'd have to start paying tolls after Exit 5... BlueMoonset (talk) 01:48, 6 August 2021 (UTC)[reply]
    OK. Whatever the majority thinks is the correct course to follow. — Maile (talk) 02:05, 6 August 2021 (UTC)[reply]
  • Question: Who declares when there is a backlog? My concern is that if this goes through, we will be switching between 24 and 12 hour sets a lot more regularly. The C of E God Save the Queen! (talk) 07:09, 5 August 2021 (UTC)[reply]
    We'll just have a little chat like we used to for the 12/24 switches before the fixed trigger points (for 12/24) were establshed. After a while we should be able to set trigger points for unrevied backlog/no unreviewed backlog as well, but that we'll need some experience first.
    I don't know where people get the idea that this proposal will cause more frequent 12/24 switches; there's no basis for that. (However, if someone could tell me about what proportion of time we current spend in 12 mode, and what proportion of time in 24 mode, then I probably could predict the effect on that of instituting this change. There's a third mode too as I recall, but I forget what it is. BlueMoonset?) EEng 07:25, 5 August 2021 (UTC)[reply]
P.S. I got the data I needed, and it turns out that the proposal will, without question, reduce the frequency of switches between 12 and 24. Details at #switches. EEng 21:34, 5 August 2021 (UTC)[reply]
  • Support per... everyone else that supported. Pamzeis (talk) 07:38, 5 August 2021 (UTC)[reply]
  • I support requiring more QPQ from experienced nominators. I oppose the proposal as written. The number of QPQs necessary should be determined and checked at nomination time. We shouldn't suddenly require an extra QPQ (or return one QPQ) depending on whether there is a backlog at a random point 1-80 days later when the nom hits the Main Page. —Kusma (talk) 12:52, 5 August 2021 (UTC)[reply]
  • Oppose no clear indication of the system is backlogged or not, and constantly switching between 1 and 2 QPQs required is just confusing for people. We should try making DYK processes simpler not much more confusing. Joseph2302 (talk) 18:19, 5 August 2021 (UTC)[reply]
  • Oppose I support this in principle. My preference would have been that we sort out the process how to determine whether we are in backlog before posting this RfC, but even without that element I would still be able to support this. What I have a problem with is what Maile points to. A single or double QPQ requirement should be clear at the time of nomination, as it's much simpler for this to be checked by the reviewer. It gets more complex for prep builders and then the admins who promote to queue having to double check when backlog kicked in and how it relates to individual nominations. I appreciate that "at the time of nomination" is the current requirement and there are good reasons for it. But the double QPQ requirement should not become a complex burden on DYK volunteers. Schwede66 21:25, 5 August 2021 (UTC)[reply]
  • TIME OUT EVERYONE I'm working on fixing what Schwede (and several others) have expressed concern about. EEng 21:45, 5 August 2021 (UTC)[reply]
    Sure, but PLEASE DON'T EVEN JOKINGLY BRING IN THE BIBLE. No place at all for that. The Rambling Man (Keep wearing the mask...) 22:09, 5 August 2021 (UTC)[reply]
  • Oppose This counter-intuitive proposal penalises those who have already contributed the most to DYK and the project in general. Writing a good article already takes a considerable effort and a thorough review even more as so many nominations are of poor quality and in need of significant remedial work. The most experienced reviewers are also the most experienced writers but will they bother to spend their time here if they have to review twice as many articles as anyone else? Philafrenzy (talk) 23:06, 5 August 2021 (UTC)[reply]
  • Oppose too complex, too many more rulez. I'd do more voluntary reviews if I didn't have to spend my time dealing with objections in discussions that get longer than the articles. - Can we make reviewing simpler perhaps? Look for basic requirement instead of fine-tuning hooks and ignoring that what's interesting to one may be boring to someone else, and what's interesting to all will likely be trivia for all. --Gerda Arendt (talk) 07:19, 6 August 2021 (UTC)[reply]
  • Oppose. The proposal as written is unclear as to what constitutes a backlog and being asked to do an extra QPQ when a nomination is finally promoted is like moving the goalposts during a game. As other have mentioned before this penalizes users who plan out QPQs in advance and spend considerable time writing one article. Flibirigit (talk) 20:17, 6 August 2021 (UTC)[reply]

