Misplaced Pages

talk:In the news: Difference between revisions - Misplaced Pages

Article snapshot taken from Wikipedia with creative commons attribution-sharealike license. Give it a read and then ask your questions in the chat. We can research this topic together.
Browse history interactively← Previous editNext edit →Content deleted Content addedVisualWikitext
Revision as of 14:18, 5 February 2021 editKtin (talk | contribs)Autopatrolled, Extended confirmed users18,713 edits COVID-19 banner: Replying to Amakuru (using reply-link)← Previous edit Revision as of 15:55, 6 February 2021 edit undoLaserLegs (talk | contribs)Extended confirmed users4,814 edits What's up with ongoing: new sectionNext edit →
Line 738: Line 738:
::By comparison, 69% of the internal clicks for ] and 67% for ] came from the mainpage, so it seems reasonable to conclude that the majority of those do represent people hitting the link while it was posted. ] (]) 22:09, 3 February 2021 (UTC) ::By comparison, 69% of the internal clicks for ] and 67% for ] came from the mainpage, so it seems reasonable to conclude that the majority of those do represent people hitting the link while it was posted. ] (]) 22:09, 3 February 2021 (UTC)
:::{{u|Andrew Gray}}, nicely explained. tl;dr{{spnd}} when looking at the above table, ignore the 'type' column. Cheers. ] (]) 03:21, 4 February 2021 (UTC) :::{{u|Andrew Gray}}, nicely explained. tl;dr{{spnd}} when looking at the above table, ignore the 'type' column. Cheers. ] (]) 03:21, 4 February 2021 (UTC)

== What's up with ongoing ==

. There are five opposes but not a single one of them addresses the staleness of the article. The ] criteria is clear "In general, articles are NOT posted to ongoing merely because they are related to events that are still happening. In order to be posted to ongoing, the article needs to be regularly updated with new, pertinent information." By that criteria, none of the opposes are actually valid. Has ] become a substitute for ]? Or should the criteria be amended to reflect what has become practice: stale articles festering in the box for weeks or months past their ] date while a dedicated cabal resists any effort to remove the article from the main page while simultaneously contributing nothing to maintain the article at a standard that would merit featuring.

I'm only using the farmers protest as an example, but there are many others such as the CAA protests, the Belarusian protests, the Venezuelan protests, the Hong Kong protests in each of these cases prying a demonstrably stale article out of the box proved to be nearly impossible. --] (]) 15:55, 6 February 2021 (UTC)

Revision as of 15:55, 6 February 2021

Please note:Please do not post error reports for Template:In the news here. Instead, post them to WP:ERRORS. Thank you.

Please do not suggest items for, or complain about items on Template:In the news here. Instead, post them to WP:ITNC. Thank you.

Please do not write disagreements about article content here. Instead, post them to WP:CEN. Thank you.
This talk page is for general discussions on In the news.
Please note: The purpose of this page is to discuss improvements to the In the news process. It is not a place to ask general questions, report errors, or to submit news items for inclusion.

Media mentionThis page has been mentioned by multiple media organizations:
Click here to nominate an item for In the news. In the news toolbox
Archiving icon
Archives

Index 1, 2, 3, 4, 5, 6, 7, 8, 9, 10
11, 12, 13, 14, 15, 16, 17, 18, 19, 20
21, 22, 23, 24, 25, 26, 27, 28, 29, 30
31, 32, 33, 34, 35, 36, 37, 38, 39, 40
41, 42, 43, 44, 45, 46, 47, 48, 49, 50
51, 52, 53, 54, 55, 56, 57, 58, 59, 60
61, 62, 63, 64, 65, 66, 67, 68, 69, 70
71, 72, 73, 74, 75, 76, 77, 78, 79, 80
81, 82, 83, 84, 85, 86, 87, 88, 89, 90
91, 92, 93, 94, 95, 96, 97, 98, 99, 100
101, 102, 103, 104, 105, 106, 107, 108, 109, 110
111, 112, 113, 114, 115



This page has archives. Sections older than 7 days may be automatically archived by Lowercase sigmabot III when more than 4 sections are present.
Archiving icon
ITNR archives

Index 1, 2, 3, 4, 5, 6, 7, 8, 9, 10
11, 12, 13, 14, 15, 16, 17, 18, 19, 20
21, 22, 23, 24, 25


RD turnover

Not sure if this is due to the WikiCup but not sure if our current RD changes are fixing the issue of having RDs up for <24 hours. In the past 24 hours we've had 11 new RDs posted to the template, and we've flexed to 7 spots as well. I IARed and "stickied" Larry King to the top of the bulleted list due to interest in having a blurb for him and perceived reader interest, but in the future this could be a slippery slope. Spencer 16:32, 24 January 2021 (UTC)

