Wikipedia talk:Good articles/GAN Backlog Drives/March 2021

Notifications

 * Eddie stated on my talk page that he sent notifications to everyone on the GA mailing list. I am going to go through the participants of the last drive and notify anyone who wasn't on the mailing list. (t &#183; c)  buidhe  04:19, 22 February 2021 (UTC)
 * There's a GA mailing list? How would I go about adding myself to that? Hog Farm Talk 04:31, 22 February 2021 (UTC)
 * It's WikiProject Good articles/newsletter participants. Thanks for participating in the drive! I have now notified everyone who participated in the last drive and hasn't already signed up. (t &#183; c)  buidhe  04:50, 22 February 2021 (UTC)

Day 18 tally

 * , I could be totally missing something, but are the numbers for 18 March (479 outstanding; 339 unreviewed) correct? At present, the nominations page gives the numbers as 484 (+5) and 347 (+8), but Legobot has run only twice since midnight, and added only two articles (JCB (song) and Geoffrey Massey). --Usernameunique (talk) 03:09, 18 March 2021 (UTC)
 * Usernameunique, it's a good question; they are correct. Here's why. There are two places to get the numbers: the Nominations page or the Reports page. I started out this drive using the Reports page, because the software there gets its numbers by parsing/counting the nominations and reviews on the GAN page at 01:00; it's a simple matter to back off from 01:00 to midnight, and doing diffs lets you know exactly when the new noms or new reviews were initiated. There have been times (maybe not tonight, but earlier in the drive) where you get there hours (or even days!) late; it isn't feasible to reconstruct numbers when that happens. The problem with the Nominations page is that it uses categories to base its numbers from (via template GAN counter), and there are a number of false positives included in the totals. Tonight, as you can see on the Reports page, the 01:00 numbers were 480 and 340; removing the sole new nomination since midnight (Geoffrey Massey at 00:35) takes the total down to 479 and the total unreviewed down to 339. (When I just looked at Category:Good article nominees, it had 485 total nominations, and in the categories, it lists 348 waiting for a review and 141 under review (which includes all review/on hold/2ndopinion noms), which totals 489 rather than 485. I've never figured out what's causing the discrepancy in these category totals, but it's why I trust the Report numbers to be more precise and use them when I get the chance. As far as the interim daily numbers, it's not going to be too far off if sometimes the Nominations page is used, and I think there was one past drive where I managed to get the Nominations page consistently for all 30 days, but for this drive and the last one I've taken the beginning and end numbers from the Reports page. BlueMoonset (talk) 05:32, 18 March 2021 (UTC)
 * Thanks for the detailed explanation, ; that makes a lot of sense. The discrepancy is odd, but in any event (and as you say), it's better to work with a consistently generated number than with a number that is subject to change every 20 minutes, with no archive. --Usernameunique (talk) 00:16, 19 March 2021 (UTC)
 * Usernameunique, I checked all three categories by copying the lists and then counting them. There was one in the nominees category that was caused by a category manually added to an article talk page back during its GAN review in 2015, but otherwise the number of entries were what I expected (485); the same for for the 141 under review. The number of nominees awaiting review was off, however: the total given was 348, but there were only 343 actual entries. I don't know why there's a discrepancy there of five entries in the count, but it's important because the Nominations page currently shows five higher than reality when it comes to "waiting for review". What are these five phantoms? BlueMoonset (talk) 01:42, 19 March 2021 (UTC)
 * , I agree that the number is off by five. The nominations page ultimately gets its number from the number of articles in Category:Good article nominees awaiting review, through the following steps:
 * The numbers at the top of WP:GAN are transcluded from WP:Good article nominations/guidelines.
 * WP:Good article nominations/guidelines takes its number of unreviewed nominations from the template "GAN counter|Wait".
 * "GAN counter|Wait" takes its number from the template "PAGESINCATEGORY:Good article nominees".)
 * The number in this category is off by five: it currently says that there are "approximately 342 total" pages in the category, but lists only 337 (page 1: "The following 200 pages are in this category, out of approximately 342 total"; page 2: "The following 137 pages are in this category, out of approximately 342 total").
 * After that, I'm a bit stuck. The category page does warn that "This list may not reflect recent changes", which might account for it; if the deviation (currently five) fluctuates over time, this might suggest it is just based on lag. Or it could be some other error. By the way, maybe we should move this discussion over to the regular talk page, which might have more eyes. --Usernameunique (talk) 08:18, 19 March 2021 (UTC)
 * Usernameunique, the This list may not reflect recent changes boilerplate on category pages is meaningless, in my experience—when something I do causes a change in the membership on any number of category pages, it happens right away, including to the count. I think you'll find that the phantom five are always there on the awaiting page. I had known about the GAN counter template and how it worked, though the "Wait" parameter gets its count from PAGESINCATEGORY with "Category:Good article nominees awaiting review". I've just updated GAN counter to use the PAGESINCATEGORY "pages" parameter, which (as far as I can tell) returns the same numbers as before, but it gets rid of the manual subtraction for the subcategories in favor of getting PAGESINCATEGORY to simply exclude them. Unfortunately, it doesn't also get rid of the phantom five for the awaiting review category, which I'd been hoping it might somehow cure, if there were some non-page unseen subcategories or files being included in the count. Oh well. BlueMoonset (talk) 00:59, 20 March 2021 (UTC)
 * Would User:SDZeroBot/GAN_sorting help with the unreviewed ones? Maybe theres some on one list but not the other. I was also thinking whether ones marked as 2nd opinion would effect the count. I'm not a coding expert. It is indeed strange. --MrLinkinPark333 (talk) 02:02, 20 March 2021 (UTC)
 * I've asked about the issue at the village pump. --Usernameunique (talk) 04:32, 22 March 2021 (UTC)

I've given out the Content Review Medal of Merit
... as it's obvious who it goes to, but I'll leave the rest of the awards to the other coords. (t &#183; c)  buidhe  01:07, 1 April 2021 (UTC)