Update: The cabal is pondering what to do. Stand by. EEng 04:21, 8 August 2021 (UTC)[reply]

  • Oppose per Philafrenzy and Gerda etc. Reviewing one article per nomination is fair on those nominating DYKs. More than that means you're forcing people to give more than they get, and linking it to some nebulous state-of-the-backlog is even more arbitrary. It also runs the risk of lowering the quality of the reviews actually performed. DYK has been running for more than 15 years now, and is arguably at its highest quality level thus far - certainly the error rate seems way down on 2-3 years ago. It will continue to function fine without this proposal.  — Amakuru (talk) 15:29, 8 August 2021 (UTC)[reply]
    Even though I've asked a hundred times now that people stop commenting since the proposal's going to be modified to address concerns expressed so far, I can't help responding to the following startling statement: you're forcing people to give more than they get. We're here to build an encyclopedia, not "get" something. It's all give. EEng 16:06, 8 August 2021 (UTC)[reply]
    Oh, I'd have to disagree on that. I will go with the majority on whatever is decided here, but this is a process outside of building an encyclopedia. And I think the giving and getting refers to completing reviews vs receiving reviews, not a traditional reward. Kingsif (talk) 19:03, 8 August 2021 (UTC)[reply]
    I'm sorry, but that's just the wrong way of looking at it. Everything we do here is about building the encyclopedia, and that includes the main page features (which contribute indirectly, by building goodwill and attracting new editors). When you nominate a hook, you're not "getting" something, you're contributing a hook to the front page. When someone reviews your nomination, they're not doing you a favor, but helping to maintain the quality of what we present. It's all give. When you start talking about give and get, you quickly end up with people saying stuff like Amakuru: why should I give more than I'm getting? EEng 21:08, 8 August 2021 (UTC)[reply]
    It's all give, yes, but I interpreted Amakuru's words to basically say why should [user A] complete more reviews than his noms receive, or: why should only few users be the ones keeping DYK running. As said, I don't really care how this ends up, there's arguments on both sides, but wanting a few editors to do the heavy lifting, especially already-established users, newcomers to DYK who are not asked to do as many, will not, thus not gaining experience, and furthermore, it may appear as a boys' club of a few regular contributors that other people can't really break into editing. I have no idea if these issues were on Amakuru's mind, but even if not, "why should X users do more reviews than Y users" is a valid question that hasn't really been discussed here. No DYK "senior" is throwing up objections that I can see, but that's not the only potential problem with implementation: e.g. experienced DYK editors also nominate a lot more than others, often doubling up in certain common areas, and won't be able to review their own noms. How's that bottleneck going to work itself out? Kingsif (talk) 23:09, 8 August 2021 (UTC)[reply]

What I think isn't getting through is that, once the current backlog is dealt with, we'll probably be in "backlog mode" something like two months out of each year, during which 25 experienced DYKers will do 2 extra reviews each, and after which that'll be it until another year goes by. Is that really worth all this gnashing of teeth and tearing out of hair? (I have no idea what this "doubling up in certain common areas" bottleneck is you refer to.) EEng 23:42, 8 August 2021 (UTC)[reply]