I'm really not keen on assigning "relative importance" to people within RD. After all, that was really the whole point of RD, where all mentions were determined to be exactly equal, and not to assign a super-notability subset. Either blurb or RD, not "super-RD". I suggest we definitely do _not_ do this because this precedent will now be called upon any time someone thinks an RD "deserves" more attention. Slippery slope indeed, and one we absolutely must not go down, please. The Rambling Man (Stay alert! Control the virus! Save lives!!!!) 16:41, 24 January 2021 (UTC)
Agree with TRM, though I know the pinning was done in good faith, the chronological order prevents chaos. --LaserLegs (talk) 17:08, 24 January 2021 (UTC)
Yeah, I'm not suggesting Spencer did anything outside good faith, and perhaps LaserLegs is right, be careful what we wished for when we discarded the chronological order. If this turnover becomes the norm, rather than the exception, we might need to think outside the box (literally). The Rambling Man (Stay alert! Control the virus! Save lives!!!!) 17:28, 24 January 2021 (UTC)
The last time this came up, there was 12 RDs posted on Jan 4.—Bagumba (talk) 01:48, 25 January 2021 (UTC)
  • Agree with TRM, LaserLegs, and Spencer. This is absolutely a slippery slope that we should not go down under, even if under the aegis of WP:IAR (though this one was done under absolute good faith). I had posted this issue earlier here. The problem is in the bursty nature of entries. As mentioned earlier, there are two variables here -- inflow rate and outflow rate. We have no control over the inflow rate (and rightly so) and the only variable that we can control is the outflow rate (i.e. the rate at which Admins pick up from ready articles and post onto the homepage). As complex as this one sounds, it is actually a relatively simple problem to solve by a process linked change. If there is a rough alignment that 24 hours is a reasonably good period to afford an article on the homepage, you can try the following. Articles should be absolutely evaluated from the bottom of the stack (both by editors and admins). Editors continue the current approach of evaluating an article as Ready and marking it as such with a tag i.e. "(Ready)". Posting admins start from the bottom of the stack and do a simple check, if the article can be posted, but, the article falling off the carousel has spent less than 24 hours, the Admin marks the article as "(Ready but hold)". When the next Admin comes along, and the article is ready to be posted (i.e. the falling article has spent at least 24 hours) they post it and change the tag to "(Posted)". If for some reason, the Admin determines that the article was not ready to begin with, they change the tag by removing the "(Ready)" tag and adding their comment on the changes requested. Ktin (talk) 17:44, 24 January 2021 (UTC)
    We're reaching the turnover point that it seems like holding items will lead to greater backlog, and unfortunately the thought that we have good control over the outflow rate assumes regular admin attention, which has proven difficult historically (there are just a few of us and I'm frequently unavailable). Spencer 21:50, 24 January 2021 (UTC)
Time to make itn and dyk longer. Sagittarian Milky Way (talk) 22:01, 24 January 2021 (UTC)
This would never have been necessary if Fuzheado had just honored the existing consensus.--WaltCip-(talk) 18:23, 24 January 2021 (UTC)
Perhaps the reservoir is the variable that needs adjusting if the inflow is fixed and the outflow is too rapid. C'mon didn't anybody else take 3rd year hydrology? Analogy aside, the number of people with articles on Misplaced Pages is ever increasing, and as such the number of eligible RDs on any given day is increasing past the point where 6 or 7 covers a 24 hour period of passings. If the container is emptied as fast as it is filled, a bigger container may be in order. - Floydian  ¢ 18:32, 24 January 2021 (UTC)
Floydian, I have tried making the case for a 'bigger container' as well in the past here and Martin tried here, but, there has not been an appetite for spilling into the third row. So, not sure what else can be done on that front. Ktin (talk) 18:41, 24 January 2021 (UTC)
... which is why (in my now restored post) I suggested "we might need to think outside the box (literally)". The Rambling Man (Stay alert! Control the virus! Save lives!!!!) 20:36, 24 January 2021 (UTC)
Has there been previous consideration into making the right column into 3 boxes instead of two? My personal opinion is that recent deaths are going to draw traffic more than selected anniversaries; perhaps some of that space can be... "annexed" as it were. - Floydian  ¢ 20:54, 24 January 2021 (UTC)
The right-hand side could be made longer: DYK often switches to two sets of hooks per day to keep their backlog down. Could just as easily have ten hooks per day and give the other side a couple more lines to use. The Rambling Man (Stay alert! Control the virus! Save lives!!!!) 21:01, 24 January 2021 (UTC)
There would have been an RD overflow, even if King had remained a blurb. It would not have drawn as much attention without a notable figure like him.—Bagumba (talk) 01:53, 25 January 2021 (UTC)
WaltCip, if you are referring to Larry King there was no strong indicator of consensus at the time of posting and the comments after the fact confirmed this. It also doesn't solve the long term problem that the current RD system is not scalable. -- Fuzheado | Talk 07:11, 26 January 2021 (UTC)
Fuzheado: What you are inherently asserting is that the admin who posted the item prior to you, in this case 331dot, interpreted consensus incorrectly. That's the problem that I have. It adds unnecessary chaos to the ITN process and diminishes the discretionary power that a single admin should have. It resembles, even if it doesn't appear to be so, wheel warring. --WaltCip-(talk) 13:30, 26 January 2021 (UTC)
I was think more like this. Granted, this was a case where non-EN conventions precluded a complete BLP. But by "biographical details" I mean DOB, DOD, education, family, etc.; things which have nothing necessarily to do with their WP:N qualities. Something like, must contain all of:
AND 5 of:
Some editors put emphasis on professional details in RDs, which is fine, but it sometimes creeps into CVs-as-RDs.130.233.213.199 (talk) 10:31, 25 January 2021 (UTC)
No, I can't agree with this suggestion I'm afraid. If anything the articles should be erring towards what they did during the significant part of their lives, that make them notable. E.g. actions while they were mayor, rather than padding out with irrelevant trivia about their private lives. The discussion process at ITN/C exists precisely so editors can scrutinise and make sure the article meets the required standards, and that's all we need. Anything else is WP:CREEP. Re the OP's point, I think the 24-hour expectation that has crept in lately is misguided. Remember, DYKs only stay up for 12 hours during periods of high activity, and there's no reason why RDs (which sometimes require less work to get up and posted than a DYK) should be any different.  — Amakuru (talk) 10:45, 25 January 2021 (UTC)
Me neither. In many cases most of these may not be available or may not apply and for good reasons which ought not to prevent an article being posted. Trying to create seemingly arbitrary objective criteria is exactly NOT what RD's transformation was about. The Rambling Man (Stay alert! Control the virus! Save lives!!!!) 16:04, 25 January 2021 (UTC)
Disagree with this proposal from IP editor. Artificial requirements should not be tapered onto WP:ITNRD just because we have not been able to smoothen our outflow rate. There are already homepage levels of hygiene that are mandated on any article that makes it to homepage, and we should stay consistent with those hygiene expectations. Nothing more and nothing less. And, definitely, not look to what can best be described as trivia (e.g. 1 juvenile fact) etc as a requirement for the article to get to homepage. I have maintained for sometime now that the amount of articles that are brought to homepage levels of hygiene as a part of WP:ITNRD is an absolute win for the community / project and if anything we should continue to encourage that and not dissuade that by adding artificial requirements (e.g. 4 family members, 2 educational institutions)! Ktin (talk) 15:59, 25 January 2021 (UTC)
I don't think there is a way to create a broad one-size-fits all requirement of ensuring this; this requires assessment of articles by ITN/C commenters and admins. And I would encourage ITN/C commenters to critically assess the quality of the articles. I won't post articles - even if marked "ready" - if they aren't off sufficient quality of posting. The period when we posted 11 RD in 24 hours, all of them were of decent quality, so this doesn't necessarily fix the underlying issues. Spencer 22:21, 25 January 2021 (UTC)
  • This appears to be an isolated spurt of RD postings and not indicative of any trend that needs addressing. When it becomes a trend that needs addressing, we can at that point address changes to the system to fix the problem. Random fluctuations are not subject to corrections by policy changes, and would still be a problem even if we changed the policy. I have seen no reason to believe this is any more than a random fluctuation in the number of RDs over a rather short period of time, and as such, can't be predicted or corrected for. If and when we have enough data over a long enough time period to say that this is a "new normal", then and only then should we be discussing policy changes. I'm fully on board with doing so if and when we can demonstrate they are needed. I don't see it yet. --Jayron32 12:49, 26 January 2021 (UTC)
    • Is there a way to see how many RDs were posted per month over the past 2 years? Seeing another couple of days with 5-6 RDs per day being posted makes me wonder what the long term trends are and how they have changed over time. Spencer 02:35, 1 February 2021 (UTC)
      • There's no way other than a manual collation of data, as far as I know. I think we are seeing more turnover as RDs are no longer going stale from being older than the last item. Rather anything in the last 7 days can now get posted if it's quality gets fixed, which seems to be happening more now, as people have more time to improve the biographies. Not sure what Amakuru thinks, as he's the one that's posted two bulk sets in the last couple of days? Stephen 02:41, 1 February 2021 (UTC)
  • Here you go with the requested data for 2020. At the cost of over-simplifying, to me it still continues to remain an outflow smoothening action. Start from the bottom. Post often. Do not batch. Ktin (talk) 03:42, 1 February 2021 (UTC)
Update 1: Updated the data with 2019 numbers as well. This gets a tad interesting. Clearly there is a jump between 2019 and 2020 and to me that could be attributable to a) more editors coming available with the lockdowns etc. and existing editors becoming more active during this period b) more effort being expended in improving articles as a community here (RD is one of the more collaborative projects that I have seen around here) and c) more RDs coming in due to the pandemic? Either way, it seems like our current situation should be compared to our handling of the outflow in 2020, we have handled high numbers then. I am still staying with my thinking that we stick to the basics. Start from the bottom. Post often. Do not batch. Thanks folks. PS: The number of articles being improved to homepage levels of hygiene continues to remain an absolute win for the community at large. We should not have any doubts or misgivings about that. Ktin (talk) 04:33, 1 February 2021 (UTC)
Expand to see the monthly RD postings data
Month RD Postings
January 2021 104
December 2020 107
November 2020 76
October 2020 76
September 2020 59
August 2020 61
July 2020 84
June 2020 55
May 2020 79
April 2020 95
March 2020 73
February 2020 50
January 2020 46
December 2019 42
November 2019 35
October 2019 46
September 2019 39
August 2019 54
July 2019 48
June 2019 28
May 2019 36
April 2019 37
March 2019 49
February 2019 57
January 2019 46
I'll have a closer look at Ktin's data later, thanks for producing that, but as ever if you want a change in practice you'll have to propose it and get consensus. Speaking just about recent turnover, I'm not convinced we have a particular problem right now. I uploaded 7 RDs on Saturday (5 of which were in a single batch in the morning, and I kept all 7 up until later in the evening), then another 6 last night. All of those ended up with 10 or more hours on the MP though, and most of the initial batch were up for a full 36 hours or so. There's no consensus at the moment for minimum posting times - we asked the community about this here and it was opposed - so I'm not sure what we should be doing differently. I don't agree with Ktin's instructions to "start from the bottom" or "avoid batching". how would that help? In general we should be posting these as soon as they're good to go, irrespective of whether it's a recent or older nom, that was what was mandated by the recent change in procedure. And in the case mentioned above there were no old RDs pushed off that shouldn't have been, so batching 5 at a time was the correct thing to do. Holding some of those back would have just risked the backlog getting full again later on. Cheers  — Amakuru (talk) 10:06, 1 February 2021 (UTC)
Amakuru, hello there. Quick update -- my statement was not meant to be a critique of your action(s) including the decision to batch up postings. I would argue that you did the right thing at that point. The larger point I was making, outside of your action, is that when we are attempting to smoothen (mathematical) the outflow function, a batch post will create a step function change, which theoretically might be alright if that batch post is done at the same time every day (think DYK posts) but not for the way we post here. It will create a situation where an article waited 24 hours to get into that batch but landed on the rightmost position and was immediately out because a single post came in right after, while all the other entries of the batch continued to stay. Cheers and thanks for all that you do. Ktin (talk) 16:30, 1 February 2021 (UTC)
@Ktin: yeah that's fine, and this all just friendly banter anyway. Re batch posting, no doubt you're right that it would be the correct thing to have them trickle through at a constant rate, sounds like a good thing to aim for. But we can only work with the circumstances at hand. I can't make any guarantees to be checking the (Ready) list multiple times a day. I might not even check it at all for several days; like everyone else I'm a WP:VOLUNTEER. And on another point, sometimes it happens that several of them become ready all at the same time because the reviewers also happen to do many all at once. So sometimes it will be necessary to batch them up. That's all really! Cheers  — Amakuru (talk) 17:35, 1 February 2021 (UTC)
Amakuru, agree with every sentence in the above statement. Cheers. Ktin (talk) 17:57, 1 February 2021 (UTC)
Fascinating data, Ktin, thank you for compiling. I think I always assumed "start from the bottom" to mean make sure to scroll all the way down to make sure that older ready noms weren't being ignored, since those sometimes are a little harder to see among all of those that have been posted. I wonder how much could be attributed to the relative ease of getting WikiCup points for an RD, but no judgement from me-- I think the process of article improvement to make something sufficient for RD posting quality has giving many (and based on the data, hundreds) of biographies up to snuff in terms of overall detail and referencing. I think the change to post anything in the past 7 days is suitable and probably accounts for the recent Dec/Jan uptick. I guess for now keep an eye on things and now be okay with ~24 hours on the template (sometimes more, sometimes less). Spencer 23:48, 1 February 2021 (UTC)
Spencer, Thanks for this note Spencer. Yup, in addition, the 'start from the bottom' ensures that RDs appear reasonably chronologically on the carousel, and more importantly when they get bumped off the carousel because of multiple reasons (including batch moves and or sequencing) that we are better off bumping off the older nominations from the carousel rather than the newer ones. I did not give Wikicup much thought, though, I personally did enroll as a participant, literally on the last day i.e. 1/31. I agree with you that the community efforts toward improving these many biographies is an absolute win for the project. The other thing that I really like about WP:ITNRD is that it is an absolutely collaborative space. Think of the number of times that someone nominates an article but someone else steps in on the article improvement actions. I have not seen many other projects display that collaborative spirit. Simple tweaks like what MSGJ made to the updaters field further increase collaboration. Now that we also got the permission to spill over briefly into the third row (by streamlining the COVID representation), I think we be WP:BOLD and afford articles as close to ~24 hours (sometimes a little more, sometimes a little less, but, close to 24 hours) just as an acknowledgement of the efforts that have gone in. Thanks for all that you and the team do! Ktin (talk) 00:14, 2 February 2021 (UTC)