Honestly, it takes maybe 10 minutes to do most DYK reviews. Sometimes doing a second one really isn't much of an imposition. —David Eppstein (talk) 23:56, 8 August 2021 (UTC)[reply]
From my first comment I noted that I don't care, I haven't !voted. You keep ignoring the logistical part; it's not much of an issue, I've got a lot of past reviews done that I won't need to do any new reviews any time soon even with two. I don't care and I think it's on everyone to review as many as we have time and energy for. But limiting it to only certain editors could be an issue whether they care or not. While the proposals may solve some issues, they may introduce others, and that's why I don't care and haven't voted: I do not think there will be any significant improvements to the process. You want a long explanation of the issue and "doubling up in certain common areas" bottleneck? Ugh, in collapse because long: Kingsif (talk) 00:29, 9 August 2021 (UTC)[reply]
tl;dr "if you want the prolific users to review more, you must consider they also nominate more, and there is no direct trade-off"

Besides making newbies feel like it's exclusive here, let me explain the doubling up comment. We get a lot of noms about NYC and opera; often, they have the same pair or 2/3 established DYK contributors as co-nominators. They can't review their own noms, and presumably reviewed 4 to be able to nom. That means if they want to nominate another DYK, say with a double link, they'll need 8 reviews, and won't be able to review their own noms. There may still be a backlog, but certain users won't be able to review, because taking into account most noms are opened for review fairly quickly and just not approved for a while, there's only a limited number of new noms a day and not enough to support the noms from these users not including their own. And if half the other noms are from other experienced users also co-nominating, you face the same problem. It becomes a bottleneck of new noms from the same users who have maxed out reviews they can do, until we get inexperienced users reviewing one of those noms, possibly only nominating one article, and slowly building up enough of a backlog to support users having to do 4/8 reviews that aren't of their own noms. Do you see the issue with that stagger? There will always be a base number of unreviewed noms in such a system unless some users take it upon themselves to review all outstanding noms without nominating themselves, something I've done but it's still mental tax, you know. Facing this scenario it would work better if all nominators had to do 2, it would clear the experienced users' noms and give them more other noms to review quicker. And that's just the first hiccup that came to mind: if you want the prolific users to review more, you must consider they also nominate more, and there is no direct trade-off (in giving review->getting review), so there will still be a backlog, but rather than being spread across all noms, the backlog will be disproportionately noms made by the experienced users. (The second level of that is that those noms typically have fewer issues, and so it might mean the average time from making noms/opening reviews to getting them approved becomes longer, also not ideal) Kingsif (talk) 00:29, 9 August 2021 (UTC)[reply]

To be honest I can't follow what you're saying, but I don't think you understand the proposal. We'll exit "backlog mode" way before the well is empty, probably when we're down to something like 80 open noms. There will always be plenty of noms needing review. David Eppstein, you're a genius computer scientist, specialist in queuing theory, and expert on the conservation of mass and the Third Law of Thermodynamics -- do you have any idea what K is talking about? EEng 01:16, 9 August 2021 (UTC)[reply]
Kingsif appears to be misunderstanding the backlog part of the proposal, thinking incorrectly that it will require 2 QPQs always, and worrying that this will mean that there won't be enough unreviewed nominations for nominators to review, preventing them from fulfilling the QPQ requirement. Because this proposal only kicks in when there is a backlog, this can't happen. —David Eppstein (talk) 01:21, 9 August 2021 (UTC)[reply]
No, David. I'm not misunderstanding. In a backlog, people will still be nominating, right. Regular DYK editors will be reviewing the backlog but also adding to it. In fact, this proposal relies on that, since it's about nom requirements. So then there comes a point where there is still a backlog but the nominations that were originally causing it have all been reviewed, just replaced with noms from people who are regulars. Other people will still be nomming and reviewing, but not at the same rate since they're not required to. Reviewing comes to a standstill because the people you want to do more reviews can't review their own noms, so can't review the backlog they've made. This creates a bottleneck as other users slowly chip away at reviewing those noms. To recap: there's a backlog, this clause kicks in, and there's a sudden rush of more noms being reviewed. Then a long period when there's still a backlog but reviews are tackled slower. For that not to happen, we have to rely on the regular editors to review *without* nominating. But this clause is making a requirement related to nominating. I don't see it improving the system, but I'll keep on as I have been, it won't really change much for me, personally, either. Kingsif (talk) 13:46, 9 August 2021 (UTC)[reply]

Take 2 -- Everyone please vote again on revised proposal

The rules for DYK's review requirement currently provide that:

If, at the time a nomination is promoted to the main page, its nominator has fewer than five DYK credits (whether or not self-nominated) then the nomination is exempt from QPQ.

Should this be replaced with more comprehensive criteria as detailed below? EEng 05:22, 9 August 2021 (UTC)[reply]

OK, you nitpicking rabble, try this:

At the time of nomination ...
  • (a) if the nominator has previously nominated fewer than 5 articles (self-nominations or otherwise), no QPQ is required;
  • (b) if the nominator has previously nominated from 5 to 19 articles, the nominator must do one QPQ;
  • (c) if the nominator has previously nominated 20 or more articles, the nominator must do two QPQs (if DYK is in "unreviewed backlog mode" at time of nomination) or one QPQ (otherwise).
A nomination lacking a required QPQ should not be reviewed (other than to note the incompleteness) until the QPQ is supplied.
The high- and low-water marks for entering and leaving "unreviewed backlog mode" will be determined by a later discussion, after experience with how quickly the new requirement eats down the backlog, and how quickly it grows back.
Multiple articles in a single hook: Suppose you have nominated 4 articles previously, and now nominate a single hook with articles A, B, and C "bolded". Then A is your fifth article and requires no QPQ; B and C are your sixth and seventh articles, and each requires a QPQ.
  • By way of namedropping, BlueMoonset and David Eppstein collaborated on this.
  • This resolves the main concern expressed to date, which is the confusing interaction between nominating "now" and the exact QPQ requirement (none? one? two?) being notionally established only "in the future", at the time of main-page appearance.
  • The backlog grows very slowly (the current one took years) so "unreviewed backlog modes" should be short and far between – something like a month or two once a year.

EEng 05:22, 9 August 2021 (UTC)[reply]