Mini-digression

Can someone rip through the last 12 months of RDs and say how many of them were due to (or at least blamed on) Covid please? The Rambling Man (Stay alert! Control the virus! Save lives!!!!) 08:39, 2 February 2021 (UTC)

The Rambling Man, did this one for January 2021 and the result is 19; details here User:Ktin/sandbox/adhoc_queries#Number_of_RD_deaths_from_COVID_-_January. This one relies on the good folks at Deaths in 2021 coding all covid deaths. For the record, we have had 189 COVID deaths in January from amongst Deaths in 2021 Ktin (talk) 09:06, 2 February 2021 (UTC)
Thanks, 189 Covid deaths out of how many total deaths list in Deaths in 2021 in January? The Rambling Man (Stay alert! Control the virus! Save lives!!!!) 12:02, 2 February 2021 (UTC)
There are 963 lines in Deaths in 2021#January that start *
Ok, thanks. So that's probably a bit of a general increase at this time at RD because of Covid. It will pass (thankfully) once the vaccinations roll out properly. The Rambling Man (Stay alert! Control the virus! Save lives!!!!) 13:10, 2 February 2021 (UTC)
Anyone got month by month figures on how many deaths there have been in the past year, compared with 12-months previously? That might give you the overall effect of COVID. Bearing in mind that some of the "COVID death" people might have died anyway. Presumably the excess will eventually be cancelled out by lower averages in future as well.  — Amakuru (talk) 15:22, 2 February 2021 (UTC)
Using the same methodology as above:
Month 2018 2019 2020 2019 (%) 2020 (%)
January 743 789 795 6.19% 0.76%
February 597 663 739 11.06% 11.46%
March 658 664 879 0.91% 32.38%
April 577 622 1099 7.80% 76.69%
May 570 620 835 8.77% 34.68%
June 589 628 864 6.62% 37.58%
July 566 639 868 12.90% 35.84%
August 579 581 826 0.35% 42.17%
September 521 566 656 8.64% 15.90%
October 564 590 653 4.61% 10.68%
November 545 598 729 9.72% 21.91%
December 567 652 884 14.99% 35.58%


I'll see about making a graph later if anyone wants (I've not got time right now). However as every month in 2019 was greater than the equivalent month in 2018, I don't know that we can say that the whole increase is due to Covid. Thryduulf (talk) 17:56, 2 February 2021 (UTC)
@Thryduulf:, if you do not mind my doing, I edited your table and added percentage increases to the table above. Feel free to revert. That said, there are two components to the increase -- increased wiki activity by editors due to lockdowns etc., new category of deaths from covid. If you have the source files can I trouble you to do the following for the 2020 deaths, for i in *; do cat $i | grep -i "covid" | wc -l; done. This will give the number of covid deaths in each of those months assuming there is a file for each month. Cheers. Ktin (talk) 18:09, 2 February 2021 (UTC)
Your edits are fine. I don't have source files per se, I just copied the wikitext and did a find/replace in text editor that reports that number of changes. I can do as you ask relatively easily though, but it will take more time than I have right now. One thing to check though is whether the earliest deaths use the term "covid" as that name was not coined until 11 February according to Wiktionary's citations. Thryduulf (talk) 18:17, 2 February 2021 (UTC)
I'm not sure anyone said the "whole increase is due to Covid". I think it's a contributing factor. Also we have at least one regular contributor (Bloom6132) who does a very good job of updating and nominating RDs, often several a day. I don't think that was the case a couple of years back (although I might be wrong). And per Ktin, lockdown certainly afforded me a lot of extra time between April and July.... The Rambling Man (Stay alert! Control the virus! Save lives!!!!) 18:20, 2 February 2021 (UTC)
Per month deaths in 2020
All deaths Covid deaths Proportion Proportion of 2019-20 increase
January 795 4 0.50% 66.67%
February 739 7 0.95% 9.21%
March 879 133 15.13% 61.86%
April 1099 254 23.11% 53.25%
May 835 97 11.62% 45.12%
June 864 59 6.83% 25.00%
July 868 73 8.41% 31.88%
August 826 56 6.78% 22.86%
September 656 56 8.54% 62.22%
October 653 58 8.88% 92.06%
November 729 111 15.23% 84.73%
December 884 167 18.89% 71.98%
Thryduulf (talk) 20:21, 2 February 2021 (UTC)

COVID-19 banner

Boldy moved to this dedicated section from above "RD turnover"Bagumba (talk) 04:06, 25 January 2021 (UTC)

Comment the other thing we could do is get rid of the COVID-19 banner. It's been going on for over a year, everyone knows about it, and there aren't any major new developments: just the ebb and flow of infections, lockdowns and vaccinations. We get rid of the banner, link to Portal:Coronavirus_disease_2019 from the ongoing line (which won't go away anyway) and we free up a line for RDs or blurbs or whatever we want. The portal looks fantastic by the way, and has right at the top all the links we're featuring in the ITN box. It's a thought anyway, and if anyone thinks it has legs, we can spin up a subsection to discuss. --LaserLegs (talk) 02:13, 25 January 2021 (UTC)

I don't think the purpose had anything to do with any administration, on this global project. 331dot (talk) 09:58, 25 January 2021 (UTC)
Possibly you could convert the links to obscure redirects and count the page views to those redirects; would go against WP:MPNOREDIRECT though. Personally I think a link to the portal in ongoing is fine, as all the other links can be reached from there. So, support as proposed - Dumelow (talk) 16:17, 25 January 2021 (UTC)
Dumelow, wow! This is a simple but a brilliant idea! If we can test this out for even 48 hours, this is brilliant -- and will help test our hypothesis before removing! Ktin (talk) 16:21, 25 January 2021 (UTC)
  • The Rambling Man, unfortunately, I have a hit a roadblock in finding this data. I have asked at a few places including the VillagePump but have not received any leads. Referral data (if not pageclick) is standard webpage analytics information that I am absolutely sure is being collected. It is just not exposed via the regular pageview tool. So, if someone has any leads, I am happy to pursue there. Ktin (talk) 16:19, 25 January 2021 (UTC)
    If you find a proper way of getting that data, let me know as I'd be interested too. Cheers. The Rambling Man (Stay alert! Control the virus! Save lives!!!!) 16:27, 25 January 2021 (UTC)
    @Ktin: You might try asking at analytics@lists.wikimedia.org? Ed  16:20, 29 January 2021 (UTC)
    The ed17, thanks much. I think just this morning, we might have got a great lead here. The numbers largely are in line with what we found with the small experiment that Amakuru setup. Marking Andrew_Gray as well. Ktin (talk) 16:27, 29 January 2021 (UTC)
    @Ktin: increasing the size of DYK wouldn't help with the ITN balancing issue - it's TFA that it sits next to. Thus the only way to make space for extra RDs would be to remove the Covid box, reduce the number of normal ITN hooks, or somehow persuade the TFA guys to have a longer box (which would be outside of the scope of this project alone).  — Amakuru (talk) 16:43, 25 January 2021 (UTC)
    Surely we can't still be bound by 1990s HTML table rows? - Floydian  ¢ 06:00, 26 January 2021 (UTC)
  • Support, ongoing is good. --Tone 15:54, 25 January 2021 (UTC)
  • Support Switch to oppose for now, see below - in general this seems a good idea, per most of the above. It anyone else wants redirects set up as suggested by Ktin above, then I can do the legwork if necessary. Any such redirects would want to have full protection on them to guard against drive-by vandalism that might not be spotted as quickly as the actual articles. If the current conversation is going to wrap up too quickly for such data to be gathered though, I wouldn't have an issue. Having an Ongoing link to either the pandemic article or the portal would be OK.  — Amakuru (talk) 16:39, 25 January 2021 (UTC)
    Amakuru, Thanks for this offer Amakuru. Given that we have no opposes. Should we try Dumelow's suggestion given that 0:00 UTC is in a few hours? And, keep those redirects on for 48 hours. Ktin (talk) 19:11, 25 January 2021 (UTC)
    @Ktin: OK, we're ready to go. I have created a version of each link with a * on the end, e.g. COVID-19 pandemic*, and full protected them. Assuming there are no objections, I will switch over the actual links in the template in an hour or two, ready for the test to begin tomorrow. Cheers  — Amakuru (talk) 19:34, 25 January 2021 (UTC)
    Amakuru, Thanks Amakuru. Much appreciated. Ktin (talk) 21:39, 25 January 2021 (UTC)
  • Comment just an aside relating to bumping up the size of DYK, in the past, eight hooks seems to have been considered some kind of "golden number" so increasing to ten may well meet with some resistance. I'm not clear on it myself, one set of ten per day would work just fine and get some extra real estate on the right-hand-side of the main page. The Rambling Man (Stay alert! Control the virus! Save lives!!!!) 16:44, 25 January 2021 (UTC)
    I always understood "main page balance" to mean that TFA and ITN should roughly line up, as should DYK and OTD. And that's what I try to aspire to when fiddling with it. Thus increasing ITN wouldn't really help much with ITN issues. Perhaps there are others who don't care whether the pastel-coloured headings line up though as long as there isn't blank space at the bottom of all four boxes. I don't know if this issue has ever been codified properly.  — Amakuru (talk) 17:30, 25 January 2021 (UTC)
Can all 4 be lengthened a line? Sagittarian Milky Way (talk) 17:40, 25 January 2021 (UTC)
Perhaps, although it might not be that easy to make TFA longer. Probably you'd have to add some length to the blurbs, which would be something the honchos at that project would have to acquiesce to. Might not be a bad thing, as I think TFA blurbs are often a bit on the short side relative to the lead of the article they represent. Otherwise the only other way to do that would be to change the relative widths of the boxes.  — Amakuru (talk) 17:47, 25 January 2021 (UTC)
I guess we could resurrect the perennial 50/50 width split of the main page which is currently (IIRC) 55/45.... The Rambling Man (Stay alert! Control the virus! Save lives!!!!) 17:51, 25 January 2021 (UTC)
  • Oppose still the top news in many if not most countries in the world, with new developments almost every day. Developments that in normal times would be blurb-worthy (e.g. first curfew since WWII in the Netherlands). Especially - especially - considering that it would just free up space for the usual barrage of inconsequential news itms and RDs. 2A02:A451:8B2D:1:1108:8856:8418:AB93 (talk) 22:00, 25 January 2021 (UTC)
    Hardly. ITN is mostly criticised for having such a slow turnover, rather than a "barrage of inconsequential itms (sic)". And the RD discussion is currently seeking, at maximum, to add one line without compromising the news items. The Rambling Man (Stay alert! Control the virus! Save lives!!!!) 22:17, 25 January 2021 (UTC)
  • Support. Although I think we should consider what the "exit criteria" are for this item from Ongoing. Spencer 22:24, 25 January 2021 (UTC)
  • Comment - per discussion with Ktin above, I have now switched the links in the Covid box over to redirects with a * at the end in order to measure how much it is being used. I suggest that this discussion remain open at least until 27 January, so that the effects of that change can be measured for a full day on the 26th. Perhaps even including the 27th as well. Cheers  — Amakuru (talk) 22:25, 25 January 2021 (UTC)
    While not intentional, I worry that any conclusions will be at risk of confirmation bias, as we did not lay out what we were looking to prove beforehand. We will see X clicks from the mainpage. What is high or low for whatever we are trying to prove? Whatever the number, can we necessarily conclude that those same readers would not be able to navigate from a different organization of the MP or even outright removal?—Bagumba (talk) 04:28, 26 January 2021 (UTC)
    Bagumba, you bring up a fair point and I had given this some thought earlier today.
    Let's start off with the objective. We started this conversation, seeking an extra row (third row) for RD spillovers and to see if removal of the COVID-19 banner could grant that extra row for RD spillovers. While there is fatigue with that banner as it stands right now, there are two categories of people. Let's call them A and B. A = those who use the homepage to land on a COVID-19 article and B = those who use the homepage to land on an RD article. Now, we know we do not have instrumentation to measure / count people, so, we flip the measure and measure the number of referrals to a COVID article from the homepage (let's call this A') and number of referrals to an RD article from the homepage (let's call this B'). Based on a large amount of hand-waving and looking at the RD articles (and honestly nothing more scientific than staring at some graphs for over six months) I would say that a typical RD article (not the celebrity equivalents) get approx 2K page views from the homepage. This is our B'. Now, the tests will show if A' would be around this number or significantly higher than this number. If it is significantly higher, it makes sense to retain the COVID banner as-is. If it is ~2K-3K it might make sense to borrow this space for an extra row of RDs and move the highest contributed links onto Ongoing.
    Regarding your second question, that is more interesting. How do we say that the same user will not find other ways to reach the article once the homepage route is shut off for her? Short answer -- you can not say that definitively without putting in place user-level instrumentation. So, not even worth bothering with that problem.
    tl;dr: Inspect the daily page views on the redirects setup by Amakuru and if it is significantly greater than ~2K to 3K page views, leave them as-is. If not, reclaim that space, and move the two pages with highest page views onto the ongoing row.
    Thanks for all your patience, folks. Ktin (talk) 04:58, 26 January 2021 (UTC)
    The intention is purely to gather some data on how much these links are being used. There are no preconceptions with regard to this, and no fixed figures defined on which we might choose to act. The decision is still ultimately down to the consensus of editors in this discussion, but this just gives them something to work with. And even if it's shown that it has a million hits a day, if everyone still says "remove the template" then that's what we do. I haven't even decided the effect this might have on my own !vote but I thought it reasonable for Ktin to request this data.  — Amakuru (talk) 08:12, 26 January 2021 (UTC)
    There are no preconceptions with regard to this ... In the event my comment came off as an accusation, please understand that was not my intent. The numbers won't be available for a day or two. The more that we can discuss the possibilities before that, the more objective our observations can be. Cheers.—Bagumba (talk) 08:30, 26 January 2021 (UTC)
    If it is significantly higher, it makes sense to retain the COVID banner as-is.: Ktin, I'm not convinced it's that straight-forward. Clicks are not the main purpose of the Main Page, though it's certainly a factor. Thus, the direct clicks that the MP generates is not the end-all consideration. Readers presumably click RD links because they either did not know the person died or the link convenience spurred them to read more about the person. Those readers would not otherwise have come looking for the RD's bio. For COVID, we can assume that readers know that it's ongoing and have some understanding on the topic. It's conceiveable that if they hadn't found the COVID banner on the main page, a good percentage would still find an alternative path and get to the related pages, regardless. As you said, we cannot definitively say with the current metrics available. That's different than RDs, where you have witnessed readership increase ~2K the day it's placed on the MP. That uptick would be attributed to new discoveries from it being on the MP as opposed to planned visits looking for the topic.—Bagumba (talk) 08:21, 26 January 2021 (UTC)
    I think we need some kind of baseline if we're going to make informed decisions based on the clicks. We need to look at how many clicks TFA gets, how many clicks other news stories are getting. Taking the Covid clicks in isolation is going to be interesting but ultimately nothing more unless we can contextualise it. Even then, as Amakuru says, it's just to inform the community, there's no golden "clickbait number" above which the template should stay in place. The Rambling Man (Stay alert! Control the virus! Save lives!!!!) 08:23, 26 January 2021 (UTC)
    Comparing the figure with the total hits of the pages in question would certainly tell you something. For example if 20% of hits come from the links, that would suggest they're having a significant effect. Of course, we will also gain significant information after we switch to the new arrangement. We may notice a big drop in the overall hits for the affected pages. Or perhaps not.  — Amakuru (talk) 08:36, 26 January 2021 (UTC)