  • I opposed the earlier version (because it seemed likely to cause delays and uncertainty in how many QPQs were needed) but this one looks good to me, so if we go back into a formal RFC it would have my support. I think that, to keep DYK running smoothly, sometimes having to do two QPQs is a reasonable price for those who nominate frequently (among whom I count myself). And I think the change of wording to count nominations rather than DYK appearances also adds needed clarity. —David Eppstein (talk) 06:06, 9 August 2021 (UTC)[reply]
  • Support That looks good. Thanks. Would support when the formal RfC starts. Schwede66 07:07, 9 August 2021 (UTC)[reply]
  • Looks fine to me but I don't see much need for an example. You could separate the actual rule changes from 'guidance information' about the stuff that would be discussed on the talk page. I'm not sure if we are still in RFC or not though? Are you closing this RFC and making a new RFC, or am I supposed to vote now? Desertarun (talk) 10:31, 9 August 2021 (UTC)[reply]
    The example explains what ought to be common sense, but without it there's technically a loophole allowing someone to make a giant multiarticle nomination and do no QPQs. Bluemoonset was concerned about that so I included the example. EEng 12:32, 9 August 2021 (UTC)[reply]
    I suspect almost nobody would know this loophole or be able to exploit it. Even if they did the 2 QPQ would be able to undo the effect easily enough. Anyway it can be left safely enough. Desertarun (talk) 13:11, 9 August 2021 (UTC)[reply]
  • Support. But I would also support at 30/3, 40/4 etc. etc. Alanscottwalker (talk) 13:55, 9 August 2021 (UTC)[reply]
    @Alanscottwalker: Your alternative proposal is clearly ridiculous. I have 277 credits; would you really expect me to do 27 reviews for nominating a single article? Schwede66 23:49, 9 August 2021 (UTC)[reply]
    With rounding it would be 28, actually :P. EEng 00:06, 10 August 2021 (UTC)[reply]
    As I think was said earlier, 'to whom much is given, much is expected,' it's not like every article we create is required to be here, but I could also support delay kick in, larger intervals, and/or a cap. -- Alanscottwalker (talk) 09:47, 10 August 2021 (UTC)[reply]
  • Not clear to me if we're voting on this yet, but I'd support, as above. The current QPQ rules allow for a non-zero number of nominations to occur without a review; we are currently relying on volunteers to deal with the the concomittant increase in the number of unreviewed nominations; based on the fact that this number frequently grows unreasonably large, relying on volunteers isn't always sufficient. Vanamonde (Talk) 16:31, 9 August 2021 (UTC)[reply]
    Yes, we're voting. I mean not-voting. We are !voting. Everyone !vote! COME ONE, COME ALL! !VOTE !VOTE !VOTE !VOTE EEng
    Where does the RfC statement begin, where does it end? The statement as copied to the RfC listings is a mess: it includes a {{collapse top}} but not the matching {{collapse bottom}}. --Redrose64 🌹 (talk) 21:22, 9 August 2021 (UTC)[reply]
    The bot probably got confused by the {collapse}. In desperation I've manually copied the appropriate text into the RfC listing, though I won't be surprised if that sends the bot into a tailspin. EEng 22:09, 9 August 2021 (UTC)[reply]
    Legobot will simply revert you. You need to correct the problem at its source, which is on this page. Legobot identifies the start of the RfC statement by the {{rfc}} tag; it identifies the end of the RfC statement by the next valid timestamp. What occurs in between must be syntactically complete: that is, for every opening construct there must be a corresponding closing construct. --Redrose64 🌹 (talk) 22:29, 9 August 2021 (UTC)[reply]
    No, the source of the problem is the stupid design of the bot; obviously the RfC statement should be a parameter to a template instead of some lame guess based on a signature. I guess people will just have to come here to find out what the proposal is. EEng 23:14, 9 August 2021 (UTC)[reply]
  • Neutral While I'm leaning toward support, I think that the current guidelines also would benefit from a training program to help people who are less familiar with the process learn the ins-and-outs. It's hard enough with the current guidelines to learn how to nominate a DYK for a page, let alone learn how to do the QPQ, promoting, and other such things. ~Gwennie🐈💬 📋22:28, 9 August 2021 (UTC)[reply]
  • Support. This is like herding cats. —valereee (talk) 22:36, 9 August 2021 (UTC)[reply]
  • Support this improved proposal. Johnbod (talk) 02:27, 10 August 2021 (UTC)[reply]
  • Support (summoned by bot) per WP:EEng is infallible. But seriously, either proposal is OK; this provides reasonable parity to production and assistance, as far as my experience has shown me this to be an issue. Urve (talk) 03:04, 10 August 2021 (UTC)[reply]
    See also User:EEng#correct. EEng 11:35, 10 August 2021 (UTC)[reply]
  • Comment - Isn't the simpler solution to do away with the 5 free noms? The nominators lacking the basic skills required, are likely to be those with a problematic and time consuming nomination. --Ykraps (talk) 05:54, 10 August 2021 (UTC)[reply]
    I think there's merit in allowing editors to go through the review process a few times to learn how it should be done first. Nominators lacking the basic skills required are also likely to carry out poor reviews just for the sake of doing QPQ and may end up creating more work for everyone. Armadillopteryx 06:15, 10 August 2021 (UTC)[reply]
    I just did my first DYK, and I'm pretty certain that if I had to review another to get the process started I would have likely not bothered. Learning one process to submit a DYK should have some time to sink in before you have to learn another process to review a DYK. ScottishFinnishRadish (talk) 11:36, 10 August 2021 (UTC)[reply]
    I would argue that anyone that has written and kept an article in mainspace, or has 5X expanded an article without being reverted, has all the skills and knowledge required. They might not be comfortable doing it, but they are capable of doing it.--Ykraps (talk) 18:13, 10 August 2021 (UTC)[reply]
  • Support per everyone else. Pamzeis (talk) 08:00, 10 August 2021 (UTC)[reply]
  • Support - seems a sensible proposal to me. Cwmhiraeth (talk) 08:35, 10 August 2021 (UTC)[reply]
  • Support. This is like herding cats, except you also have groom others editors' scabby kittens. Martinevans123 (talk) 11:33, 10 August 2021 (UTC)[reply]
  • Support but I'd support even more stringent requirements (1 QPQ per every 5 articles, or 10 articles after 5 "free" DYK reviews, instead of these if...elseif...else loops). But since I don't expect these requirements to go that far, that's a good step forward. Szmenderowiecki (talk) 15:09, 10 August 2021 (UTC)[reply]
  • Oppose as per all my previous comments on how we should be making DYK less complicated, instead of more complicated. Joseph2302 (talk) 15:20, 10 August 2021 (UTC)[reply]

Scott Lee Kimball

This dyk nomination for Scott Lee Kimball looks good but I'm not sure if the article has copyvio. It has a lot of similarities on Earwig (65% with one article) but some of them seem like generic phrasing. Better safe than sorry so I'm asking for a second opinion before approving. BuySomeApples (talk) 05:07, 7 August 2021 (UTC)[reply]

It is a copyvio. There are multiple instances of close paraphrasing. Desertarun (talk) 13:29, 9 August 2021 (UTC)[reply]

Still searching for the missing

Six months since Yoninah went missing in action. There's not a day that goes by at DYK, that I don't wish she'd show up and tell us it was only a rumor. Sigh ... — Maile (talk) 20:41, 8 August 2021 (UTC)[reply]
Indeed, in her absence, things that she would have dealt with don't always get done, special occasion hooks get missed, and badly-worded hooks are less likely to be improved. We are currently in a position where we are running two sets a day, but struggling to keep up, particularly with prep set building. There are currently four empty prep sets, and the three that are filled were all promoted by me so I will be unable to move them on into the queue. So prep set builders are urgently needed, as well as queue movers. The number of approved hooks is reducing, but slowly, so we will have to wait awhile before we can get back to one set a day. Cwmhiraeth (talk) 08:57, 9 August 2021 (UTC)[reply]

I'm a bit uncomfortable with this "supposedly haunted house" line, which doesn't really convey why it might or might not be haunted. And the wording in the article is a little problematic too. The text says "Likely due to its historic nature, the Morris–Jumel Mansion is considered by some to be a haunted house", which is WP: WEASEL because it doesn't clarify who exactly considers it haunted. This at least should be reworded with attribution IMHO, and probably the hook wording modified slightly to remove the term "supposedly". Pinging @The Spirit of Oohoowahoo, Epicgenius, Ceoil, and The C of E: who were involved with the hook. Cheers  — Amakuru (talk) 18:04, 9 August 2021 (UTC)[reply]