    I did a cursory look at recent items added/removed from Template:In the news/special-header, but it's hard to tell if the traffic changes for those specific items are tied more to the popularity of the respective link's topic or its placement on the MP.—Bagumba (talk) 08:42, 26 January 2021 (UTC)
  • Oppose portal as a target destination. I very much disagree that it "looks fantastic". Issues I notice just quickly glancing over it, several of which are significant:
    • The terrible machine-read autogenerated video which we removed from the article ages ago remains, and moreover is super prominent.
    • The disease progress box lists the cases/deaths/recoveries twice.
    • COVID-19 pandemic, the most important of all our COVID-19 pages, is not transcluded or otherwise prominently featured.
    • The Did you know box has a "poster shown" label with no poster actually shown.
    • The scrolling partial collapse of the reference box presents accessibility concerns.
I'm sure that additional issues would show up if I looked longer. And I don't want reopen the whole portal wars, but portals are just not how our readers expect information on Misplaced Pages to be organized, or what they seem to seek out (despite both being linked currently, the portal has gotten only 30k views in the past month, compared to over 1 million for the pandemic page, more than 30k per day. All of this is to say that I think linking COVID-19 pandemic would be a much better move than linking the portal. Linking the article is what we normally do for ongoing events anyways, and it contains prominent links in the lead to the other pages if anyone wants to seek them out. {{u|Sdkb}}02:48, 26 January 2021 (UTC)
  • Oppose for now. Covid remains the dominant news story in most countries around the world. Major media organisations still have multiple stories in the top few slots of their home pages, special sections devoted to it etc. We could think about trimming some of the lesser-used links in the box, but removing it entirely (or relegating to ongoing) seems inappropriate, especially if we would just be wasting the recovered space on more RDs. It's not as if we have a steady stream of blurbs that need to go up. Modest Genius 11:30, 26 January 2021 (UTC)
  • Oppose for now Covid has not lost its sui generis status among reliable source and remains an unprecedented event, as such, I think the banner still represents Misplaced Pages's reflection of that. --Jayron32 12:52, 26 January 2021 (UTC)
  • Support - ongoing is sufficient. Misplaced Pages should provide solid background information, not up-to-the-minute news, about the pandemic. The link target should be the main pandemic article, COVID-19 pandemic. This link will help readers find the information they're looking for about the pandemic. Levivich /hound 17:58, 26 January 2021 (UTC)
  • Question there is disagreement among supporters about whether to link the article or the portal, but would both be possible? Something like "COVID-19 pandemic "? Thryduulf (talk) 19:07, 26 January 2021 (UTC)
  • Comment. Preliminary Results are out here, and it seems the main pandemic page got 400 clicks from the banner, compared with over 40,000 for the page itself. So for me that suggests it won't be missed much if we axe it. I haven't checked the other pages yet, but no doubt someone can do so shortly.  — Amakuru (talk) 00:21, 27 January 2021 (UTC)
Axe now, ask questions later! InedibleHulk (talk) 00:42, 27 January 2021 (UTC)
  • Amakuru, not sure if there is a replication lag etc., I did a quick check for all the other links and all of them are in the same ballpark. i.e. ~1% of the traffic to those pages are coming from the homepage. For some reason I was expecting higher. Also, for some reason I was thinking the pandemic portal would get more clicks, but, apparently not. Anyway, if someone wants to bring a link onto the ongoing row, it might have to be the main pandemic article page. Also, if someone has friends on the pandemic project, might be worthwhile to ask them to consider a banner equivalent on the disease page, in case we do away with the banner. But, that is beyond our brief. Thanks everyone. Ktin (talk) 00:45, 27 January 2021 (UTC) Striking this out based on the updated numbers from Amakuru below. I think we had to wait for the numbers to populate / replication delays to catch up etc. Thanks. Ktin (talk) 13:46, 27 January 2021 (UTC)
  • Oppose it's in the news every day, everywhere around the world. We could have it in Ongoing I suppose, but at that point, why'd we make a banner in the first place? Banedon (talk) 00:59, 27 January 2021 (UTC)
It's about freeing up space for more RD or blurbs. Ongoing will be a permanent fixture with some poor quality article about low grade protests anyway, so we put COVID-19 in there and free up space in the box. That's it. Still on the main page, with room for more stuff too. --LaserLegs (talk) 01:16, 27 January 2021 (UTC)
  • Comment I don't care too much about the target, and I get the feeling none of the other supporters do either. If the main pandemic article is better then fine, lets use that. Seems to be fairly clear consensus, and the stats from Amakuru show that mostly no one cares about the banner anyway. --LaserLegs (talk) 01:19, 27 January 2021 (UTC)
  • Oppose. In addition to it still being a major news item worldwide, it's also the news item most likely to directly impact Misplaced Pages's readership. As much as I'd like to see more breathing room for RDs, making information about the pandemic immediately available to the living is more important than showcasing more articles of the dead. jSarek (talk) 09:45, 27 January 2021 (UTC)
  • Comment a quick look at the "redirect" stats shows that between 8% and 14% of traffic to these pages comes via main page clicks, which means that between 86% and 92% does not come from ITN. Pretty stark evidence that ITN is not really seminal in helping people find the information they're after. The Rambling Man (Stay alert! Control the virus! Save lives!!!!) 09:51, 27 January 2021 (UTC)
  • Comment again - (edit conflict) apologies, I was probably too hasty in looking at the stats last night. The figures look different this morning - 5,886 for the redirect, 39,076 for the page as a whole. Here are the full results:  — Amakuru (talk) 09:57, 27 January 2021 (UTC)
Article Article hits Redirect hits Percentage
COVID-19 pandemic 39,076 5,485 14.0
COVID-19 pandemic by country and territory 32,827 4,312 13.1
COVID-19 vaccine 21,633 1,788 8.3
Coronavirus disease 2019 19,513 967 5.0
Severe acute respiratory syndrome coronavirus 2 6,256 576 9.2
Portal:Coronavirus disease 2019 760 789 103.8
Impact of the COVID-19 pandemic 713 598 83.9
Thanks for those numbers. That's a decent amount of traffic, more than most RDs get. I think that justifies keeping at least the banner entries that are getting a thousand clicks per day. Modest Genius 14:38, 27 January 2021 (UTC)
Nah, main page get s more than six million hits per day, so this is peanuts. The Rambling Man (Stay alert! Control the virus! Save lives!!!!) 14:44, 27 January 2021 (UTC)
2-5k per day is more than a typical RD or DYK entry. Modest Genius 16:28, 29 January 2021 (UTC)
Modest Genius, Not really. We have just reasonably seen that an RD gets ~2-3K hits per day. We should definitely carry COVID-19 pandemic to the ongoing line. Where I am on the fence is COVID-19 pandemic by country and territory. If there is a way to show it as Ongoing: COVID-19 pandemic (by Geography), or something similar, that might be a good median outcome. Ktin (talk) 16:35, 29 January 2021 (UTC)
  • Comment I maintain my position to remove the banner and simplify to a link. Everyone is aware of COVID and has some working knowledge of it. No reason to believe anyone looking for information would not find it without the banner. Contrast with blurbs or RDs, which are likely new news for a lot of readers. Based on the numbers, I'm ok with "Ongoing" linking to COVID-19 pandemic instead of the portal.—Bagumba (talk) 11:33, 27 January 2021 (UTC)
  • information Administrator note was considering if this was ready for closure, but think it needs more opportunity for discussion - I've invited in editors from Talk:Main Page. — xaosflux 00:47, 28 January 2021 (UTC)
  • Comment Reading through these, and I notice that there are many !votes cast apparently with the notion that the banner is going to be removed with nothing to replace it. It should be clear here that the Main Page will still have a link to COVID-19; it will reside in the Ongoing section of ITN, where it will require a nomination and strong consensus to remove (i.e., it will not "roll off").130.233.213.199 (talk) 08:41, 28 January 2021 (UTC)
  • Admin Action Requested. Can I request an Admin to evaluate consensus / determine next steps? The latest three day traffic report can be accessed here. Seems like the numbers posted by Amakuru are holding. I personally was surprised by the low number of clicks, but, it is what it is. Ktin (talk) 08:49, 29 January 2021 (UTC)
  • Support. We can just put "COVID-19 pandemic" at Ongoing. Making a dedicated banner makes it seem like we're prioritizing the pandemic over others. It doesn't advantage readers, anyway. GeraldWL 09:50, 29 January 2021 (UTC)
  • Support removing the banner and having a link to COVID-19 pandemic in Ongoing. the wub "?!" 12:54, 29 January 2021 (UTC)
  • Removed and moved to ongoing, as I interpret the consensus here. Please check. --Tone 16:33, 29 January 2021 (UTC)
  • Comment. For folks who are interested in any sort of analysis, we managed to find a way based on feedback here here to get the raw referral numbers for the month of December. It is available only monthly (and not daily) but, has great information re: referral data from the mainpage. Feel free to ping me if you'd want any query checked. Ktin (talk) 22:04, 29 January 2021 (UTC)
mysql> select * from december_views where referer = 'Main_Page' order by pageviews desc limit 10;
+-----------+--------------------------------------------+-------+-----------+
| referer   | target                                     | type  | pageviews |
+-----------+--------------------------------------------+-------+-----------+
| Main_Page | Hyphen-minus                               | other |   2951391 |
| Main_Page | Lists_of_deaths_by_year                    | other |   1130422 |
| Main_Page | Misplaced Pages                                  | link  |    214705 |
| Main_Page | COVID-19_pandemic                          | link  |    159008 |
| Main_Page | COVID-19_pandemic_by_country_and_territory | link  |    120652 |
| Main_Page | Great_conjunction                          | link  |    106300 |
| Main_Page | Tigray_conflict                            | link  |     81721 |
| Main_Page | 2020_Indian_farmers'_protest               | link  |     78387 |
| Main_Page | Encyclopedia                               | link  |     71174 |
| Main_Page | Hayabusa2                                  | other |     62045 |
+-----------+--------------------------------------------+-------+-----------+
Seems like the great conjunction was the most clicked non-corona ongoing story of the month. Sagittarian Milky Way (talk) 00:04, 30 January 2021 (UTC)
Sagittarian Milky Way, It seems to have been a story that a lot of our readers have clicked on. I will add one caveat, the pageviews are monthly and not average daily, so, for articles that rotate, they have not been normalized for number of days spent on the homepage. Ktin (talk) 01:28, 30 January 2021 (UTC)
What's up with the hyphen? Sagittarian Milky Way (talk) 01:43, 30 January 2021 (UTC)
- is literally a redirect I wonder if it's from collapsing tables or something --LaserLegs (talk) 02:29, 30 January 2021 (UTC)
Calls to {{IsValidPageName}} perhaps. - Floydian  ¢ 04:19, 31 January 2021 (UTC)
  • @Ktin: would you "select sum(pageviews) as sum_pageviews from december_views group by `type` where target = 'COVID-19_pandemic'"? I'm interested in the number of link referrals vs external (which my grep-fu counted up to around 700k). Else I can just load it into some RDBMS but since you've already done it.... --LaserLegs (talk) 02:27, 30 January 2021 (UTC)