... that Washington Heights, Manhattan has a house, where the musical Hamilton was written, that is haunted according to Ghost Adventures and The Today Show? SL93 (talk) 18:46, 9 August 2021 (UTC)[reply]
That's an improvement, although perhaps the hookiness is then compromised? If those sources are attributed properly in the article, it would probably be OK for the hook to say something along the lines of:
"... that the Morris–Jumel Mansion in Washington Heights, Manhattan, where the musical Hamilton was written, has been described as a haunted house?"
Amakuru I took it from the source - "Featured on THE HOLZER FILES, GHOST ADVENTURES, HAUNTED USA: NEW YORK, and the TODAY SHOW, the Morris-Jumel Mansion is perceived by many as a paranormal site and has attracted academics and investigators such as Hans Holzer, Zak Bagans, the Tennessee Wraith Chasers, and now you!". I do have the links to The Today Show and Ghost Adventures which might be better links. If someone can reword the article's sentence, I would support your alt. SL93 (talk) 19:26, 9 August 2021 (UTC)[reply]
@SL93: Indeed, and I'm not questioning what you said at all. No doubt those are the correct attributions. Just pointing out that as far as I know, the hook doesn't absolutely have to attribute something, as long as the article does. I might try to reword it later on if nobody else does, and as we await further input from the nominator(s).  — Amakuru (talk) 19:32, 9 August 2021 (UTC)[reply]
I think we may have misunderstood each other. I just mentioned those links because the original source isn't independent. SL93 (talk) 19:34, 9 August 2021 (UTC)[reply]
OK then! It looks like The Spirit of Oohoowahoo has just updated the wording in the article, so that looks better now. @The Spirit of Oohoowahoo, Epicgenius, Ceoil, and The C of E: would you be OK if I update the hook to my suggested wording above? Or is there some other variant you'd prefer? Thanks  — Amakuru (talk) 21:08, 9 August 2021 (UTC)[reply]
Sure, that sounds good to me. Sorry about the late response. Epicgenius (talk) 21:14, 9 August 2021 (UTC)[reply]
Yes, this is better.Ceoil (talk) 21:23, 9 August 2021 (UTC)[reply]
Amakuru I revised my previous edit, we could do either stick with how you worded it or say "... that the Morris–Jumel Mansion in Washington Heights, Manhattan, where the musical Hamilton was written, has been rumored to be haunted for over two centuries?" Since I put sources talking about how it was considered haunted even before the Jumels bought it we could emphasize the historical aspect that way. The Spirit of Oohoowahoo (talk) 21:28, 9 August 2021 (UTC)[reply]
@The Spirit of Oohoowahoo: I suppose the only problem with this is that if we talk about "rumors", then we're once again dealing with unattributed statements. Perhaps in this case it's a moot point, since from an encyclopedic point of view it's not really a point of fact whether or not the house is haunted, it's simply a case of who believed it to be and when. The source says "rumor had it that the place was haunted" and "the superstitious said a Hessian soldier sometimes appeared on the winding stairs" which is annoyingly vague. Maybe we can simply attribute it to the NY Times and leave it at that? e.g. "according to a 1981 article in the New Yotk Times, there were rumors that the mansion was haunted by a Hessian ghost". With that in place, I'd probably be happy with your revised hook. Would that work for you?  — Amakuru (talk) 22:34, 9 August 2021 (UTC)[reply]
Amakuru I think it's better to stick with your original wording that Epicgenius and Ceoil agreed on, never mind my suggestion. The Spirit of Oohoowahoo (talk) 23:25, 9 August 2021 (UTC)[reply]

 Done. Thanks to all for the satisfactory discussion.  — Amakuru (talk) 23:45, 9 August 2021 (UTC)[reply]

Pictures 2

And there are some hooks that would work better without an image, like "... that Black Rock (pictured) is mostly concrete?" - the image reveals it isn't a rock, and that makes it distinctly not-hooky. Just a PSA for promoters to consider that kind of thing, too, when picking image slots. Kingsif (talk) 20:05, 9 August 2021 (UTC)[reply]

It's not even a good-quality photo. Schwede66 21:33, 9 August 2021 (UTC)[reply]
Kudos on your proper hyphenation. EEng 22:11, 9 August 2021 (UTC)[reply]
I agree; this is precisely why I choose not to include pictures with most of my hooks (e.g. Talk:Paradise (nightclub) and Talk:Industry Bar). Images often take the hookiness out of those hooks that depend on wordplay/ambiguity. Armadillopteryx 22:10, 9 August 2021 (UTC)[reply]
Agree- this would have worked much better in the quirky hook slot in my opinion. Joseph2302 (talk) 22:27, 9 August 2021 (UTC)[reply]

Queue 6

Independent admin check

I've just promoted Prep 6 to queue and for one of the nominations, I was the reviewer. Could another admin please do the usual admin checks for that hook?

... that Lisa Warrington was responsible for painting the doors of Allen Hall Theatre red?