    LaserLegs, done. Pinged you here. Ktin (talk) 02:59, 30 January 2021 (UTC)
  • Drop all but main parent article. Thank you Yassen for the email about this....was looking for days.... was under the wrong impression the portal was up for deletion. That said. ....and as the portal maintainer - I suggest we link the main pandemic article over the portal every time. Yes they both present roughly the same info all transcluded, but best to link to information presented in a normal format that are readers are familiar with and has navigational aids like a Toc. Usability, functionality and accessibility shoild be our primary concern when choosing a linked page thus not a portal or oddly formated page. I take care of the portal simply to make sure it's mobile accessible (as the majority of our portals are not) but a normal formated page is always best because of reader familiarity and its were they have access to verification by way of sources.--Moxy 🍁 02:54, 30 January 2021 (UTC)
  • Oppose This decision strikes me as hasty, localized and fairly unbiased. Omitting a banner that contextualizes an unprecedented, historic global event on Misplaced Pages that continues to strike with second and third waves in various countries and has seriously reshaped economies and lifestyle. This is well above the relevance policy bar that Misplaced Pages prays, and rewrites a new standard for handling event on a global scale. The banner itself had to be restructured to better illustrate the scenario. I think that, above any sudden preference to remove the banner, to give a localized group consensus right now is to counter a huge amount of news, policies and scenarios that are updated at all hours about COVID-19. We are not talking about a brief event to bring it to consensus yet. Also, the article is massive to be referred to in a single link, especially in a mobile experience. The direct links that the banner contained helped to direct to the different articles with enormous information made by editors. --JungleWiki (talk) 05:30, 30 January 2021 (UTC)
    This decision strikes me as hasty: For the record, there was 4.5 days of discussion before it was pulled.—Bagumba (talk) 08:02, 3 February 2021 (UTC)
  • Comment - it's been several days since there was action in this thread, but I think given the stats that have been shown up by the special asterisked versions of the links, that there is evidence the main page links are still useful. The disease, the pandemic and the country-by-country breakdown links are all getting tens of thousands of hits per day. I'm therefore amending my !vote above to Oppose.— Preceding unsigned comment added by Amakuru (talkcontribs) 15:27, 2 February 2021 (UTC)
    @Amakuru: COVID-19 pandemic page views have actually gone up since the special header was removed on Jan 29. Not sure we can conclude that the people looking for COVID-19 pages aren't finding them because of the change. What is your analysis?—Bagumba (talk) 08:19, 3 February 2021 (UTC)
    @Bagumba: urggh... I hadn't even noticed that the proposed change had already been enacted. I assumed the conversation had died a death for lack of consensus! As JungleWiki says below, this was done way too hastily, particularly as we'd been actively analysing the data. Seven days would be the normal period to let a survey like this run, as we would at AFD or RFCs. To answer your question, my source for saying the links were accessed tens of thousands of times per day was Ktin's link to the page-view stats above, based on the special asterisked versions we set up. That showed that a significant percentage of all hits for the affected articles were coming from the main page links. Anyway, since the change has now been enacted, we have the opportunity to actually see the real effect of the change. I've run a like-for-like comparison of daily averages between the three days from 23 Jan to 25 Jan (with the banner in place), against the same three weekdays one week later when it was gone (30 Jan to 1 Feb). Results are:
Results of week on week hits comparison
Article Previous hits New hits Change % Change
COVID-19 pandemic 38,879 40,118 1,239 +3.2
COVID-19 pandemic by country and territory 30,897 23,353 -7,544 -24.4
Coronavirus disease 2019 21,888 15,410 -6,478 -29.6
COVID-19 vaccine 21,193 18,713 -2,480 -11.7
Severe acute respiratory syndrome coronavirus 2 6,251 5,223 -1,028 -16.4
Portal:Coronavirus disease 2019 1,215 506 -709 -58.4
Impact of the COVID-19 pandemic 1,019 470 -549 -53.9
TOTAL 121,341 103,792 -17,549 -14.5
  • So yes, although there has been a very slight uptick in the COVID-19 pandemic views, this has been more than offset by a large loss of views on all the other affected pages. Across all of them it's a 14.5% decline in views. I'm aware that initially I was in favour of dropping the box, but it's clear from the stats and the !votes that keep coming in opposing the change, that this was a mistake and the box should be reinstated ASAP. Cheers  — Amakuru (talk) 10:55, 3 February 2021 (UTC)
    Amakuru, while I do not have any preference on what is done here eventually, the metric that you are using 14.5% is not the right one to check. By the law of small numbers, the change as a percentage is always going to look big. Bottomline is that the based on the above data, the cumulative COVID banner was resulting in ~17K views a week (now showing up as nett lost traffic; nett = traffic lost  – traffic made up from other avenues e.g. ongoing link). This translates to approx 2.5K page views per article day. This is right about at or lower than any other typical link on homepage including a average RD (non celebrity) which gets ~4.5K. So, from a numbers standpoint you are serving more readers by creating space by removing the banner (and taking the high traffic articles onto an ongoing link). However, if Admins do not use this space by either using it for a news blurb or for additional RDs spill over (and using the space for expression of articles), you are better off retaining the COVID banner. Because, something is better than nothing. Good luck. Ktin (talk) 03:16, 4 February 2021 (UTC)
    @Ktin: i don't really follow your logic. These are not particularly small numbers, and in any case it's clearly the % that's important here. If the pages were generally getting 1 million views per day, then 17,000 more or less is small change and wouldn't be significant. But with the actual numbers we have, that 17,000 is a big drop off and I think we've definitely done our readers a disservice by removing it. COVID is and will remain the overriding story of our time for the foreseeable future, and getting our readers to the relevant parts of that, in particular NPOV scientific information on the vaccine, is really quite important right now. If that comes at the expense of stories which are more than a week old, and basically forgotten, such as Kaja Kallas becoming PM of Estonia, then that's a good thing. @Tone: is any chance you could reconsider your decision to remove the box, because this really looks to have been the wrong move. Cheers  — Amakuru (talk) 22:55, 4 February 2021 (UTC)
    Amakuru, let me explain in a different way, and see if that helps. Again, I will start this off with a caveat that I do not have a preference for what is done on this specific topic, eventually.
    When you are measuring links on the homepage and their effectiveness, you are looking at how valuable are those links for what someone is looking for or for where they go next. One way to do that is by looking at what folks are clicking at. Now, the way you would do that is by measuring "how many people are clicking on a link on the homepage" and then asking the question "what fraction or % of the home page users are clicking on a particular link". Just by way of data, currently the COVID linked articles in that banner are (or were) getting ~2.5K clicks per day. That is lower than the other category of links on the ITN box / homepage.
    Regarding your later points on COVID being the zeitgeist of the current times and how presenting NPOV information is important, I absolutely don't deny that. That is a qualitative feedback (opinion) independent of what the data on clicks say. It's not data driven, but, it is definitely an opinion that merits independent consideration. It boils down to 2.5K clicks (but more meaningful, per our own subjective opinion) more important than 4.5K clicks on general additional information. I am definitely not making the case that one category of clicks is more important than the other.
    What I am definitely against is, the current state, where the banner was removed to make space for "blurbs and RD," but then Admins refusing to use up even more than 25%-50% of row #2. So, space was created by removing the banner but, is being used for real whitespace. Good luck. Ktin (talk) 23:31, 4 February 2021 (UTC)
    Ktin: The !votes were generally not tied to a specific use of the freed banner space. AFAICS, the main page is now operating as it did before COVID. WP:ITNRD and WP:ITN/A remain the same. Admins are not "refusing" to do anything, except perhaps IAR in a way you might prefer them to. Can you elaborate on the implication that "25%-50% of row #2" is unused? —Bagumba (talk) 01:21, 5 February 2021 (UTC)
    Bagumba, my disenchantment stems from yesterday's posting Misplaced Pages:In_the_news/Candidates#(Posted)_RD:_Charan_Gill as with a few other. The article fell off the carousel ~14 hours when ~50-75% of row 2 was available for the article could have stayed on (for at least a few more hours) without even spilling into row #3.
    In my view, this conversation regarding removal of COVID banner was an offshoot of a conversation that intended to find space (i.e. an additional row) for "blurbs or RDs". Ktin (talk) 02:02, 5 February 2021 (UTC)
    @Ktin: Some have stated that 12h is enough time, on par w/ DYK. It's not clear where consensus is re: 24h. As for space on row 2, it's probably dependent on one's screen resolution. I have a 2560x1600 screen, and if I look at the MP using the ITN template at the time Gill was removed, the second row is 80% full, and the second column is already slightly longer and "unbalanced" from the first column.—Bagumba (talk) 12:43, 5 February 2021 (UTC)
    I don't particularly care about RD tenure, I think 12h is more than enough, but that was one of the rationales for this removal. However, the bigger point was that people felt it wasn't useful any more. But the stats we have seen show that they emphatically were being used and frankly to suddenly remove this prominent link for no discernible gain, when we're bang in the middle of a second (or third/fourth/fifth depending on your viewpoint) wave of the disease, coupled with a worldwide vaccine roll-out, makes us look quite tone-deaf frankly.  — Amakuru (talk) 13:28, 5 February 2021 (UTC)
    It's amazing how times change. It was only a few years ago that about 95% of these RDs would never get onto the main page, let alone have the luxury of 12 hours (or more) of exposure. Easy to forget. The Rambling Man (Stay alert! Control the virus! Save lives!!!!) 13:36, 5 February 2021 (UTC)
    Amakuru, I will start off by saying where I do not disagree with you. I do not disagree about the relevance of the COVID topic in a world as we have it today. However, I do not think it is right to hold my insistence for a 24 hour consideration window for RDs as the reason why the banner has been or should be removed. Quoting from my post earlier I have been consistent in my request for a third row for RDs to spill over onto. However, I do not believe that this should come from the COVID banner (at least until we see the data). My point is that, having removed the banner, not using that whitespace, is bad. Regarding your statement that the stats show usage 'emphatically' that this banner was being used. Not really. Just as a summary, there are 6M daily views on the homepage, of which 1.1M views result in some sort of a click, of which the banner articles attracted 2.5K clicks on an average. But, that said, the argument (zeitgeist etc.) sans the stats is a reasonable argument to advance. Ktin (talk) 14:18, 5 February 2021 (UTC)