Much appreciated. Schwede66 22:58, 9 August 2021 (UTC)[reply]

Heidi Grant Murphy hook

Further to this queue. Going through the hook checks, the Heidi Grant Murphy hook does not look right. I can't get to the source (paywall), but there's a direct quote from it in this article that states that the process of becoming a singer "takes work on your psyche, your innermost being." Whereas in the approved hook, it says:

... that soprano Heidi Grant Murphy, who has given over 200 performances at the Met, said that opera works on one's innermost being?

It taking work on your innermost being, and opera working on one's innermost being, are not the same thing, or are they? Schwede66 23:35, 9 August 2021 (UTC)[reply]

Pinging Gerda Arendt for soprano hook. SL93 (talk) 00:09, 10 August 2021 (UTC)[reply]
Sorry, should have pinged everyone who was involved (LouisAlainNetherzoneNarutolovehinata5). Schwede66 00:20, 10 August 2021 (UTC)[reply]
I don't have access to the sources, either, but the phrase in the article is in quotations. And I think it's written as a different meaning in the article, “the instrument in your throat” is not enough, and that the process of becoming a singer “takes work on your psyche, your innermost being.” As written in the article, that comes across as opera coming from within, and the individual has to work on their inner self to project it correctly. — Maile (talk) 00:26, 10 August 2021 (UTC)[reply]
That's how I interpret that, too. So it's the individual who has to do the work, and it's not opera that's doing the work for the individual, as the hook seems to imply. Schwede66 01:00, 10 August 2021 (UTC)[reply]
Hello all, thank you for catching that and for the ping. Would it be correct to change it to:
...that the soprano Heidi Grant Murphy, who has given over 200 performances at the Met said that becoming a singer works on one's innermost being? Netherzone (talk) 01:08, 10 August 2021 (UTC)[reply]
Perhaps we could avoid doing any interpretation of our own by using the quote in the hook, i.e.:
... that soprano Heidi Grant Murphy, who has given over 200 performances at the Met, said that becoming a singer "takes work on ... your innermost being"?
Armadillopteryx 01:26, 10 August 2021 (UTC)[reply]
Would it be too long to use the entire quote, "takes work on your psyche, your innermost being." There is something about the ellipsis that looks odd to me. Netherzone (talk) 01:31, 10 August 2021 (UTC)[reply]
It would still be under 200 characters (163, to be exact) if the full quote were used, so it's within policy, at least. Armadillopteryx 01:34, 10 August 2021 (UTC)[reply]
Is everyone ok with this as the revised hook:
...that the soprano Heidi Grant Murphy, who has given over 200 performances at the Met, said that becoming a singer "takes work on your psyche, your innermost being."
(courtesy ping Gerda Arendt) Netherzone (talk) 01:53, 10 August 2021 (UTC)[reply]
Looks fine to me, though I note that there is no "the" before "soprano" in the original hook (and since the hook is now a little longer, I'd leave that word out since it's unnecessary). Armadillopteryx 02:06, 10 August 2021 (UTC)[reply]
Agree! Netherzone (talk) 03:00, 10 August 2021 (UTC)[reply]

 Done Thanks. I've updated the hook; please have a look at Q6. Schwede66 04:47, 10 August 2021 (UTC)[reply]

Fine by me late riser. --Gerda Arendt (talk) 07:29, 10 August 2021 (UTC)[reply]
Sorry, Gerda. I should have considered waiting a bit longer as I'm aware which time zone you are in. Schwede66 09:59, 10 August 2021 (UTC)[reply]

Older nominations needing DYK reviewers

The previous list is running low on hooks and should have been archived tonight by the bot; this new list below includes 39 nominations that need reviewing in the Older nominations section of the Nominations page, covering everything through July 30. We currently have a total of 234 nominations, of which 115 have been approved, a gap of 119, which has decreased by 25 in the past seven days. Thanks to everyone who reviews these.

Over one month 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) 05:28, 10 August 2021 (UTC)[reply]