Election head of state / government simplification

Suggest overhauling the heads of state/government section to be as follows:

Changes to the leaders whose offices constitutionally administer the executive of their respective state/government as listed at List of current heads of state and government when:

  • The leadership changes as the result of an election
  • The succession of leadership when not the result of an election

Changes to other heads of state or government who do not constitutionally administer the executive of their respective state/government are discussed on their own merit.

Feel free to clean up my poor grammar as necessary. --LaserLegs (talk) 20:52, 26 January 2021 (UTC)

Presumably this is an anti-powerless or low powered head of state without fame rule, not aimed at Charles. He would be posted quick whether automatic or not. Sagittarian Milky Way (talk) 00:42, 27 January 2021 (UTC)
"Changes to other heads of state or government who do not constitutionally administer the executive of their respective state/government are discussed on their own merit." if you really think we'd not post the succession of the British royal family, I'm not sure what to do for you. Also that happens what? Once every 5 or 6 decades? Why is this obvious change so controversial? --LaserLegs (talk) 01:15, 27 January 2021 (UTC)
I'm saying the opposite, the British monarch abdicating would be posted quick whether there's an automatic rule or not. Sagittarian Milky Way (talk) 02:27, 27 January 2021 (UTC)
And Queen Elizabeth is not the head of state of just one country. Of course it would be posted. I disagreed with removing head of state from ITNR, but it's done, and that does not mean they can't be posted, just that it isn't automatic. 331dot (talk) 01:25, 27 January 2021 (UTC)

Question: What happens to legislative elections that do not lead to changes in whoever is in the List of current heads of state and government? Examples are US and Philippine midterm elections, French and Indonesian legislative elections and Japanese House of Councillors elections. One could argue that the US legislative elections, including the one that led to a runoff in Georgia are no longer ITNR. Are those still ITNR, or ITN will pick whoever has the more white dudes in a country east of the Atlantic (let's face it this is how it'll boil in to)? Howard the Duck (talk) 15:06, 29 January 2021 (UTC)

General elections are ITNR irrespective of the head of government. The vast majority of changes to heads of government were posted with an election, but the recent change covers occasions where the occupant of a position changes outside of an election.(such as Theresea May to Boris Johnson) 331dot (talk) 15:32, 29 January 2021 (UTC)
It can be argued that some of those elections are not general elections. We have ITN regulars opposing US midterm elections (LOL). You could argue that the Japanese House of Councillors elections are not general elections (none have been posted since 2010). I believe French Senate elections are of similar nature but that is an indirectly elected body unlike the Japanese House of Councillors. Howard the Duck (talk) 15:38, 29 January 2021 (UTC)

December at ITN

Sorry for the spam, but, thought this might be interesting to the curious minded. This is a view of articles that we discussed in WP:ITN / WP:ITNRD in December and their relative interest on the mainpage. Thanks to Andrew Gray for his leads toward getting to the source of this information. Ktin (talk) 04:44, 31 January 2021 (UTC)

Caveat 1: This is a monthly aggregate view and hence is not normalized for time spent on the mainpage. E.g. one article could have spent a couple of hours while other could have spent days on the mainpage.

Caveat 2: These are articles that were introduced for discussion in December. Does not include articles that were discussed in late November, but, spent time on the mainpage in the early part of December. Similarly, does not include articles that were discussed in December but, landed on the mainpage in early January.

Click 'show' to expand / see December data
mysql> select * from december_views where target in (select article from itn_list) and referer = 'Main_Page' order by pageviews desc;
+-----------+--------------------------------------------------+-------+-----------+
| referer   | target                                           | type  | pageviews |
+-----------+--------------------------------------------------+-------+-----------+
| Main_Page | Great_conjunction                                | link  |    106300 |
| Main_Page | 2020_Indian_farmers'_protest                     | link  |     78387 |
| Main_Page | Hayabusa2                                        | other |     62045 |
| Main_Page | Valéry_Giscard_d'Estaing                         | other |     60510 |
| Main_Page | Chang'e_5                                        | other |     57153 |
| Main_Page | 2020_World_Rally_Championship                    | other |     31841 |
| Main_Page | Charley_Pride                                    | other |     28819 |
| Main_Page | 2020_Ghanaian_general_election                   | other |     25883 |
| Main_Page | Stella_Tennant                                   | other |     25735 |
| Main_Page | John_le_Carré                                    | other |     23500 |
| Main_Page | Jeremy_Bulloch                                   | other |     22165 |
| Main_Page | Awesome_Again                                    | other |     21405 |
| Main_Page | Paolo_Rossi                                      | other |     20369 |
| Main_Page | Maria_Piątkowska                                 | other |     16874 |
| Main_Page | Doug_Anthony                                     | other |     16625 |
| Main_Page | Fanny_Waterman                                   | other |     14994 |
| Main_Page | Karima_Baloch                                    | other |     14896 |
| Main_Page | John_Barnard_Jenkins                             | other |     14654 |
| Main_Page | Brodie_Lee                                       | other |     14517 |
| Main_Page | George_Blake                                     | other |     14460 |
| Main_Page | Gérard_Houllier                                  | other |     14050 |
| Main_Page | Jerry_Relph                                      | other |     14032 |
| Main_Page | Flavio_Cotti                                     | other |     13852 |
| Main_Page | Soedardjat_Nataatmadja                           | other |     13191 |
| Main_Page | Jack_Steinberger                                 | other |     13166 |
| Main_Page | Òscar_Ribas_Reig                                 | other |     12531 |
| Main_Page | Robin_Jackman                                    | other |     12320 |
| Main_Page | 2020_Petrinja_earthquake                         | link  |     12287 |
| Main_Page | Luke_Letlow                                      | other |     12062 |
| Main_Page | Doug_Scott                                       | other |     12002 |
| Main_Page | Harold_Budd                                      | other |     11976 |
| Main_Page | Barry_Lopez                                      | other |     11956 |
| Main_Page | Walter_E._Williams                               | other |     11838 |
| Main_Page | Huang_Zongying                                   | other |     11727 |
| Main_Page | Phil_Niekro                                      | other |     11635 |
| Main_Page | James_Flynn_(academic)                           | other |     11589 |
| Main_Page | Alison_Lurie                                     | other |     11523 |
| Main_Page | R._N._Shetty                                     | other |     11516 |
| Main_Page | Totilas                                          | other |     11474 |
| Main_Page | Iman_Budhi_Santosa                               | other |     11271 |
| Main_Page | Ann_Reinking                                     | other |     11249 |
| Main_Page | Michael_Jeffery                                  | other |     11178 |
| Main_Page | Jim_McLean                                       | other |     11141 |
| Main_Page | Mohamed_Abarhoun                                 | other |     10761 |
| Main_Page | Betsy_Wade                                       | other |     10602 |
| Main_Page | Donald_Fowler                                    | other |     10549 |
| Main_Page | Paul_Sarbanes                                    | other |     10210 |
| Main_Page | Narinder_Singh_Kapany                            | other |     10192 |
| Main_Page | Maria_Itkina                                     | other |      9668 |
| Main_Page | Roddam_Narasimha                                 | other |      9279 |
| Main_Page | Brian_Kerr,_Baron_Kerr_of_Tonaghmore             | other |      9255 |
| Main_Page | Astad_Deboo                                      | other |      9133 |
| Main_Page | Martin_Lambie-Nairn                              | other |      9042 |
| Main_Page | Romell_Broom                                     | other |      9030 |
| Main_Page | James_Odongo                                     | other |      8915 |
| Main_Page | Gerard_Stokes                                    | other |      8802 |
| Main_Page | Maria_Fyfe                                       | other |      8657 |
| Main_Page | K._C._Jones                                      | other |      8562 |
| Main_Page | Barbara_Windsor                                  | other |      8544 |
| Main_Page | Roger_Moret                                      | other |      8379 |
| Main_Page | Barbara_Rose                                     | other |      8365 |
| Main_Page | Bill_Fitsell                                     | other |      8313 |
| Main_Page | Adele_Rose                                       | other |      7840 |
| Main_Page | George_Ross_Anderson_Jr.                         | other |      7715 |
| Main_Page | Minoru_Makihara                                  | other |      7659 |
| Main_Page | Udyavara_Madhava_Acharya                         | other |      7507 |
| Main_Page | U._A._Khader                                     | other |      7464 |
| Main_Page | Benjamin_Abeles                                  | other |      7422 |
| Main_Page | Ferenc_Tóth_(politician)                         | other |      7176 |
| Main_Page | Dawn_Wells                                       | other |      7114 |
| Main_Page | Fou_Ts'ong                                       | other |      7029 |
| Main_Page | Yuichiro_Hata                                    | other |      6988 |
| Main_Page | Kim_Ki-duk                                       | other |      6685 |
| Main_Page | Elaine_McCoy                                     | other |      6650 |
| Main_Page | Manglesh_Dabral                                  | other |      6550 |
| Main_Page | Paul-Heinz_Dittrich                              | other |      6468 |
| Main_Page | Raymond_Hunter                                   | other |      6465 |
| Main_Page | Hugh_Keays-Byrne                                 | other |      6418 |
| Main_Page | Jutta_Lampe                                      | other |      6411 |
| Main_Page | Joseph_Bachelder_III                             | other |      6329 |
| Main_Page | Pumza_Dyantyi                                    | other |      6282 |
| Main_Page | Tabaré_Vázquez                                   | other |      6152 |
| Main_Page | Muhammad_Mustafa_Mero                            | other |      6137 |
| Main_Page | Dharampal_Gulati                                 | other |      5894 |
| Main_Page | Belinda_Bozzoli                                  | other |      5595 |
| Main_Page | Herman_Asaribab                                  | other |      5585 |
| Main_Page | Lorraine_Monk                                    | link  |      5420 |
| Main_Page | Amelia_Lapeña-Bonifacio                          | other |      5148 |
| Main_Page | Howard_J._Rubenstein                             | other |      4789 |
| Main_Page | Zafarullah_Khan_Jamali                           | other |      4630 |
| Main_Page | Shamsur_Rahman_Faruqi                            | other |      4509 |
| Main_Page | Jack_Lenor_Larsen                                | other |      4426 |
| Main_Page | Sugathakumari                                    | other |      3908 |
| Main_Page | MF_Doom                                          | other |      3407 |
| Main_Page | Brandon_Bernard                                  | other |      2951 |
| Main_Page | Hasu_Yajnik                                      | other |      1291 |
| Main_Page | Joseph_Safra                                     | other |      1262 |
| Main_Page | Garry_Runciman,_3rd_Viscount_Runciman_of_Doxford | other |       836 |
| Main_Page | Kevin_Greene                                     | other |       784 |
| Main_Page | Tommy_Docherty                                   | link  |       766 |
| Main_Page | Aldo_Andretti                                    | other |        23 |
| Main_Page | H._Jack_Geiger                                   | other |        15 |
| Main_Page | John_Fitzpatrick                                 | other |        14 |
| Main_Page | Ronald_Atkins                                    | other |        13 |
+-----------+--------------------------------------------------+-------+-----------+
104 rows in set (0.02 sec)

Cool, and surprising that Great conjunction topped the list, a nom which had a fair amount of opposes on notability.130.233.213.199 (talk) 06:34, 3 February 2021 (UTC)

Personally I think it shows a disconnect between what editors think is notable, and what actually draws reader interest. - Floydian  ¢ 06:58, 3 February 2021 (UTC)
Huh, people like space. Cool. 🚀  Nixinova T  C   07:05, 3 February 2021 (UTC)
@Ktin: I only did a cursory read of m:Research:Misplaced Pages clickstream, but wouldn't we be interested more in type=link i.e. links on the MP that are clicked to get to the target, and not type=other?—Bagumba (talk) 07:55, 3 February 2021 (UTC)
@Bagumba: I was originally a bit puzzled by this as well, but it seems this is an artefact of the way the data's generated - it defines link/other on the basis of a single point in time, which seems to be the end of the month. Makes sense for most cases (page to page links are reasonably stable) but gives odd results for things like the main page that rotate frequently.
There seems to be a background level of "main page clicks" for items which aren't linked - you can see it above with Ronald Atkins, for example, which wasn't on the main page until January. These probably reflect people coming to the main page and then using the search box; if they go straight to the target page it'll be logged as an "other" click to that page. I ran the numbers for some "mundane but popular" articles that weren't apparently linked from the MP in this period (eg Roman Empire, The Beatles), and found that their mainpage clicks weren't usually more than about 5% of the recorded internal clicks.
By comparison, 69% of the internal clicks for Charley Pride and 67% for Paul Sarbanes came from the mainpage, so it seems reasonable to conclude that the majority of those do represent people hitting the link while it was posted. Andrew Gray (talk) 22:09, 3 February 2021 (UTC)
Andrew Gray, nicely explained. tl;dr – when looking at the above table, ignore the 'type' column. Cheers. Ktin (talk) 03:21, 4 February 2021 (UTC)

What's up with ongoing

Second attempt to remove the Indian farmers' protest. There are five opposes but not a single one of them addresses the staleness of the article. The Misplaced Pages:In_the_news#Ongoing_section criteria is clear "In general, articles are NOT posted to ongoing merely because they are related to events that are still happening. In order to be posted to ongoing, the article needs to be regularly updated with new, pertinent information." By that criteria, none of the opposes are actually valid. Has WP:VOTE become a substitute for WP:CONSENSUS? Or should the criteria be amended to reflect what has become practice: stale articles festering in the box for weeks or months past their Best by date while a dedicated cabal resists any effort to remove the article from the main page while simultaneously contributing nothing to maintain the article at a standard that would merit featuring.

I'm only using the farmers protest as an example, but there are many others such as the CAA protests, the Belarusian protests, the Venezuelan protests, the Hong Kong protests in each of these cases prying a demonstrably stale article out of the box proved to be nearly impossible. --LaserLegs (talk) 15:55, 6 February 2021 (UTC)

Category: