This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.
Helper script: Add an option to disable watchlisting user talk pages
If I review a lot of drafts, my watchlist quickly piles up with user talk pages I never wanted on my watchlist in the first place. Please add an option to disable adding these user talk pages to my watchlist. —pythoncoder (talk | contribs) 04:21, 9 February 2023 (UTC)
Hm, I was just coming here to report the opposite - talk pages are, correctly, not showing up on my watchlist (this is desired behaviour), but I would really appreciate it if the script subscribed me to the individual discussion that it leaves on Talk pages when you accept/decline an article. I've noticed that editors are using the "reply" button on these and it's not tagging me. -- asilvering (talk) 04:27, 9 February 2023 (UTC)
@Asilvering. Mind linking to a page with a section that you'd like subscribed? I don't think accepting a draft creates any talk page sections by default, so I think I am misunderstanding you or missing something. –Novem Linguae (talk) 19:47, 9 February 2023 (UTC)
Here's one: [1]. This is a decline, not an accept, but the principle is the same. It creates a new "Your submission..." section and there is a reply button next to my name. I don't get notified unless I've also been the creator of the Talk page, as mentioned in the sub-thread below this one. -- asilvering (talk) 20:33, 12 February 2023 (UTC)
@Pythoncoder. Do you have a steps to reproduce? As far as I know, AFCH never watchlists anything (a search for the word "watch" in the codebase returns 0 hits), so this may actually be a different script/gadget/preference. –Novem Linguae (talk) 19:45, 9 February 2023 (UTC)
Yeah, this sounds more like a Preferences thing, there is an option to watchlist any page that one moves, and obviously accepting a draft would tick that box. Same is true for editing etc. See if you have those options turned on? Primefac (talk) 20:13, 9 February 2023 (UTC)
Oh, that must be it. I'll turn that off then. —pythoncoder (talk | contribs) 20:55, 9 February 2023 (UTC)
Looks like it's actually caused by my setting to "Add pages I create and files I upload to my watchlist", which would make sense because many draft contributors (especially the drive-by types) are new to Wikipedia and thus don't have talkpages yet. —pythoncoder (talk | contribs) 20:57, 9 February 2023 (UTC)
I keep my watchlist open in a separate tab and just click on the 'x' to un-watch them manually... Mattdaviesfsic (talk) 19:56, 9 February 2023 (UTC)
Think it's Wikimedia software doing that if the Talk page doesn't exist when article accepted since we banner the talk page. Xtools lookup of my account seems to confirm this. Slywriter (talk) 20:17, 9 February 2023 (UTC)
Yes, creating pages tends to put them on one's watchlist, barring Preferences changes (mentioned above). Forgot to mention that earlier. Primefac (talk) 20:57, 9 February 2023 (UTC)
Uploaded draft in wrong language
I wrote an article in Bulgarian thinking I was in the Bulgarian Wikipedia, then uploaded the draft, but it turned out I was in the English one. Can I somehow move the draft to the Bulgarian Wikipedia? Vassil01 (talk) 04:58, 9 February 2023 (UTC)
@Vassil01 you can just copy and paste the draft over to the Bulgarian wiki. – robertsky (talk) 05:06, 9 February 2023 (UTC)
Thanks! Vassil01 (talk) 11:54, 14 February 2023 (UTC)
I accepted this yesterday, but felt for no real reason that it was borderline. I still feel this today. This is a (currently) unused disambiguating page.
Was I right to accept this
If I was not right, should it now be offered for deletion
It's good that we self review and ask for the opinions of others 🇺🇦 FiddleTimtrentFaddleTalk to me 🇺🇦 14:33, 8 February 2023 (UTC)
I'd support that. It's an ambiguous term, but that target is much more all encompassing than the current disambiguation page could ever hope to be without being a copy of List of people who have received a state funeral. Hey man im josh (talk) 15:22, 8 February 2023 (UTC)
When I saw the acceptance the hairs on the back of my neck stood up as I remembered one of my kerfluffles when I resumed Admin duties. I think redirecting it is the right thing; dabs have a specific purpose, but if the topic of the collection of those links can be handled in a more general fashion, such as a list article, then that's the better solution. - UtherSRG(talk) 15:33, 8 February 2023 (UTC)
I see nothing wrong with repurposing the text as a list. I took it at face value and decided that it had a better than 50% chance of surviving an immediate deletion process, so accepted. 🇺🇦 FiddleTimtrentFaddleTalk to me 🇺🇦 15:59, 8 February 2023 (UTC)
I'm curious as to why an article being a copyright violation is a decline reason rather than a reject reason (along with vandalism). IT would make more sense for this and the vandalism one to be a reject reason considering drafts can be deleted for both of these reasons so users shouldn't be resubmitting them. ― Blaze WolfTalkBlaze Wolf#6545 21:11, 16 February 2023 (UTC)
@Blaze Wolf, because copyright violations can be removed. If the draft consists soley or largely of copyvios, it can be tagged for deletion. — Qwerfjkltalk 21:13, 16 February 2023 (UTC)
To add slightly, for copyvio there are two options: Remove and decline or Delete. Reject never plays a part because we can't leave a copyvio published. Slywriter (talk) 21:40, 16 February 2023 (UTC)
Also, a draft can be about a notable individual (and/or potentially notable individual), but if the majority is copyrighted then it is often not appropriate to accept it, even in redacted form. Primefac (talk) 13:44, 18 February 2023 (UTC)
Blocked / banned
Misterrrrr blocked as sockpuppet. They approved AfC submissions, one of which caught my attention as it was clearly not-notable. Just FYI for those who want to go back to their aprovals for cleanup. CNMall41 (talk) 21:54, 15 February 2023 (UTC)
I have examined their three listed approvals at User:Misterrrrr/AfC and have some concerns about sourcing in Diwali in Muzaffarnagar since some of the coverage is by no means significant. I think it is on the right side of the borderline, but can be convinced otherwise, and have no objection if the community gets an AfD look at it which will tell us either way. The other two appear to be serving politicians.
This does not appear to the the AFCH script's AFC Log, though, thus is misleading. A full query would tell us more. 🇺🇦 FiddleTimtrentFaddleTalk to me 🇺🇦 22:06, 15 February 2023 (UTC)
Thanks. I am using the AfC Reviewer History tool now and trying to go through what is listed. --CNMall41 (talk) 22:17, 15 February 2023 (UTC)
Should this be cross-posted at WT:NPP? S0091 (talk) 22:38, 15 February 2023 (UTC)
Probably. I'm trying to WP:AGF but I am noticing a pattern with the approvals in the edit history. --CNMall41 (talk) 22:46, 15 February 2023 (UTC)
AGF with socks is very challenging. I tend to fail dismally! 🇺🇦 FiddleTimtrentFaddleTalk to me 🇺🇦 23:19, 15 February 2023 (UTC)
I get that. I never do when it comes to socks, but there is something else that jumped out at me. If no one notices without my input then I am likely overthinking. --CNMall41 (talk) 23:45, 15 February 2023 (UTC)
Worth letting the CU on the case know by email privately, probably. 🇺🇦 FiddleTimtrentFaddleTalk to me 🇺🇦 23:48, 15 February 2023 (UTC)
Done. --CNMall41 (talk) 00:47, 16 February 2023 (UTC)
I don't want to edit your table, but to save others the trouble: # 1, 2, 5, 7, 9, and 10 are all WP:NPOL passes. -- asilvering (talk) 00:53, 16 February 2023 (UTC)
Asilvering when we do these types of reviews the tables/lists are there to be edited by all to keep track and avoid wasting people time by duplicating work. Cheers KylieTastic (talk) 11:17, 16 February 2023 (UTC)
I seem to be the only one who signed my answers though! 😇😈 🇺🇦 FiddleTimtrentFaddleTalk to me 🇺🇦 19:31, 16 February 2023 (UTC)
Now we have checked all the accepts the "error rate" is high. Three reversions, two AfDs and one potential AfD. How do these acceptances equate wth the comments about a UPE ring? The only common thread appears to be NPOL passes. 🇺🇦 FiddleTimtrentFaddleTalk to me 🇺🇦 08:27, 16 February 2023 (UTC)
I've been suspicious of Misterrrrr from the start, but evidence of UPE was elusive. I think their AfC error rate is no more than we might expect for any new reviewer and agree there's no obvious pattern. Curb Safe Charmer (talk) 17:52, 16 February 2023 (UTC)
In my opinion UPE and COI editors' cases should be handled carefully while reviewing articles in AfC. Twinkle1990 (talk) 13:13, 21 February 2023 (UTC)
Should we check the original authors of these drafts to see if there's any UPE/Sock connections? - UtherSRG(talk) 13:31, 21 February 2023 (UTC)
I support your proposal. We should get a clear instruction to accept or not to accept {{UPE}} and {{CoI}} editors/creators' and submitters' AfC drafts. Twinkle1990 (talk) 13:45, 21 February 2023 (UTC)
The general instruction is to treat COI drafts like normal drafts and apply the normal criteria. Once the problems that make them obviously COI are fixed, such as NPOV and unreliable sources, then they are just normal articles. UPE is never allowed, but is hard to prove. –Novem Linguae (talk) 14:10, 21 February 2023 (UTC)
Help finding copyvio
Is there any chance anyone could find the source for Draft:LÜ Bing-Chuan? It doesn't show up with earwig, but if it's not copyvio, it's two editors sharing the same account. I suspect it might be deep-L on one of the Chinese sources. Here's the other draft article by the same user: Draft:Zhuang Jin-Cai. The difference in writing is dramatic. -- asilvering (talk) 04:16, 21 February 2023 (UTC)
@Asilvering Ah.. Opps... I okayed the article without checking this talk page earlier. I did a quick check on the references, and they are roughly aligned with the text, but maybe embellished a bit. The sources cited do line up with the text in the article. I don't think it is copyvio. DeepL, maybe, but even if I do a mental translation into Chinese, I don't see/feel how it may be generated in this manner. Upon seeing your message here, I relooked at the article, and I think it is a translation of the zhwiki article: zh:呂炳川. Maybe an attribution somewhere is in order? Same with Draft:Zhuang Jin-Cai, it is a translation of zh:莊進才. – robertsky (talk) 15:31, 21 February 2023 (UTC)
Could you drop a note on the user's Talk page about attributing translations from other wikis? I think you might be better doing this than me since you can read the originals. -- asilvering (talk) 15:33, 21 February 2023 (UTC)
I don't know about y'all but I miss @DoubleGrazing. One of the many things they did was answer AfC Helpdesk queries. That also means @331dot you cannot leave lol. Thanks to all the folks who pitch in (now and always). S0091 (talk) 18:31, 22 February 2023 (UTC)
Thanks for sharing. I find the Marketing Company Mediation addition a little strange, so I have asked a question about it. –Novem Linguae (talk) 15:34, 22 February 2023 (UTC)
My concern about the Marketing Company Mediation section was resolved. In brief, mediation may open the door for WMF to take cost-effective legal action against some UPE farms, whereas before it was cost-prohibitive. –Novem Linguae (talk) 22:52, 22 February 2023 (UTC)
Checking AfC pending drafts, I have witnessed a sudden growing of WP:Academic submissions by editors who are editing their first edit. Onel5969 is kind to catch WP:UPE and WP:COI. But it's a challenging issue for us to handle as there are many editors active in WP:UPE per noticed in Freelancer.com and Upwork. We need a crystal clear scenario of WP:UPE as we can't detect them, but I have found a path to identify them all - "They all uses hyperlinking in the draft articles and places references before coma and the period." Who else have noticed the same? Twinkle1990 (talk) 13:43, 23 February 2023 (UTC)
So, are you suggesting something around UPE like changing WP rules around it, or some kind of section for how reviewers have detected editors being paid? Could you clarify the purpose of this section, please? Mattdaviesfsic (talk) 18:40, 23 February 2023 (UTC)
What should we do with the WP:UPE submissions? Twinkle1990 (talk) 01:14, 24 February 2023 (UTC)
Can you link to some examples? I do a lot of WP:NPROF. I've noticed both hyperlinking in drafts and references before punctuation in articles before, but they're not uncommon characteristics of drafts in general so I'd like to see if there are other tells. Is there some reason not to get a CU to check for a sock farm here? That seems the obvious next move, to me? -- asilvering (talk) 00:13, 24 February 2023 (UTC)
UPE is, by its very nature, extremely difficult to detect. That is, there isn't a necessarily a single "tell", but a broad series of them. Can you provide some examples that you have seen? Curbon7 (talk) 01:57, 24 February 2023 (UTC)
It's not always great to discuss UPE tells on wiki. Might be better for email or Discord. For the UPE angle, SPI seems like a good approach if there's enough evidence. For the COI angle, I think we're supposed to treat these like any other draft. If they get improved to the point that they pass our policies and the COI is no longer detectable (all the WP:UNDUE, bad sources, etc. are fixed), then they can probably be accepted like a normal draft. Of course I could be way off, let me know if you disagree. –Novem Linguae (talk) 02:33, 24 February 2023 (UTC)
Women in Red March 2023
Women in Red Mar 2023, Vol 9, Iss 3, Nos 251, 252, 258, 259, 260, 261
--Lajmmoore (talk) 12:51, 26 February 2023 (UTC) via MassMessaging
Accepting drafts that are already redirects
I am having trouble accepting a few drafts with the AfC helper script, which already exist in the mainspace as redirects. Is there a way to accept these without having to do it manually? 141Pr {contribs/Best page} 16:51, 21 February 2023 (UTC)
@Praseodymium-141: you can tag the mainspace redirect with {{db-afc-move}}, and you can come back to it in a couple hours and an admin will probably have deleted the redirect. Alternatively, users with the Page mover right can swap the two pages, which has the added benefit of preserving the redirect's history. — Ingenuity (talk • contribs) 17:09, 21 February 2023 (UTC)
I generally also leave an AfC comment stating I am waiting for the redirect to be deleted to accept it. I think that helps the admin know the request is by an AfC reviewer and also occasionally it can take some time for deletion so it lets other review know as well in case I forget in the interim. S0091 (talk) 20:01, 21 February 2023 (UTC)
We should build non-trivial redirect detection and {{Db-afc-move}} tagging into the AFC helper script. It is our most common question on this talk page, approximately 5 times a year. I'll try to write a patch if I get time. Anyone else is welcome to beat me to it though :) –Novem Linguae (talk) 20:36, 21 February 2023 (UTC)
This would be really helpful, thank you! -- asilvering (talk) 22:13, 21 February 2023 (UTC)
As a general note, please make sure that if you are marking the redirect as {{db-afc-move}} that you also mark the submission as "under review". This will save time and duplicating work. Primefac (talk) 09:45, 28 February 2023 (UTC)
ChatGPT and other AI generated drafts
I needed to unwind from work, and dove into AfC submissions for about an hour or so. Out of the 10 drafts I acted on, 3 of it, with 1 as collateral (it was reviewed and decline as copyvio earlier) bugged me the wrong way (Draft:Corporate EnglishUser:Hoangluong007/sandbox, Draft:Virtualization and containerization in testing environments. and collateral: Draft:E-Library)... They looked artificial. I did a check on these with writefull's GPT checker, and they returned with rather high confidence (>90%) that they may be AI generated. These were also lacking in-line references. I have declined them with either hoax/joke, essay, or exists, and left a link to Wikipedia:Large language models as comments.
I am wondering, how the rest of the reviewers here deal with this situation? – robertsky (talk) 15:40, 21 February 2023 (UTC)
I was just wondering about this earlier today. I haven't seen any LLM-written ones yet - at least, not ones with faked references, which is the obvious tipoff. If they had no references I'd be declining anyway, and then immediately forgetting about them. A warning to anyone reading this: LLM-checkers have a very high rate of false positives, so you should never use this alone as a reason for rejection. They will probably score "wikipedia style" writing as higher in general for two reasons: 1) the models are trained on wikipedia data, and 2) they tend to rate writing by autistic people as AI-generated (and what is Wikipedia if not a collection of a great many people's special interests?). I don't mean to imply here that robertsky has done anything wrong (obviously, no references is already a problem on its own), but I urge caution to AfC reviewers in general. Always check the refs - chatGPT will generate extremely plausible sounding but utterly fake ones. -- asilvering (talk) 22:12, 21 February 2023 (UTC)
Haha, that "in summary" at the end does appear to be a tell as well... very common in undergraduate essays (and thus will also be common among genuine wikipedia newbies), but it does really stand out on QAAOps, doesn't it. -- asilvering (talk) 22:20, 21 February 2023 (UTC)
I honestly think we will ultimately need "Potentional issue" tagging for these in the New pages feed, similar to what we have for possible copyright vios. I have reviewed a couple LLM drafts which I declined for other reasons but did not know they were LLM until @Rsjaffe tagged them afterward. S0091 (talk) 15:42, 22 February 2023 (UTC)
I've found about 60 drafts so far that had obviously been LLM-generated, and a bunch more that were suspicious. I haven't had a single person object to the tag, but I've been very conservative in tagging—using > 99.8% confidence level if using the outdated GPT-2 detector, or using GPT zero for a sentence-by-sentence judgement, and only testing for LLM-generation when it appeared to my eye to be a high probability case. This method appears to have a low false-positive rate.
Looking for very regular sentence structure, perfect English, and repeated use of the title of the page without variation or abbreviation are some of the tipoffs. There may be references, but if those are generated by GPT, the references will probably be fictitious. I've seen some people retroactively add references that were real.
Sometimes, the AI-generated text is only a portion of the total article. If that's the case, it could be deleted based on WP:V and then judge the rest on its own merits. Here's an example of a draft that had some AI-generated text: Special:Diff/1135929240.
And just to give you an idea of how pernicious the lying can be in an LLM-generated article, I fact-checked one item in the text I removed from the Honda Big Red draft. It states The engine is paired with a manual 5-speed gearbox, which provides smooth and responsive gear changes. Sounds great, but the Big Red has an automatic transmission. — rsjaffe🗣️ 20:19, 22 February 2023 (UTC)
The ever busy and helpful @Novem Linguae, curious about your thoughts. This obviously impacts NPP as well and you support both AfC and NPP. S0091 (talk) 20:28, 22 February 2023 (UTC)
Thanks for the ping. I have a negative impression of AI-generated articles so far. They seem to be hard-to-detect hoaxes, with fluent sounding language, but incorrect information and fake references. I'd like to see Wikipedia move towards CSDing these. However right now there is resistance to even getting the maintenance tag to say to delete this text on sight (I was reverted). Instead the maintenance template encourages users to comb through the AI-generated text and spend time salvaging it. We'll see what happens.
From an NPP/AFC standpoint, maybe we can incorporate an open-source LLM detector into some of our software. I'll make a PageTriage ticket. –Novem Linguae (talk) 22:30, 22 February 2023 (UTC)
I tried getting a new CSD tag for these articles, but perhaps raised the issue prematurely, as we did not have any policy relating to machine-generated text, nor a lot of experience at AfD/MfD about this either. — rsjaffe🗣️ 22:42, 22 February 2023 (UTC)
Oh dear. fwiw, I'm extremely skeptical about LLM detectors (several academics who work in AI have found they will state with high confidence that portions of their own books are faked), but combing through AI-generated text and salvaging it is a hopeless waste of editors' time. It's just plagiarism laundering. We should be assuming good faith and willing to work with new editors who tripped n LLM detecting filter or don't have the hang of "Wikipedia style" yet, but this is a reason to invoke WP:PACT if there ever was one. -- asilvering (talk) 01:53, 23 February 2023 (UTC)
That's why it's useful to prescreen before using a detector. False positive rate is strongly affected by the prevalence of the problem in the population being tested. The more you can get a "concentrated" sample, the less of a problem false positives are. See this discussion of positive predictive value. This is my approach, and it's tuned to miss a lot of stuff but be fairly confident when getting a positive result. — rsjaffe🗣️ 03:08, 23 February 2023 (UTC)
Agree with the skepticism about (the currently available) LLM detectors, see below.
Regarding WP:PACT, given ChatGPT's well-documented inability to provide valid references, I wonder if this may be a good moment to strengthen the emphasis on newly created articles needing to have valid citations. (although future AIs may not have this obvious weakness)
Regards, HaeB (talk) 09:29, 24 February 2023 (UTC)
@Rsjaffe have you seen bad LLM drafts be accepted yet? S0091 (talk) 20:42, 22 February 2023 (UTC)
Yes. I'm having trouble re-locating them all (I believe I found three). Here's one: Pavilion of Harmony. It had ChatGPT output mixed with apparently human output. That user had another article in draftspace flagged by me, and the user subsequently deleted that flagged article. You'll see in the edit history where the AI stuff was removed (I had mentioned it in a village pump discussion).
Here's one that was in mainspace but got moved to draft, apparently by its creator: Draft:Nordic states game industry. The main editor of the article, after I tagged it for AI, appropriately edited to remove the AI-generated text and removed the tag. — rsjaffe🗣️ 21:41, 22 February 2023 (UTC)
Note that these had segments of AI-generated text rather than being completely computer-generated. As I noted above in the Big Red article, the risk of persuasive lying is high when computer generation is used, which makes identifying these text segments important. — rsjaffe🗣️ 21:43, 22 February 2023 (UTC)
Oh dear. That editor is using text generators quite a bit, it seems. At least now this is being mentioned in their edit summaries, but I think this is a bit of a ticking time bomb for wp from a copyright cleanup perspective. I'm not aware of any similar lawsuits for LLMs yet, but there's this one now against Stable Diffusion: [2]. -- asilvering (talk) 02:04, 23 February 2023 (UTC)
Look at the draft LLM policy Wikipedia:Large_language_models. There seems to be fewer copyright issues than you may fear, at least for the end-users, as the generated text rarely paraphrases or quotes things. I think the current suits are against the people making the AIs, as they are scooping up lots of intellectual property and digesting it to develop their programs.
I'm much much more worried about misinformation than I am about copyright. — rsjaffe🗣️ 02:57, 23 February 2023 (UTC)
I'm certainly worried about misinfo. But the position of the lawsuit against Stable Diffusion is that these kinds of models are per se copyright violations, and I see no reason why the same legal logic would not apply to LLMs. I don't think we want to find out and then have to clean up the mess. -- asilvering (talk) 03:00, 23 February 2023 (UTC)
I think the same about misinformation. The problem with Chat GPT articles is that – unless you absolutely know the article's subject – it is very difficult to spot errors because Chat GPT can write very convincing text that is total nonsense. I strongly support tagging AI-generated articles as hoax articles, i.e., CSD'ing them. Best regards, --Johannes (Talk) (Contribs) (Articles) 07:50, 23 February 2023 (UTC)
With all due respect, following the discussion about your Signpost article I have some doubts about the validity of your assumptions regarding these detectors. Regards, HaeB (talk) 09:29, 24 February 2023 (UTC)
Worth noting that in one of the examples provided by Robertsky above, the author strongly disputes his conclusion. Regards, HaeB (talk) 09:29, 24 February 2023 (UTC)
I could write some code to automatically check new drafts against a LLM detector, if anyone knows an API avaliable. — Qwerfjkltalk 20:13, 23 February 2023 (UTC)
GPTZero has an API: see bottom of https://gptzero.me/. I'll look to see if there are others. GPTZero would be a good one if we could get it to work for our purposes. — rsjaffe🗣️ 23:36, 23 February 2023 (UTC)
We are also thinking about integrating this into PageTriage. Please feel free to join the discussion at phab:T330346. –Novem Linguae (talk) 00:05, 24 February 2023 (UTC)
It would be helpful to have this as an edit filter on drafts, if that wouldn't be too many API calls. Or if it would flag articles in the same way that we currently have "possible spam" or "2 previous declines" when looking at the lists of AfC submissions by topic. -- asilvering (talk) 00:09, 24 February 2023 (UTC)
Do note that if the article is composed from different sets of generated texts, the detector may return as a negative. e.g. Draft:Virtualization and containerization in testing environments as a whole would pass a detector's check, but if you break down by sections, all the sections returned as positive individually. – robertsky (talk) 08:41, 24 February 2023 (UTC)
Currently, the available detectors are being regarded as "definitely not good enough" to use for important decisions, due to frequent false positives and false negatives (and are often intended for outdated models like 2019's GPT-2). This includes OpenAI's own tool for detecting ChatGPT-generated text. This situation may change of course, especially if OpenAI goes forward with their watermarking plans for ChatGPT. (Also, Turnitin announced last week that they have developed a detector that "in its lab, identifies 97 percent of ChatGPT and GPT3 authored writing, with a very low less than 1/100 false positive rate" and will make it available to their customers "as early as April 2023." But even there it's worth being skeptical about whether they can keep up these levels of sensitivity and specificity.)
Regards, HaeB (talk) 09:29, 24 February 2023 (UTC)
How is this different than the copyright violation detectors we use as far as false-positives or misses (or spam)? I see LLM as being similar. The flag is just an indicator and human review is required. I may be thinking about it incorrectly, though and there's more nuances to it. S0091 (talk) 16:31, 28 February 2023 (UTC)
One very obvious difference is that copyvio detectors usually also identify the (potential) original that the content was obtained from, so human review and excluding false positives is much easier. Regards, HaeB (talk) 17:06, 28 February 2023 (UTC)
Hello! So while attempting to decline User:Heisenburg7/sandbox for being in Serbian and not English, upon pulling up the AFC reviewer script (or UI or whatever), for whatever reason it causes that browser tab to hang. I don't know what's going on here but could someone possibly decline it for me? ― Blaze WolfTalkBlaze Wolf#6545 21:12, 6 March 2023 (UTC)
I am having to same problem and tried a couple different things but they didn't work. S0091 (talk) 21:25, 6 March 2023 (UTC)
Blaze Wolf, known issue that I think is still an outstanding ticket - cause is too many spaces next to headings - I fixed with this then could complete the review. It had ~38 spaces after the "Рани живот" section. Cheers KylieTastic (talk) 21:29, 6 March 2023 (UTC)
@KylieTastic you are "Tastic" for a reason. Thanks for your help! S0091 (talk) 22:12, 6 March 2023 (UTC)
I have been through a check in Freelancer.com and found this. In my opinion, we should be more cautious before approving any AfC draft. Twinkle1990 (talk) 13:05, 28 February 2023 (UTC)
What do you mean by "more cautious"? Primefac (talk) 13:10, 28 February 2023 (UTC)
@Primefac Should we approve such PAID, UPE and COIN? We need guidance. Twinkle1990 (talk) 15:09, 28 February 2023 (UTC)
There is absolutely nothing wrong with PAID and COI editing, provided that the user in question is following best practices, so please separate that from UPE, which is against our policies. A PAID or COI submission that meets our acceptance criteria should be accepted, because that's the entire point of AFC. Primefac (talk) 15:18, 28 February 2023 (UTC)
@Primefac thanking you for your valuable inputs, we are supposed to accept declared WP:UPE but not the otherwise? Twinkle1990 (talk) 16:03, 28 February 2023 (UTC)
@Primefac I am confused, why we we are accepting paid articles! It's strange for me. Twinkle1990 (talk) 16:06, 28 February 2023 (UTC)
A paid editor who declares and submits to AfC or a COI who declares and submits to AfC has complied with WMF ToS and En-Wiki policy. There's no reason to decline on that basis. An undeclared Paid Editor violates Terms of Service, which would be a basis to decline. Slywriter (talk) 16:13, 28 February 2023 (UTC)
There is a contradiction in your statement: declared WP:UPE. UPE is undisclosed paid editing. So declared undisclosed paid editing? Paid editors are encouraged to submit drafts for AfC reviewers to vet through for GNG, NPOV, PROMO, ADV among other possible issues. As long as the content they write meets our guidelines and policies, including disclosing their paid status, there is no issue with accepting such drafts. In my opinion, we expect such articles meet a much higher/stricter bar since the writers are paid for their work. – robertsky (talk) 16:14, 28 February 2023 (UTC)
I should refrain from accepting any AfC that is connected to WP:UPE and WP:COI. Twinkle1990 (talk) 16:15, 28 February 2023 (UTC)
You are welcome to do so; we do not force reviewers to review drafts they are not comfortable reviewing. Primefac (talk) 19:32, 28 February 2023 (UTC)
Sidebar on reviews
I briefly browsed through Twinkle1990's recent AfC-related edits and frankly, I am a bit concerned. Editors who obviously disclose their paid editing should not be told that their disclosed paid editing is a violation of the terms of use, and their AfC submissions should not be declined and tagged as UPE. This is extremely harmful to the effort we have to undertake against UPE, because it gives paid editors who disclose the impression that they violate the terms of use, and this, as a result, will likely lead to disruptive editing. I think that checking Twinkle1990's recent AfC history is a good idea. Best regards, --Johannes (Talk) (Contribs) (Articles) 22:11, 28 February 2023 (UTC)
As a procedural note on that point, they are a probationary reviewer so if there are concerns they can be removed from the project without a formal review. I have also seen some problematic declines but have not had time to dig into them to make a judgement. Primefac (talk) 10:09, 1 March 2023 (UTC)
@Johannes Maximilian I might be a probationary reviewer, but I have added notes in declines. Nevertheless, I take time to read the sources before accepting/declining. My one decline was false positive by WP:CV check which was fixed by @Primefac. But, many of my declines led CSD completed and editor blocking. Yeah, you rightly said "checking Twinkle1990's recent AfC history is a good idea" which should be considered per through checking of my reviewing. Day by day AfC submissions are increasing, and as a probationary reviewer, I don't think I have violated any rule of WP:AFC. Thanks for your concern. Twinkle1990 (talk) 12:32, 3 March 2023 (UTC)
Paid editing is very unpopular, but is allowed under our rules if it is disclosed. I think that's what Johannes was worried about above. I think Johannes found an example of where you were too harsh to a disclosed paid editor. Disclosed paid editors are encouraged to use AFC under our rules (see Wikipedia:Paid-contribution disclosure#Conflict of interest guideline), and their drafts should be judged on their merits like any other draft. Undisclosed paid editors are different. Hope that makes sense. –Novem Linguae (talk) 12:41, 3 March 2023 (UTC)
Yes, absolutely! Editors who have disclosed shouldn't have their drafts tagges as "contains undisclosed paid editing". The UPE tag should only be used if an obviously paid editor doesn't disclose. Best regards, --Johannes (Talk) (Contribs) (Articles) 13:04, 3 March 2023 (UTC)
I am against both paid and UPE. I shall raise this concern for the AfC reviewing process. Going through check on Freelancer and Upwork, I shall say that we really need a new consensus at WP:AN as paid editing through mentioned above job sites are growing rapidly. Twinkle1990 (talk) 13:16, 3 March 2023 (UTC)
I have this feeling that you don't understand what paid editing in a Wikipedia context is. Several editors have told how it works, now, here is my approach: The set of "paid editing" has two proper subsets: disclosed paid editing (perfectly acceptable), and undisclosed paid editing (not acceptable). Presuming that you are a reasonable editor (per WP:AGF), then what you are saying ("I am against both paid and UPE") is illogical. Now, it is okay if you oppse paid editing (i.e., if you oppose both disclosed and undisclosed paid editing), but then you should do yourself a favour and don't review (and don't decline) AfC drafts that you figure are any kind of paid. I'm not this type of editor who points out mistakes for the sake of pointing out mistakes, but I'm trying to be clear, and straightforward: It is not acceptable if – in cases where paid editors have disclosed their paid editing according to the ToU – you decline their drafts, tag their drafts as UPE, and attack them on their talk pages by telling them that they will be blocked, just because they are paid editors.
The growth of paid editing offers/requests on job sites is a nuisance, yes, but the paid editing that one can buy at said sites for less than 1000 € is of such low quality that it is very easy to spot for the average Wikipedian. Also keep in mind that in cases that are difficult, I suppose fellow AfC reviewers would be willing to help. E.g., I can assess most German-language sources and say whether they are acceptable for a given article. Best regards, --Johannes (Talk) (Contribs) (Articles) 08:40, 4 March 2023 (UTC)
@Johannes Maximilian For a note, I have already stopped reviewing UPE/PAID/COI submissions. I don't want to be in trouble. Twinkle1990 (talk) 12:31, 4 March 2023 (UTC)
I don't want to add any fuel to the flames, but are you aware that AfC is the place for UPE/PAID/COI submission on Wikipedia? I'm not saying that being an AfC reviewer and avoiding said submission is impossible, but I am almost certain that even the best AfC reviewers cannot do that reliably. Best regards, --Johannes (Talk) (Contribs) (Articles) 10:28, 5 March 2023 (UTC)
"I have already stopped reviewing UPE/PAID/COI submissions". You have not stopped doing that. Nobody says that you have to, but I find this pretty confusing. Best regards, --Johannes (Talk) (Contribs) (Articles) 16:39, 6 March 2023 (UTC)
@S0091 I think that we both see the same issue. Twinkle1990's recent AfC accept history is concerning. I have taken 5 minutes of my time to look at the references sections of all the accepted drafts, and I have only checked for very obvious errors/flaws (i.e, citation errors or sources marked as unreliable). I found that roughly 30 per cent of all accepted drafts were flawed just by performing the most obvious check that an AfC reviewer can perform. This leads me to conclude that Twinkle1990's claim "Nevertheless, I take time to read the sources before accepting/declining" makes no sense. If I had no AGF, I'd even argue that it's either a lie, or indicative of a lack of competence. And this leaves the foreign-language sources (that are not marked as unreliable, but that I doubt Twinkle1990 understands) out of the equation. Best regards, --Johannes (Talk) (Contribs) (Articles) 07:33, 7 March 2023 (UTC)
Thank you all for the constructive feedback. I have removed Twinkle1990 from the AFCP list, with no prejudice towards re-addition in the future if they can indicate they have overcome some of the biases indicated here. Primefac (talk) 10:35, 7 March 2023 (UTC)
Thanks for removing me. I was about to request you to remove me. Twinkle1990 (talk) 11:44, 7 March 2023 (UTC)
@Johannes if you provide me (the few/a few) accepted drafts as you see problematic acceptance, it would be a great help for my future here, if I don't get blocked for those. Regarding your say "Twinkle1990's claim "Nevertheless, I take time to read the sources before accepting/declining" makes no sense.", I used to have a check per WP:THREE. I accepted drafts if met notability criteria. The single exception is Chinnapat Panwisawas which is under WP:PRD after my accepting and patrolled by WP:NPP, which was massively edited by the page creator after acceptance. I rest my final reply to you. Twinkle1990 (talk) 12:18, 7 March 2023 (UTC)
I don't want you to be blocked, and I don't think that you have done things so terribly wrong that a block is warranted. It's just that your actions have been confusing to me and other Wikipedians:
On 28 February, you asked whether we should accept "declared UPE", which makes no sense because UPE is always undeclared, and can never be declared. It's like asking whether ice cream shall be served hot.
On 28 February, you asked a paid editor – who had disclosed – to stop editing Wikipedia (Special:Diff/1142107212) and tagged his draft as UDP (Special:Diff/1142106991), which makes no sense.
On 3 March, you claimed that you "take time to read the sources before accepting/declining", which S0091 and I found to be not the case (drafts with foreign language sources / drafts with referencing flaws).
On 4 March, you said that you had stopped reviewing drafts that included paid contributions, however, you continued reviewing drafts that included paid contributions (and I had previously mentioned that being an AfC reviewer and avoiding reviewing paid contributions is virtually impossible).
I suppose this is a very unfortunate chain of events, but is it clear why you have caused so much confusion?
Now, regarding the drafts that you accepted that contain obvious referencing flaws (and I have ONLY conducted a very basic, superficial test for each article):
@S0091 for a note, could you accept Draft:Edwin Key Yager? Regarding that draft's decline, it is already proved that the draft was not not ready to accept. Would you accept the draft as it's waiting for AFC review? Can you prove me that draft passes reliable sources? To be honest, I must reiterate, I did first mistake by accepting the Draft:Bernard Kramer (U.S. Army) which was quickly reverted by patrollerOnel5969, after which I realized what to not do. From that day, I was strict on reviewing. I don't have any wrong feeling to lose AFC reviewer right, but I wanted to clarify a few questions raised against me. Twinkle1990 (talk) 12:31, 7 March 2023 (UTC)
I don't understand what you are asking but the issue was not that you declined the draft. The issue was the reason you declined the draft along with your responses in the help desk discussion.
I agree with Johannes in that I don't think anything you have done from an AfC perspective warrants a block. Removal from the project is sufficient in that regard. I also hope you learn from your WP:Articles for deletion/Clement Richardson nomination, which is a subject that clearly meets WP:NPROF and is probably one the most misunderstood notability guidelines. I don't have a great grasp of it either unless it is obvious and being president of a public university certainly meets #6. S0091 (talk) 19:01, 7 March 2023 (UTC)
How can we improve the feedback in order to encourage newcomers?
Background: I have a draft that has been declined three times: Draft:Main Directorate of Deep-Sea Research. I have received the same message about notability as an explanation. I have asked the reviewers about more pieces of advice and I got some hints. (For that draft, you can comment here).
My general question: How do we improve the feedback on the declined drafts? It would be good to have the articles accepted in the next round, as it will lower the burden on reviewers.
My background: I have been active on Wikipedia for more than 15 years. I have seen the whole spectrum of contributions. I have created 30 articles on ENWP and 129 on SVWP, where I am an administrator.
I know that the reviewers are doing it voluntarily and they deserve many thanks for their work. The point is to identify the possible articles and to suggest concrete improvements in order to be able to accept them. Per W (talk) 08:22, 6 March 2023 (UTC)
@Per W may I seek the freedom to take part in this conversation? Twinkle1990 (talk) 12:35, 7 March 2023 (UTC)
Hello there! The Active reviewers section on the page Wikipedia:WikiProject Articles for creation/Participants states that "Please note that as of 13 September administrators do not need to add their name...". I think this phrase should also mention the year in which the motion was appended to minimise confusion, as right now even I have no idea if it was 13 September 2022/21/20/19... etc. ❯❯❯ Raydann(Talk) 18:46, 8 March 2023 (UTC)
Date removed; it's been long enough now. Primefac (talk) 19:29, 8 March 2023 (UTC)
114 refs, for those curious. Primefac (talk) 18:17, 8 March 2023 (UTC)
Oh so now they are faking American and Canadian websites for there promo spam - yuk :/ - and yes I have G11ed as this behaviour needs a good slapping. KylieTastic (talk) 18:54, 8 March 2023 (UTC)
As a general rule of thumb, a decent draft will probably have around 5 - 15 sources. Any more than that, and it can be reasonably expected to be spam. So if a draft has lots of sources, spot check three or four of them (ideally using a locked down sandbox if you don't trust loading them into your main browser) and if they don't mention the subject or don't verify the claim given, you're on safe ground tagging for G11. The Times of India is particularly notorious for hosting self-published content and paid advocacy. Ritchie333(talk)(cont) 13:17, 9 March 2023 (UTC)
Change to cv decline
A recent thread at WP:AN got me thinking - how can we improve the cv decline to make it more obvious to reviewers that if there is non-infringing content that can be left on the page (i.e. it's not a {{db-g12}} candidate) that the reviewer needs to (at the very least) tag the page for {{revdel}} so that it can be dealt with? The decline notice already gives a (very tiny) note to reviewers not to leave copyvio, but clearly that is not working. My first thought is to make that note very large and very obtrusive, or somehow making it so that it must be dealt with, but I would like to get consensus before doing so. Primefac (talk) 09:48, 28 February 2023 (UTC)
I posted the linked thread. I'm also guilty of having done the requested revdel but not switched the template - I didn't know we needed to until yesterday. I see that changes were made since yesterday so that Category:AfC submissions declined as copyright violations is now a subcategory of the RD1 category, and is also populating the top-level Category:Candidates for speedy deletion, but I don't like either of these approaches. Making the decline category a subcategory of RD1 doesn't flag a count in the admin dashboard, and also Category:AfC submissions cleaned of copyright violations is a subcategory of the decline category, so in this setup drafts that are cleaned are still in the RD1 category, though buried two subcategories deep. As for populating the speedy deletion category there are two problems: 1) there shouldn't be anything in that category (they should be in a subcategory), and 2) these aren't actually candidates for speedy deletion: the desired action is to clean the violation and purge the history, but leave the rest of the draft if it has any non-violating content; deletion is only warranted if there's nothing left. There's also 2.1) that there are still too many admins that delete anything flagged for speedy deletion without reviewing at all.
It seems to me that the simplest way to handle this is to just have the {{AfC submission}} template also populate Category:Requested RD1 redactions, with CLEAR instructions to administrators (i.e. not buried behind a small-text wikilink) to change the flag after reviewing. Those instructions could be added to the category page itself. That category shows up with a red highlight at the very top of {{Admin dashboard}} if it has more than 0 members, and whether or not the AfC reviewer has followed through on tagging the article with {{copyvio-revdel}}, review and deletion is nonetheless required, so we might as well just use the category that already exists for that. Ivanvector (Talk/Edits) 15:18, 28 February 2023 (UTC)
Just as a small point of note, Ivanvector, I added the cat to RD1 a year ago, not a day ago ;-) Primefac (talk) 15:33, 28 February 2023 (UTC)
I would be fine with bumping the cat to be the straight-up "pages needing RD1" category. Primefac (talk) 15:35, 28 February 2023 (UTC)
I might also be helpful to have a bot run through the category and leave a reminder notice on the reviewer's talk page after a week or two that some type of action is needed (id the offending content and/or update the template). S0091 (talk) 16:18, 28 February 2023 (UTC)
As an RD1 regular I'm a bit concerned about putting the requests straight into CAT:RD1, as that category usually doesn't have many entries in it and it's rare for it to have more than 10-20. Category:AfC submissions declined as copyright violations currently has 141 entries, so they will completely flood this process, and as I understand it lots of them don't actually need any revdel. Even the fact that this category is a subcategory of CAT:RD1 is problematic, because it means this category never shows up as empty on dashboards. Hut 8.5 20:19, 28 February 2023 (UTC)
Any draft that is rejected for copyvio must need revdel, or else it shouldn't have been declined for that reason. Copyvios should be dealt with ASAP, even if the person who first identifies it isn't familiar with all of our processes, so while this isn't elegant, it's another trigger for action under a policy with legal implications. There is a backlog now because of a gap in our processes, but the 141 entries have built up over quite a while: some of the ones I reviewed yesterday had been rejected more than a year prior but still hadn't been cleaned. If they were in the more visible category, I feel that more admins monitoring it would be spurred to act more quickly, and the backlog would be a one-time thing that won't happen again. Ivanvector (Talk/Edits) 22:46, 28 February 2023 (UTC)
What should we do with drafts that have an incorrect copyvio decline? Remove the review altogether and procedurally resubmit? Artificially mark it as cv-cleaned to get it out of the category? Or something else? /wiae/tlk 00:20, 1 March 2023 (UTC)
I suppose it depends on what you mean by "incorrect". If there is no copyvio, then the decline should be reverted so the page can be re-reviewed (and not just resubmitted). If the amount of infringing content is so minimal that cv as the only reason for decline is a bit excessive, then clean it, mark for {{revdel}} and (potentially) resubmit. Primefac (talk) 10:11, 1 March 2023 (UTC)
Copyvio is one of the AFC quick-fail criteria: a draft with copyvio doesn't need any other reason to fail, and I assume that means reviewers aren't looking for any other deficiencies once a copyvio is identified. Also, in my experience, very often once a draft is cleaned of copyvio it's left in a state not suitable for submission, because large parts will be blank, and many are left as a one-sentence stub. Probably we should notify the authors that their draft was cleaned and is ready for them to work on, rather than resubmitting a draft that's probably going to fail again. But if a draft is rejected for copyvio and the reviewing admin finds that no copyvio is present, I'd support reverting the decline and letting the AfC reviewer know. Ivanvector (Talk/Edits) 14:21, 1 March 2023 (UTC)
Can't argue with that (and just for further clarification, my second "wrong" cv example was not the situation you mention, but one where there might be a paragraph that violates cv in a seven-paragraph draft). Primefac (talk) 14:36, 1 March 2023 (UTC)
If someone does not tick "Nominate the submission for speedy deletion" when declining as cv maybe AFCH could include User:Enterprisey/cv-revdel.js and auto launch? KylieTastic (talk) 12:07, 2 March 2023 (UTC)
That's a good addition. As another RD1 regular I don't want to be faced with 100+ entries where the cleaning work hasn't been done. It also means RD1 acts as a double check that there was a copyvio. A bit more awareness for some AFC approvers might be needed as we do see a lot of RD1 where the person tagging (not just on AFC submissions) doesn't appreciate which revisions need to be revdel'd.
On a wider point should copyvio be an AFC auto-fail? Typical scenarios are that the drafts are a) G12 candidates which is an auto-fail or b) there's something left which merits review and pass/fail for some other reason. Nthep (talk) 19:22, 2 March 2023 (UTC)
In a general sense, yes. However, there is a c) a section or paragraph is copyvio. In this case it is not necessarily a pass/fail issue. Primefac (talk) 20:11, 2 March 2023 (UTC)
Yes it is a pass/fail issue as the remaining content has to be assessed after the cv content is excised. Nthep (talk) 20:18, 2 March 2023 (UTC)
Rejig the cv decline
I did a silly thing. See below. Primefac (talk) 14:43, 1 March 2023 (UTC)
Maybe also have a note for the admin doing the RD1 redaction to set the cv-cleaned if it hasn't been done? Galobtter (pingó mió) 04:46, 3 March 2023 (UTC)
Done, admins will now see Administrators: if the page has been cleaned and you are seeing this notice, please change the cv to cv-cleaned in the {{AfC submission}} call.Primefac (talk) 08:07, 3 March 2023 (UTC)
If you would like to continue working on the submission, click on the "Edit" tab at the top of the window.
If you have not resolved the issues listed above, your draft will be declined again and potentially deleted.
If you need extra help, please ask us a question at the AfC Help Desk or get live help from experienced editors.
Please do not remove reviewer comments or this notice until the submission is accepted.
Where to get help
If you need help editing or submitting your draft, please ask us a question at the AfC Help Desk or get live help from experienced editors. These venues are only for help with editing and the submission process, not to get reviews.
If you need feedback on your draft, or if the review is taking a lot of time, you can try asking for help on the talk page of a relevant WikiProject. Some WikiProjects are more active than others so a speedy reply is not guaranteed.
To improve your odds of a faster review, tag your draft with relevant WikiProject tags using the button below. This will let reviewers know a new draft has been submitted in their area of interest. For instance, if you wrote about a female astronomer, you would want to add the Biography, Astronomy, and Women scientists tags.
Please note that if the issues are not fixed, the draft will be declined again.
The new notice is way too big and bright. Reviewing a rejected draft for CSD is already a mess of red boxes, I don't need even more of it. Is this really necessary? Isabelle Belato🏳🌈 21:23, 6 March 2023 (UTC)
It's kind of the point, since previously no one has paid attention to previous notice. Keep in mind also that when the draft has been cleaned, the big notice goes away. Primefac (talk) 11:00, 7 March 2023 (UTC)
The trouble is this is shown on the users talk page as well and the message makes no sense to them and is distracting from the info they should read. Can it be hidden in on a talk page (i.e. use {{talk other}}) ? KylieTastic (talk) 17:01, 7 March 2023 (UTC)
Ah, good idea. Done, and with all (7) user talk notices removed. Primefac (talk) 09:48, 8 March 2023 (UTC)
Working through category
I am working my way through Category:AfC submissions declined as copyright violations, which is down to under 50 (plus another 15 or so waiting to be revdeled so that they can be marked as cleaned). In terms of high-level trends, most declines looked good to me. A couple of drafts were declined for only a sentence or two of copied material; I guess it's up to AFC folks to decide whether that's how they'd like the copyright decline to be used. I also found a few drafts that were declined when the text in question was public domain (typically from a US government source), suitably licensed (I think there was a CC BY-4.0 journal article with the license hidden on an unintuitive part of the page), a bibliography / works cited, or a backwards copy. This wasn't all that common, but maybe it's something that we should flag for reviewers to keep in mind. /wiae/tlk 13:45, 6 March 2023 (UTC)
Down to seven, though with the categories now changed the regular RD1 crowd might be able to chip in and help. Two of the pages are at CCI and will likely get nuked anyway, with another couple I'm still debating G12'ing based on what I'm seeing (but annoyingly enough, not finding). Primefac (talk) 14:01, 9 March 2023 (UTC)
Place both types (clean and "unclean") into the same category, with the un-cleaned versions populating the RD1 category for review.
Thoughts and suggestions welcomed. If I don't hear anything I'll probably just go with the second option. Primefac (talk) 14:01, 9 March 2023 (UTC)
Bugger, just thought of a third option, which is slightly more appealing to me... saves on unnecessary recategorisation and confusion, since the RD1 cat will take care of the cleaning itself. Primefac (talk) 14:00, 10 March 2023 (UTC)
It's minorish but I'd like pages that have been referred to CCI to come out of the RD1 cat as the CCI may need time to consider the editor's edits in totality and any RD1 can't be dealt with immediately. Nthep (talk) 15:15, 10 March 2023 (UTC)
I doubt we'll have pages like that going forward - the pages in the category were old and hard to deal with, and going forward it will be easier to ask the reviewer for the source if it's non-obvious. That being said, I am trying to find a way to make the two mutually exclusive. Primefac (talk) 17:28, 10 March 2023 (UTC)
Warn users who resubmit when they've made no edits
I think there should be some way to warn users who resubmit their article when they've made no edits since the previous decline, saying that if they choose to resubmit again without making any improvements that their submission will most likely be reverted or the article will simply be declined again. I feel that this might reduce the number of submissions in which the user jsut immediately resubmitted after it was declined when the user has made no attempt to address the concerns. Obviously it should either go through if they try and submit it again (if the warning should stop them from resubmitting it) or have a button that says something like "I know what I'm doing." or "Submit anyway". ― Blaze WolfTalkBlaze Wolf#6545 17:13, 9 March 2023 (UTC)
@Blaze Wolf As someone who has done that before, my motive is the expectation that anything I submit will take 3 months before anyone even sees it, so I see myself as submitting the article after three months of improvement. I would not be surprised if this is a very common attitude among draft submitters which contributes to backlog growth. Immanuelle ❤️💚💙 (talk to the cutest Wikipedian) 17:20, 9 March 2023 (UTC)
I understand that, however a user will often just submit the article (which means that they think it's ready to be reviewed) seemingly expecting a different result. You should improve the article before you submit it. ― Blaze WolfTalkBlaze Wolf#6545 17:22, 9 March 2023 (UTC)
The idea is pretty good. However, I'd rather have some way of preventing a draft submitter from resubmitting an unimproved draft. That's better than "warning" people, because I doubt that those who resubmit unimproved drafts care about/understand warnings. Best regards, --Johannes (Talk) (Contribs) (Articles) 18:04, 9 March 2023 (UTC)
I genuinely can't think of a technical way to do that. Primefac (talk) 18:06, 9 March 2023 (UTC)
The reason I propose a warning is because sometimes there is a valid reason for a user to resubmit an unimproved draft (for example if the reviewer declined the draft but another reviewer thinks it's good to accept) ― Blaze WolfTalkBlaze Wolf#6545 18:08, 9 March 2023 (UTC)
It says next to the blue 'resubmit' button "Please note that if the issues are not fixed, the draft will be declined again." If someone chooses to ignore that, they'll probably ignore other similar warnings as well? -- DoubleGrazing (talk) 18:27, 9 March 2023 (UTC)
Maybe. But maybe if something actually makes them stop for a moment then they might read it. I know some platforms actually make you read things like the Terms of service and stuff before clicking the button saying that you read it (although those are much much longer and sometimes include a bunch of legal stuff the normal user wouldn't care about and require scrolling, this would be much shorter). ― Blaze WolfTalkBlaze Wolf#6545 18:37, 9 March 2023 (UTC)
Do not expect the average AfC draft submitter to have a reasonable understanding of English – this is, at least in my opinion, one of the most common reasons why drafts are resubmitted without having their issues addressed. Best regards, --Johannes (Talk) (Contribs) (Articles) 19:03, 9 March 2023 (UTC)
That is a fair point, although most of the time the English is at least good enough to be somewhat comprehensible from my experience. ― Blaze WolfTalkBlaze Wolf#6545 15:36, 11 March 2023 (UTC)
I don't think that the average AFC draft submitter doesn't reasonably understand English; that is, I think that the average AFC draft submitter does understand some English. But I agree that there are a substantial minority who do not. They fall into at least two subclasses, those who understand some other language, and should maybe be editing its Wikipedia, and those who just completely clueless in any language. Robert McClenon (talk) 02:39, 11 March 2023 (UTC)
Alternatively we could change the "submit" button from blue to red so that it's even more tempting to press![Joke] ― Blaze WolfTalkBlaze Wolf#6545 19:08, 9 March 2023 (UTC)
See, this is why technical changes are so hard for these sorts of things. We very well could have it be so that you must make an edit before the "submit" button will show up (we could do this based on the CURRENTID or similar), but then invariably folks would figure out you just have to add some whitespace to "turn on" the button... Primefac (talk) 10:42, 10 March 2023 (UTC)
When the OP asks about a warning for resubmission without improvement, what sort of a warning do they want? I use either of two templates in AFC, which are {{noimprove}} or {{rapidresub}}. You can use them in addition to any decline text. They don't have a warning level, but they do say what I want to say to editors who resubmit without improvement or resubmit within a few hours without improvement. Robert McClenon (talk) 02:39, 11 March 2023 (UTC)
I didn't know those templates existed. I was sort of meaning a popup box or something, but I have a feeling that's not technically possible. ― Blaze WolfTalkBlaze Wolf#6545 15:37, 11 March 2023 (UTC)
User:Blaze Wolf - You didn't know that those templates existed because I never publicized them. I created them primarily for my own use because I wanted to say the same things to multiple submitters over and over. I have a lot of these templates. Maybe two of the earliest ones were {{compsays}} and {{whichmusic}}. for declining corporate drafts and musician or band drafts, respectively. They stand for: "That is what the COMPany SAYS, but notability is based on what third parties say", and "It isn't clear WHICH of the MUSIC notability criteria applies." You are welcome to use any of them. Robert McClenon (talk) 00:35, 12 March 2023 (UTC)
The templates belong to a category, Category:AfC comment templates, and are meant to be included in the comment field either on a Decline or a Comment. Robert McClenon (talk) 00:41, 12 March 2023 (UTC)
Thank you. Will save me some typing Slywriter (talk) 01:46, 12 March 2023 (UTC)
Ip reverting submissions
Hi, today i've come accross an ip 174.212.224.64 (talk·contribs) who has reverted two afc submissions which also included reverting improvements to one of the drafts by the submitter which seems to be disruptive in my view. Here are the diffs here and here. One of the drafts has been turned down a few times but the submitter has not been told by an approved reviewer to stop submitting, imv Atlantic306 (talk) 02:18, 15 March 2023 (UTC)
I agree it is disruptive but looking at the range 174.212.224.0/21, which I believe is largely the same person, it is not typical behavior. Usually they make comments or do some cleanup work on drafts, among other things across Wikipedia and they appear to be an experienced editor. Hopefully they see your note and do not do it again but still something to keep an eye out for. S0091 (talk) 17:30, 15 March 2023 (UTC)
I'm Shaun Morgan and my team wanted to start creating my page they are always went to auto rejection and deletion every time they submit the page to be published. We are a new team and we would like to seek your great assistance to get us started.
Dr. Shaun Gregory Morgan This question is kinda misplaced but I will answer you, I may move this to the main AFC Help Desk later. We prefer to call the parts of the encyclopedia articles, and not "pages", this may change your mindset somewhat. Articles are not for the benefit of the subject in any way. There may be benefits, but those are on the side and not our primary goal. Our primary goal is to summarize what independent reliable sources with significant coverage have chosen on their own to say about the subject(a person in this case), showing how they meet the special Wikipedia definition of a notable person. Wikipedia is not interested in what someone wants to say about themselves, only in what sources completely unconnected with the person choose to say about them and how they are important/significant/influential. Please see Your First Article.
I examined what you wrote(I can view deleted content) and to be frank it read as a social media style page written by a marketing team, not as a neutral encyclopedia article that summarizes independent reliable sources.
I'm not sure whom else has edited about you, but if they work for you, the Terms of Use require them to declare as paid editors. 331dot (talk) 23:11, 19 March 2023 (UTC)
List-articles are different animals, and it is my personal experience that AFC is often/usually incapable of reviewing and accepting them when they obviously do meet standards and would absolutely not be deleted by any AFD. For example, Draft:Preserved locomotives in the United States, in its current version as of 21 march 2023 is obviously valid IMHO, although it will not meet personal (non-policy) standards of some/many regular AFC volunteers. Towards allowing these to be handled in a different track, I started up Wikipedia:WikiProject Articles for creation/List-articles to list editors willing and able to handle reviewing them, and linked that from the "Reviewers" section. I only put myself as a willing party there; please feel free to add yourself if you have a basic willingness to address these. It happens that I myself am limited by an editing restriction and have had negative experiences with list-articles that I have myself created and submitted to AFC. I am myself willing to review list-articles, although I believe I will have to ask others for assistance at the end of an acceptance. --Doncram (talk,contribs) 20:33, 21 March 2023 (UTC)
Well, I was instantly reverted by User:Primefac, which is okay, as I suppose my note in the "Reviewers" section at Wikipedia:Articles for creation was too specialized for such a highly trafficked general page. And they pointed out at my Talk page that "we do have Wikipedia:WikiProject Articles for creation/List of reviewers by subject, of which 'Lists' could considered to be someone's 'interest'." That's fair enough, but I currently think it would still be okay/good to have some note on the general page, probably in the "Reviewers" section, advising people that, for certain types of articles they may wish to make requests to specific reviewers willing and able to review those.
And it remains: who is willing and able to review list-articles? Does anyone agree with me that these are different and do require a different track? --Doncram (talk,contribs) 21:03, 21 March 2023 (UTC)
Change to plot summary decline
Well since recently we've made changes to other AFC decline templates, how about this one? Currently it says:
The proposed article is not suitable for Wikipedia. Because Wikipedia is an encyclopedia, articles on fictional subjects should cover their real-world context and contain sourced analysis, offering detail on a work's development, impact or historical significance—not just a summary of the plot. You may wish to add this content to an existing article. As anyone can edit Wikipedia, you are free to do so yourself.
The issue with this is that it doesn't immediately tell the user that it was declined because it's mostly a plot summary. I think instead it should read:
The proposed article is not suitable for Wikipedia because it is mostly a plot summary. Because Wikipedia is an encyclopedia, articles on fictional subjects should cover their real-world context and contain sourced analysis, offering detail on a work's development, impact or historical significance—not just a summary of the plot. You may wish to add this content to an existing article. As anyone can edit Wikipedia, you are free to do so yourself.
I feel like at the end, instead of simply mentioning it might be better in an existing article (because, in the case of the most recent draft I declined Draft:Stars and Stripes Forever (Novel) the content may not be relevant in an existing article), it should first mention how the user should improve their draft, and then mention that if they can't it may be better to add the content to a pre-existing article. ― Blaze WolfTalkBlaze Wolf#6545 15:00, 20 March 2023 (UTC) Added formatting, no content change. Primefac (talk) 20:32, 21 March 2023 (UTC)
We should not decline a draft solely because it is mostly a plot summary. If it meets the notability criteria it should be accepted then tagged with {{All plot}} or {{Long plot}}. Otherwise it should be declined for notability, verifiability, etc. in addition to the plot summary issue more so for guidance. S0091 (talk) 20:51, 21 March 2023 (UTC)
Really? I was going to say that it would violate WP:NOTPLOT but that would be if its only a plot summary. ― Blaze WolfTalkBlaze Wolf#6545 01:14, 22 March 2023 (UTC)
WP:NEXIST. Just because an article is in terrible shape doesn't mean the topic isn't notable. Many just plot summaries are not notable but if let's say it's a Marvel movie, it will inevitably be notable and so would survive AfD. I can think of at least one WP:NSCHOLAR I've stubbed and then sent to mainspace from AfC because notability wasn't in doubt but draft was disaster.
After 6 million plus articles, whether this is still a good policy is debatable, but there is pretty much zero chance it would ever change. Slywriter (talk) 01:33, 22 March 2023 (UTC)
I think plot, much like ilc, is a supplementary decline rationale that should be used with a primary/deal-breaking decline rationale. In other words, decline as something like v with plot to indicate that the sourcing sucks, but also that the plot needs trimming. Very rarely should one expect to see a plot decline by itself. Primefac (talk) 09:30, 22 March 2023 (UTC)
Women in Red April 2023
Women in Red Apr 2023, Vol 9, Iss 4, Nos 251, 252, 262, 263, 264, 265, 266
--Lajmmoore (talk) 07:50, 27 March 2023 (UTC) via MassMessaging
'Editing consultancy' offer
After a draft was declined, a user received an offer for 'free editing consultancy' from "Wiki Submissions" at a seemingly Peru-based (or spoofed) number. I've advised them to give it wide berth. Any other action needed here? -- DoubleGrazing (talk) 06:19, 29 March 2023 (UTC)
Wiki Submissions' business model is fraud. Their method is sophisticated but yet simple: They look for drafts on living persons, and then contact these persons using their publically available contact information. This works, because in many cases, the AfC submitter and the person described in the draft are the same. (This also works for firms, quite obviously.) The reason why Wiki Submissions are doing it that way is because it doesn't create any onwiki evidence (e.g., no talk page messages, nor emails sent through the Wiki email system). Victims usually believe that the Wiki Sumbmissions emails they receive must come from Wikipedia (since most people register their accounts using email addresses). I also believe that it is very difficult for the average person to tell the difference between a legitimate Wikipedia email and a fraud email – who would expect a legitimate Wikipedia email to have no HTML formatting and a weird sender's name? Anyways, Wiki Sumbissions don't have any Wikipedia accounts. They quote an attractive price for "fixing" an AfC submission, and they ask for payment in advance. After they have received the money, they disappear. Pretty simple. Best regards, --Johannes (Talk) (Contribs) (Articles) 07:20, 29 March 2023 (UTC)
Another note: Wiki Submissions don't even get the "declined/rejected" right, they also tell people whose drafts were not declined (or even accepted) that the drafts were indeed declined… --Johannes (Talk) (Contribs) (Articles) 07:22, 29 March 2023 (UTC)
About reviewing drafts I improved by myself
Hello! Before I get to my first AfC reviews (by the way, thanks to @Primefac for giving me the green light), I just wanted to ask if I can still take over drafts I've helped improve by myself: I was specifically thinking about Taha Ayari and Lorenzo Lazzari.
[Especially in the first case, I expanded the draft significantly, so I just wanted to make sure I avoid any "conflict of interest"...]
Hi @Oltrepier, welcome to AfC! Yes, you can accept drafts you have improved. Technically, you can accept drafts you created but that is generally frowned upon. S0091 (talk) 19:41, 29 March 2023 (UTC)
Anyway, thank you very much for your help! Oltrepier (talk) 19:47, 29 March 2023 (UTC)
Just to add to the above, there is no requirement (unless there is a topic ban or similar) for a user to go through the AFC process. Generally speaking, if you are going to move a draft you heavily worked on and/or created yourself, it is preferable to just manually move it (i.e. don't "accept via AFC") to avoid the appearance of using AFC as an unofficial stamp of approval. Primefac (talk) 19:51, 29 March 2023 (UTC)
Welcome Oltrepier, some drafts are notable but bad when submitted, some reviewers just accept and tag but improving and accepting is even better if your so inspired (I find it topic dependent). I have a collection of drafts I improved to the point of being the main contributor like you did with Taha Ayari then accepted. Your edit to Draft:Lorenzo Lazzari (footballer) is relatively minor, mostly copy editing so no issues at all. As the others have said, just avoid accepting your own creations. Unlike what Primefac said I still accept via AfC even if I've "heavily worked on" so the original submiter is encouraged with the acceptance notice, but I would agree and either move or leave to another to accept if notability is borderline. Cheers KylieTastic (talk) 20:20, 29 March 2023 (UTC)
For what it's worth, I don't necessarily disagree with KylieTastic's position; working on and then accepting a clearly-notable-at-the-time-just-not-good draft is something probably most of us have done at one point or another. I was referring more to drafts that one comes across that are potentially interesting and notability is found/determined because of the AFC reviewer putting in the leg work. That starts getting into nuance and nitpicking, so I'll just reiterate my general point that no one is obligated to go through AFC ;-) Primefac (talk) 09:21, 30 March 2023 (UTC)
Redirecting after draft decline
I've just done something which made sense to me when I was doing it... and now I'm not so sure anymore. I reviewed Draft:Mohammad Eghbal Shahnavazi, which IMO wasn't notable, failing WP:BLP1E and possibly also WP:NOTMEMORIAL. However, I thought it might make sense as a redir (essentially as WP:BLAR, although I realise I'm working on a draft, not a published article), pointing to the article on the 2022 Zahedan massacre which it relates to. But as I can't remember if I've come across this before, and can't immediately find a guideline covering it, I thought I'd ask "better late than never". Views? -- DoubleGrazing (talk) 08:03, 19 March 2023 (UTC)
Declining with the "mergeto" reason might be better than redirecting from draftspace to mainspace. I usually only redirect drafts if there is a duplicate draft, and the other draft is better/bigger. Hope this helps. –Novem Linguae (talk) 11:16, 19 March 2023 (UTC)
Good point, thanks @Novem Linguae; I'll do that. Makes much more sense! -- DoubleGrazing (talk) 12:45, 19 March 2023 (UTC)
Is this discussion about redirecting a draft to an article after declining it, or about redirecting a draft to another draft? The situation that I often review is where a draft and an article exist with the same title on the same topic. The first question that I have when there are both a draft and an article is whether the draft and the article are essentially the same. If so, the next question is whether the article and the draft were created primarily by the same editor. If so, the draft should be declined, and redirected to the article (just as if the draft had been accepted). If the draft and the article were created by different people, but are the same, it is probably a copy-paste job, and a history merge is required. If the draft and the article have different or slightly different information, then I ask which of the following is the case:
The article is more complete than the draft. If so, decline the draft, and redirect it to the article.
The draft and the article each have information that is not in the other. If so, tag the draft to be merged into the article, and decline the draft. Do not redirect the draft until the merging is done.
Does any other editor have other ideas? Robert McClenon (talk) 04:55, 31 March 2023 (UTC)
Is this discussion about redirecting a draft to an article after declining it, or about redirecting a draft to another draft? Neither, it is about redirecting a Draft to a related Article.
Also, I will note that if it is a copy/paste pagemove, and the page creator is the only content editor for both copies, a histmerge is not necessary. I don't mind declining histmerges for odd cases, but if it's obvious it doesn't need a tag. Primefac (talk) 07:39, 31 March 2023 (UTC)
Maybe I wasn't clear. I was trying to say that a copy-paste requires a history merge if the draft and the article are created by different editors but are otherwise the same, and I think that is also what User:Primefac has said. Also, I was asking about declining and redirecting a draft because I think that it is a courtesy to a submitter to decline a draft before redirecting it. Also, I was saying that a draft should be redirected to an article only if the draft is a subset of the article, that is, either the same as the article or less than the article. If the draft and the article each contain unique information, the draft and the article should be tagged for merge (not history mere). I think that some editors may not understand that. Robert McClenon (talk) 15:18, 1 April 2023 (UTC)
Ditto, Earwig's copyvio detector. Already missing it desperately. -- DoubleGrazing (talk) 12:38, 3 April 2023 (UTC)
Draft redirects
If the page "Draft:Foo" is reviewed and approved it is moved to "Foo" in mainspace, and "Draft:Foo" becomes a redirect to it. Should it be left that way, or should the redirect be deleted? Cambalachero (talk) 17:24, 3 April 2023 (UTC)
I noticed that using the Template:Draft categories actually makes hotcat add categories into the template.
It only does so when a category is already present though
Do you think we could change the draftify script to put all the categories in this template instead of turning them into links, and possibly somehow transclude draft categories into Template:Draft or Template:AfC submission in order to make it so that people can edit categories more naturally and not lose them in the process of drafting? Immanuelle ❤️💚💙 (talk to the cutest Wikipedian) 03:09, 4 April 2023 (UTC)
Update of instructions to reviewers
I'm asking for help about a very frustrating problem that comes up repeatedly. The instructions to AfC reviewers at WP:AFCSTANDARDS still contain the following advice: "Avoid declining an article because it correctly uses general references to support some or all of the material. The content and sourcing policies require inline citations for only four specific types of material, most commonly direct quotations and contentious material (whether negative, positive, or neutral) about living persons."
This advice might once have been correct, but is now completely at odds with what reviewers do, it's against Wikipedia's policies, and it merely misleads those of us who edit, but do not review, should we come across it. There is no point in pretending that general referencing is permitted. It isn't. Elemimele (talk) 12:05, 5 April 2023 (UTC)
Elemimele although I don't like the use of general references I was not aware there had been and policy change. It is still listed in WP:GENREF and Wikipedia:Inline citation still does not says require in all cases. What policy are you referring to with your claim "It isn't"? Regards KylieTastic (talk) 12:31, 5 April 2023 (UTC)
I assume this is is reference to the comment on Draft:Lothar Abel. Unfortunately while they are technically valid they do make reviewing much more difficult unless it's just a stub. For the same reason it also makes if difficult for readers to verify which is why i do not like them. The comment that Greenman left does not actually say they need to be inline, but it's easy to see that interpretation, however they could have just been saying that your general sources they could not access should cover all points. I do agree that the official policies do no match how many apply them and it would be preferable if we dropped the use of general references, but that is not something for this project but the wider community. Cheers KylieTastic (talk) 12:49, 5 April 2023 (UTC)
I hear you, and I don't disagree, but bear in mind that part of the reviewer's job is to check that the draft is reliably sourced, and it can be very difficult to do that if the author hasn't indicated at all which of the sources provides which bit of the content (and therefore, how much of it is just OR/synth, or entirely made up even), especially when citing offline sources. So yes, general refs are accepted, as are offline sources, as are non-English ones... but if all these are pushed to the limit, at some point it becomes nigh-on impossible to verify the information, and a reviewer may just give up and move on. (This is just a personal observation, without commenting on policy.) -- DoubleGrazing (talk) 12:32, 5 April 2023 (UTC)
Hi KylieTastic and DoubleGrazing, it's not just a response to Greenman's comments on Lothar Abel, though prompted by this. It's something I've noticed generally, and not just in my own translations (I can't find the link, but it came up recently in the teahouse with another translator-from-German too, who was giving the same advice: don't even attempt general referencing. They were again pointing out the difference between German and English WPs).
The problem is an interaction of AFC and the core policy, which I quote: "All material in Wikipedia mainspace, including everything in articles, lists, and captions, must be verifiable. All quotations, and any material whose verifiability has been challenged or is likely to be challenged, must include an inline citation". Thus if AfC reviewers habitually challenge everything that doesn't have an inline citation, it has been challenged, and therefore must have inline citation, by definition.
I totally accept that without inline citations, you'd need to read the entire general source to check whether each fact is verifiable, which is too much for any volunteer. I think, in any case, you have an impossible task: if you doubt that the person who wrote the original article actually read the sources they cite, there's no guarantee their inline citations are any more reliable than their general sources. DoubleGrazing is correct that there's a risk of the checks being nigh-on impossible; if you pursue this to its logical conclusion, AfC can't accept anything unless the reviewer has managed to track down and understand a reliable, scholarly and respected book written in Lithuanian in 1970; Wikipedia would degenerate into a collection of facts available electronically, with open access, in English. A bad outcome. I don't know the answer. Elemimele (talk) 14:55, 5 April 2023 (UTC)
FYI, Template:Ifu w(edit | talk | history | links | watch | logs) has been nominated for deletion. It appears it was never migrated to the newer {{Ffu b}} template set. Though, the FFU template set should probably also be updated to match the corresponding Afc template (ie. {{Ffu c}} equivalent for {{Afc c}} ). And {{ifu c}} should probably be merged into {{afc comment}} -- 64.229.90.172 (talk) 09:40, 7 April 2023 (UTC)
Since I'm an admin, I can delete articles. However, when I'm accepting a draft via AFCH, if the destination is occupied, I have to go manually delete the destination. It would be handy if AFCH could see my admin bit and realize that it can use my perms to delete the destination. Perhaps give a checkbox or somehting to acknowledge that the destination will be deleted? - UtherSRG(talk) 17:43, 7 April 2023 (UTC)
Sure, good idea. I've created a feature request. Although we should probably prioritize #243 first, a similar issue that is asked about multiple times a year here. –Novem Linguae (talk) 18:08, 7 April 2023 (UTC)
Isn't 243 basically the same thing? Non-admins can't move over redirect anyway, and we shouldn't be accepting drafts over existing articles... Primefac (talk) 18:20, 7 April 2023 (UTC)
I wrote a draft that ties into something I want to bring forth to WP:DYK this Easter...
...and given the rising backlog the team's facing, how do I (or whom should I contact to) bump it up the priority queue if that's even possible? (Subject in question: Lonzo Anderson (draft) [1905–1993], author of Two Hundred Rabbits [1968]; preferred deadline, Sunday or Monday.) To @Theroadislong:/@S0091: Any ideas? --Slgrandson (How's myegg-throwing coleslaw?) 01:21, 9 April 2023 (UTC)
I know one can't put the cat back into the bag in this particular case, but I would strongly discourage doing this sort of thing in the future; we should not be encouraging people to request a review to fit a deadline (or just a review in general), because soon it will mean this talk page will be flooded with requests for review. Primefac (talk) 07:19, 10 April 2023 (UTC)
Fair enough. I'll add that I'm happy to be asked about literature/authors/academics drafts personally (ie, on my Talk page), provided I haven't already reviewed them. -- asilvering (talk) 16:12, 10 April 2023 (UTC)
New backlog drive?`
With pending submissions standing at almost 4,000, I can't help but think we really do need a new drive. As many articles are over 2 months old, could I suggest a more motivating points system such as:
1 point per review
+ 1 for 2 weeks to 1 month (4 weeks) - 2 points
+ 2 for 1 month (4 weeks) to 2 months (8 weeks) - 3 points
+ 3 for over 2 months (8 weeks) - 4 points
No points for a review which failed re-review (same as Jan backlog drive)
Anyone have any thoughts on this? Mattdaviesfsic (talk) 07:26, 31 March 2023 (UTC)
We don't need drives, we just need a larger contingent of regularly-active reviewers. The last drive was great, but if you look at the graph it did nothing to change the rate of new open drafts after it was over (and if anything, it got worse). We should focus less on "let's bash out as many drafts as we can in as short a period as possible" and focus more on a consistent level of activity. I would much rather have a perennial 2k drafts pending than try and always be reaching 0. Primefac (talk) 07:37, 31 March 2023 (UTC)
The July 2021 backlog drive was a success, albeit still suffering from the lack of consistent reviewing after - but it did clear the very old and rose again slower after. The January 2023 backlog drive I think was a failure as we already had a failure level of reviewing before, that just continued after. I feel I've been pushing harder after to try to slow the growth (and depressingly failed), so it feels like the last backlog hurt the project. So if done well it can work, but only if enough good reviewers join in. However, I think unless we get the review/submission rate to level out first it's not worth it. I'm on a break anyway with too many real-life issues so unlikely to participate, also I've done enough reviews this year and just burnt out. KylieTastic (talk) 08:50, 31 March 2023 (UTC)
Right, that's kind of what I mean; if you look at 2022 we were pretty steady between 2k-3k pages in the queue. I'll be honest, I am 90% okay with that. The sudden uptick in submissions (and/or the lack of reviewing) in December led to panic, which led to the backlog drive, which didn't solve the issue of a lack of a lack of reviewing to match the submission rate and so here we are back at the same point. You (Kylie) shouldn't have to carry the burden entirely by yourself, nor should any of our usual "top 10" editors. We've just never found a good way to motivate people to basically do one of the most tedious and soul-crushing jobs on Wikipedia. Primefac (talk) 09:16, 31 March 2023 (UTC)
Agree with all your points Primefac. For a long while there we were steady with 2.3-2.5k...not bad. I, like Kylie have slowed down because of burnout and I have never reviewed nearly the volume they do. How do we get more active reviewers though? I think a newsletter would help but the newsletter audience is current reviewers so limits the outreach. S0091 (talk) 17:29, 31 March 2023 (UTC)
Actually, maybe we should have a drive focused on getting more reviewers. S0091 (talk) 17:36, 31 March 2023 (UTC)
Backlogs in any realm are interesting. My primary doctor's office always has a one-month backlog. (I know that a doctor's office is much closer to a FIFO queue than Wikipedia article reviewing is.) I presume the rate at which people make appointments, and are seen, is about equal. If not, the backlog would grow or shrink. Maybe this is self-limiting, since those who are told the wait is six months would presumably find a different provider.
I fantasize about a situation where doctors are "as busy as they want to be", but appointments could always be had the next day.
Is it odd that Wikipedia can reduce the article backlog with a drive, but can never seem to keep the pile small? I know that reviewing is hard work. David10244 (talk) 08:20, 11 April 2023 (UTC)
Has any detailed analysis been done on drives beyond just looking at numbers in the queue? Does it bring in new reviewers? Do they stay around afterwards? Does it reactivate inactive ones? What are the impacts on active reviewers? If they are more active during the drive, and less active afterwards, how much so? Is there a net benefit? Greenman (talk) 10:32, 31 March 2023 (UTC)
I suspect that backlog drives net out to the same amount of articles reviewed, just that it causes the graph to go way down, then back up due to post-drive burnout, then it levels out again. However, even if backlog drives vs no backlog drives zeros out to the same number of articles reviewed, I like backlog drives due to their other benefits. Mainly that it fosters a sense of community among AFCers. They are fun, receiving barnstars is fun, they encourage friendly competition, the feeling of reaching zero articles in the queue is really cool, etc. One of my favorite Wikipedia memories was participating in the 2021 AFC backlog drive, where I did 225 reviews in a month, and we got the backlog from 5000 to zero. –Novem Linguae (talk) 17:02, 31 March 2023 (UTC)
Apologies for going off on a tangent, and I don't know if this has come up before, but: could we put a limit on how long a draft can sit in the pool? By which I mean, if a draft has waited for, say, 30 days (since its initial submission, or most recent resubmission) without anyone picking it up, it gets automatically released into the wild and goes on the NPP list instead. It seems to me that the worst dross gets often dealt with on the day of submission, and the persistent problems with multiple declines are on several reviewers' watch lists anyway, so the chances of anything too horrendous slipping through are probably relatively low. (And maybe some sort of extra safety mechanisms could be implemented, like preventing automatic release of drafts flagged up as COI/UPE, non-English content, blanks, protected titles, etc.) I realise this wouldn't solve the underlying problem of too many drafts coming through and there being not enough active reviewers, and NPP probably wouldn't be too happy about this, but from the draft creators' point of view it would at least put a backstop on the long waits and stop the complaints about AfC being just a big black hole where drafts go to die. Thoughts? -- DoubleGrazing (talk) 06:44, 2 April 2023 (UTC)
It seems a bit harsh to hand over a lot more work to NPP, when they're only able to pick up that slack after having just recently done a lot of work to clear their own overlarge queue... -- asilvering (talk) 20:36, 2 April 2023 (UTC)
AFC and NPP want the same thing in the end. Both are fundamentally just measures to assist with detecting problems early. Articles that are unreviewed for 90 days are still indexed by Google. I'm hopeful that the tooling improved for NPP will improve, and only provide more recognition and retention of patrollers. ~ 🦝 Shushugah (he/him • talk) 21:02, 2 April 2023 (UTC)
DoubleGrazing I have considered this as a possible improvement for a couple of years (with suitable protections from abuse). Draft/AfC should be a control not a block and after a while let the consensus rule. For those that think Draft/AfC is required to stop junk easily getting to main-space it still works; for those that hate Draft/AfC and think we avoid the consensus of AfD it works beter. On my more optimistic days I've thought about starting an RfC but in reality I think it would just fail in Wikipedia bureaucratic resistance to change. If it was going to pass it would need a robust bot to check submissions: check it has really been listed for x days; had not been tagged/declined for key issues before (even if removed); a way to mark submission to not be auto promoted; not submitted by editors with restrictions; etc... I would also start at longer than 30 days, maybe 3 months, and then adjust down as required. As long as people can tag to not auto promote (the opposite of {{Promising draft}}) then it just turns us into a pre-approval system rather than a blocking system. Reviewers who are concerned would just have to check the new 0 day category at the other end of the queue and submitters would know a max wait time that would remove a key negative impact of the current system. Cheers KylieTastic (talk) 21:00, 2 April 2023 (UTC)
I think this could actually be a really good idea with some work-shopping. My only concern is with COI editors whose drafts shouldn't be accepted without actual review. But I think a a reason a lot of the drafts sit for more than two months is because they are borderline notable, making it hard to decline, but also hard to accept without fearing criticism. Exposing them to mainspace might be the best way to resolve the issue.
Of the ~200 submissions per day, only about ~40 per day make it 30 days, so it wouldn't add too much work to NPP. I think a lot of the articles sitting at the end of backlog have been seen by many many reviewers, and just no ones wants to decline or accept. Galobtter (pingó mió) 21:21, 2 April 2023 (UTC)
I like this idea as well but I agree it needs to be workshopped here first. If it seems there is consensus among AfC reviewers to move forward, then start a discussion with the broader group of NPP reviewers to get their view (some AfC reviewers are NPP as well so I expect we will get some NPP input beforehand). Then only after that, depending on the feedback, start an RfC. Thoughts? S0091 (talk) 22:06, 2 April 2023 (UTC)
That seems like a good plan - I think this is a case where any premature RfC would likely fail quickly and prejudice future reform attempts. Galobtter (pingó mió) 03:16, 3 April 2023 (UTC)
FWIW, I'm also at NPP, and I certainly wouldn't want to unduly increase the patrolling workload as I think NPP is already more arduous than AfC anyway. However, going to patrol is the default position for any new non-autopatrolled articles that don't go through AfC, so NPP was always going to be busy no matter what we do here. -- DoubleGrazing (talk) 10:28, 3 April 2023 (UTC)
I think we have to survey why many of us are not actioning on the older drafts first. – robertsky (talk) 03:25, 3 April 2023 (UTC)
Speaking for myself as a not very active reviewer, I think they are generally borderline drafts that prompt skipping to the next one rather than dealing with it. Something like, to pick a random example from the 2 months queue, Draft:Choo Mandhirakaali, has some reviews so might be notable, discouraging a decline, but also has some issues like the plot section and the references being non-English (making them harder to review), discouraging an accept. Galobtter (pingó mió) 05:06, 3 April 2023 (UTC)
Personally speaking, when I review something at the back of the queue, I will generally spend at least 10-15 minutes on the page. As mentioned above, they are often either borderline notable or have issues that a quick check cannot simply rule out (e.g. lots of low-quality refs with maybe a single good ref, native advertising, etc); it's just not a quick job to do, and even at my most active (and most ruthless) I could maybe get a half-dozen pages reviewed in an hour. Primefac (talk) 10:07, 3 April 2023 (UTC)
I'm sure there are many different approaches to personal work order and we need a mixture. When active I first try to review all 0-day submissions for attack pages, copyvios, spam, promo, personal info of youngsters, easy basic declines: no refs, blank, exists etc. If I have any energy/time left I treat the rest of the pile as one and work on subjects I'm most comfortable with first and I hope together we cover all the subject areas. I generally have no interest in most BLPs and organisations which make up much of the end of the queue so yes I do less of the oldest. KylieTastic (talk) 09:14, 3 April 2023 (UTC)
@Galobtter: another suggestion which I've floated here before and which got some support (but may have died a quiet death since?) was to give the reviewer a way of indicating their confidence in the draft when they accept it, sort of like 'borderline, and I realise it's borderline, but giving it the benefit of the doubt' vs. 'absolutely solid, no issues at all, doesn't even need patrolling'. This might make it easier to clear many of the drafts that otherwise end up sitting there for weeks or months on end, where nobody wants to put their 'reputation' on the line and accept a borderline case, but doesn't really find a solid reason to decline, either. -- DoubleGrazing (talk) 10:33, 3 April 2023 (UTC)
I might have said the same thing the last time you brought this up, but this is what talk pages are for. When I accept a borderline draft I will generally leave a talk page note to that effect, usually along either something like your idea or along the lines of "please don't take AfC acceptance as AFD-proofing it". I think adding "add a talk page message about the accept" to AFCH would just make for too much clutter, especially since it would likely be used somewhat infrequently, but if folks think it's a good idea I'm not going to oppose it. Primefac (talk) 14:34, 3 April 2023 (UTC)
I think this is a good idea; could be a checkbox in the accept stage. Primefac is right in that theory you can just leave a talk page note. But I think this is more of a social thing (as you mention with the "reputation" thing): a newer AfC reviewer might be worried about accepting a borderline draft if it's not "officially" supported/endorsed by the script. I personally don't care if something I pass through AfC is AfDed or someone complains about my accept, but also I don't have to worry about someone removing my AfC perms/really making an issue of it. Plus having it in the script removes a hurdle to accepting these borderline drafts. Galobtter (pingó mió) 18:48, 3 April 2023 (UTC)
Split from my original, as it pertains to a few different threads A year or so ago Enterprisey and I were discussing the possibility of setting up a multi-person review system, where we would have an AFC-reviewer-only checkbox system where we could split up the work. For example, I could review the references (good/bad/somewhere in the middle), someone review copyvio (yes/no) and/or prose (tone, primarily), etc. If a page passed each criteria, then it could be approved. A single reviewer could do all of these steps, of course, but for some of these larger/older pages it might make the work a bit easier to manage. If there's enough interest in pursuing this avenue of thought further, we can probably resurrect the idea. Primefac (talk) 10:07, 3 April 2023 (UTC)
Great. I think there's a free one at https://gptzero.me/ (scroll down a bit to see the form). I'd be interested in if there's differences between that public form and the API. I'd also be interested in how accurate the detector is. Please keep us posted. –Novem Linguae (talk) 16:48, 30 March 2023 (UTC)
@Qwerfjkl Why the green text on certain letters? David10244 (talk) 08:09, 11 April 2023 (UTC)
Given the sample human text, it returns:
@David10244, I messed up the syntax highlighting. — Qwerfjkltalk 09:28, 11 April 2023 (UTC)
{'documents':[{'average_generated_prob':0,'completely_generated_prob':4.188728382629251e-08,'overall_burstiness':81.27423095703125,'paragraphs':[{'completely_generated_prob':0.11111111111111108,'num_sentences':1,'start_sentence_index':0},{'completely_generated_prob':0.11111111111111108,'num_sentences':1,'start_sentence_index':1},{'completely_generated_prob':0.11111111111111108,'num_sentences':1,'start_sentence_index':2},{'completely_generated_prob':0.11111111111111108,'num_sentences':1,'start_sentence_index':3},{'completely_generated_prob':0.11111111111111108,'num_sentences':1,'start_sentence_index':4},{'completely_generated_prob':0.11111111111111108,'num_sentences':1,'start_sentence_index':5},{'completely_generated_prob':0.11111111111111108,'num_sentences':1,'start_sentence_index':6},{'completely_generated_prob':0.11111111111111108,'num_sentences':1,'start_sentence_index':7},{'completely_generated_prob':0.11111111111111108,'num_sentences':1,'start_sentence_index':8}],'sentences':[{'generated_prob':0,'perplexity':20,'sentence':"Climate change has likely led to the decline of some of Scotland's mountain plants, according to new research."},{'generated_prob':0,'perplexity':71,'sentence':'Scientistssaidmanyofthespeciesreliedonsnowcoverremaininghighonhillsuntillatespringandevensummertoensureamoistenvironment.'},{'generated_prob':0,'perplexity':112,'sentence':'Theyalsosaidplantsthatthrivedonlowergroundinwarmerconditionswerespreadingtomountainhabitats.'},{'generated_prob':0,'perplexity':169,'sentence':'Speciesfoundtobeindeclineincludesnowpearlwort,alpinelady-fernandalpinespeedwell.'},{'generated_prob':0,'perplexity':26,'sentence':'TheresearchbytheBotanicalSocietyofBritainandIreland(BSBI)hastaken20yearstocompleteandhasbeenpublishedinthenewPlantAtlas.'},{'generated_prob':0,'perplexity':61,'sentence':'Datausedtoproducethereportincludedmorethanthreemillionplantrecordsof2,555speciescollectedbyhundredsofbotanistsacrossScotland.'},{'generated_prob':0,'perplexity':25,'sentence':'Climatechange,habitatlossandthespreadofnon-nativespecieswerefoundtokeythreatstothehealthofBritishandIrishnativeplants.'},{'generated_prob':0,'perplexity':265,'sentence':'BSBIsaiddevastatinglossesofspeciesinScotlandwereamongthefindings.'},{'generated_prob':0,'perplexity':130,'sentence':"Almost the entire British population of snow pearlwort is found on Ben Lawers, but half of the Perthshire mountain's known colonies have disappeared over the last 40 years."}]}]}
I just posted this on Wikipedia talk:Large language models, but noticed this discussion here as well. I independently did some testing using one of my own articles - it variously returned "likely entirely AI" (lead only), "likely partly AI" (lead plus early sections, or early sections on their own), and "likely entirely human" (later sections). I don't know what's causing the discrepancy, but it makes me a bit worried about how accurate these are, and whether they're perhaps being thrown off a bit by WP style. Andrew Gray (talk) 17:01, 1 April 2023 (UTC)
Considering ChatGPT has been trained on a huge amount of Wikipedia text, I am quite skeptical of its ability to distinguish between human and AI created text for Wikipedia articles. Trainsandotherthings (talk) 21:08, 1 April 2023 (UTC)
I think a much more useful tell will be "did the original draft have any citations". If no, and they're added after... that's a big LLM tell. Otherwise LLMs make up citations, so simply checking them will help. -- asilvering (talk) 20:37, 2 April 2023 (UTC)
Draft:Unnecessary surgery
Would appreciate another pair of eyes on this Draft:Unnecessary surgery. First, it looked like a polemical essay to me. Then I noticed that several of the refs are to the website of a law firm that deals in medical malpractice etc., so I thought maybe it's just admasq. But there are also a couple of secondary sources and a number of academic papers cited, which suggest that there could be some notability to this, although even then, I'd say there is almost certainly OR/synth included. I don't think it's acceptable as it stands, but beyond that, I'm not sure quite what to do with it? -- DoubleGrazing (talk) 07:41, 12 April 2023 (UTC)
Essay-ish. US heavy. May require a rewrite. The topic is definitely not insignificant on the international level, i.e. India, UK, etc., and on certain demographics, i.e. intersex children. – robertsky (talk) 09:52, 12 April 2023 (UTC)
I don't think it's possible for this to be done without violating some form of OR/synth/npov. Otherwise, these items belong according to topic: c-sections, MRI, etc. -- asilvering (talk) 20:31, 12 April 2023 (UTC)
"Undeleted edits"
Hey there! I'm looking to become a participant, and the criterea state I have to have "500 undeleted edits". How do I check the number of undeleted edits I have? If at all? And what does it mean by undeleted edits? Current edits or ones that were reworded, or what? AugustusAudax (talk) 21:05, 11 April 2023 (UTC)
Check here: [3]. As I write this, you have 528 undeleted edits. -- asilvering (talk) 21:13, 11 April 2023 (UTC)
A deleted edit is one made to a page that has since been deleted. — Qwerfjkltalk 21:29, 11 April 2023 (UTC)
For clarification, the required is 500 undeleted edits to articles, a count of which for the OP currently sits at 345. Primefac (talk) 09:32, 12 April 2023 (UTC)
Do draft edits count? If not, is there some reason it doesn't say "edits to mainspace" instead of "edits to articles"? -- asilvering (talk) 20:28, 12 April 2023 (UTC)
No, those are Draft edits. No to your second question, other than potentially the fact that "article" is much more clear than "edits to mainspace". Primefac (talk) 16:48, 13 April 2023 (UTC)
John Pork
Just a heads-up, I'm seeing quite a lot of crap lately coming through AfC on this silly John Pork meme. Managed to get some of it deleted, others not. Apparently TikTok have blocked them, so I don't know if the idea is to target Wikipedia next. -- DoubleGrazing (talk) 17:36, 13 April 2023 (UTC)
Accepting article to redirect?
Hi, I'm trying to accept Draft:O'Clock (album), but O'Clock (album) already exists as a redirect. How do I go about dealing with this? Cheers! :3 F4U (they/it) 22:42, 13 April 2023 (UTC)
@Freedom4U: you can tag the mainspace redirect with {{db-afc-move}}, and an administrator will probably delete it within an hour or so. — Ingenuity (talk • contribs) 23:08, 13 April 2023 (UTC)
Thank you! :3 F4U (they/it) 23:13, 13 April 2023 (UTC)
Adding a scam warning directly onto the decline template?
I'd want to see evidence of how prevalent this is in regarding to declined AfC submissions, since every additional bullet point means people are less likely to actually read the decline notice. Galobtter (talk) 04:23, 14 April 2023 (UTC)
Agreed. I don't think we're at the point of too much information in the decline notice, but I genuinely have no idea how we'd cludge in a "oh by the way don't pay for help, it's a scam" into our bullet points. Our warning is already pretty large on the main WP:AFC page. Primefac (talk) 09:09, 14 April 2023 (UTC)
I think Jimbo's acquaintance fell for a scam that had to do with deletion, which may not have involved drafts or AFC at all. The exact quote from ARC is the mark has had their biography deleted several times over the past decade and was easily duped into the belief that they were paying the big bucks to buy off administrators and arbitrators. Adding something to AFC declines may still be a decent idea, but just wanted to point this out. –Novem Linguae (talk) 05:39, 14 April 2023 (UTC)
Professors
Draft:Anne-Marie Chang. I cleaned this one up a bunch. I'm pretty much ready to accept it. With an h-index of 31, probably passes WP:NPROF. My hesitation is the only sourcing is the university website's bio, and her scientific papers. I'm a bit hesitant to accept a BLP that is 1 bio website and a bunch of non-independent sources. Think we should accept this one, or are we supposed to require better sourcing? Thanks. –Novem Linguae (talk) 08:40, 11 April 2023 (UTC)
Depends on who you ask. There are some that believe once PROF is met it doesn't matter how many non-primary sources there are. Ideally I would like to see a couple of secondary sources, but honestly if you think PROF is met I wouuld chuck it into the Article space and let the masses deal with it. Primefac (talk) 09:02, 11 April 2023 (UTC)
The instructions for AFC advise us to accept a draft if we think (subjectively) that there is at least a 51% chance that the article will be kept on AFD. There are a number of editors who are likely to say 'Keep - Passes PROF', so that it is reasonable (if a little lazy) to accept. (And being lazy on one draft gives time to review other drafts or improve articles.) Robert McClenon (talk) 16:49, 16 April 2023 (UTC)
Talk about timing!
I've just picked up a draft that's been sitting there for three months, Draft:Intellectual Property (album). Purely by coincidence, it's about a new album that came out yesterday. I was going to accept it... but as it happens, only a few hours ago someone has independently of this created Intellectual Property (album). So I had to decline the draft. I expect its author won't be best pleased. :( -- DoubleGrazing (talk) 06:41, 15 April 2023 (UTC)
I suppose that's not about bad timing or bad luck; Wikipedia is an encyclopedia, which means that by principle, Wikipedia describes what is considered established knowledge. Now, what is known about something unreleased, i.e., something unknown? That should make it obvious, but people not understanding "encyclopedia" is a never-ending story I reckon. Best regards, --Johannes (Talk) (Contribs) (Articles) 15:51, 15 April 2023 (UTC)
Yeah, I was mainly just commenting on the timing of it all. A few hours either way, and it could have gone differently. -- DoubleGrazing (talk) 16:53, 16 April 2023 (UTC)
Rules for draft and article on same title
As Johannes Maximilian says, this is not an uncommon situation, to have a draft by one author and an article by another, often about music or films, which have release dates. My rules for dealing with this situation are:
First, check whether the article appears to have been copy-pasted from the draft by a different editor. If so, tag the article to have the history of the draft merged in. If the copy-paste was done by the original author, it is not a history merge situation, but an editor creating two copies, one in draft space, the other in article space, which is a different issue.
Second, check whether the draft is a subset of the article. That is, is everything that is in the draft also in the article? (If two sets are equal, each is a subset of the other.) If all of the information that is in the draft is also in the article, redirect the draft to the article.
Third, if the draft contains at least some information that is not in the article, tag the draft to be merged into the article (or tag the article for a Merge-From of the article).
In this case, the draft has a Singles section that is not in the article, and I have tagged the draft to be merged into the article. (So the author of the draft can get attribution of edits to the article.)
I hope that this is clear. Robert McClenon (talk) 16:44, 16 April 2023 (UTC)
I'm not sure why the article was rejected. It has 2 inline referenced that meets the 4 criteria that is mentioned in the rejection notice...Pvmoutside (talk) 18:34, 17 April 2023 (UTC)
It was declined NOT rejected, two mentions in poor quality sources is not enough to show any notability. Theroadislong (talk) 18:41, 17 April 2023 (UTC)
I'd like some people to take a look at this article. It seems pretty good, however there's a 2 sentence paragraph about his membership of the Church of Jesus Christ of Latter-day Saints and him baptizing his wife. I'm not aware of any guideline in WP:BLP relating to how a person's religion should be covered in the article if they've publicly disclosed it so I'm not entirely sure how it should be handled here. ― Blaze WolfTalkBlaze Wolf#6545 19:15, 17 April 2023 (UTC)
Maybe just apply normal Wikipedia policies to the paragraphs. Is the paragraph WP:DUE? Is it sourced to WP:RS? The source is an interview in Deseret News, which is generally reliable at WP:RSPSOURCES and is certainly better than, say, Twitter. Up to you if it's DUE. An interview is not independent and would certainly not work for GNG, but may be OK for sourcing their religion. Hope this helps. –Novem Linguae (talk) 20:33, 17 April 2023 (UTC)
South Asian location stubs
Wonder if there's an editathon on? I've declined several South Asian stubs on streets/localities today. All just a single paragraph of text, no refs, no links, nothing else. Alternatively, could be some sort of AI thing, I guess? -- DoubleGrazing (talk) 16:14, 19 April 2023 (UTC)
A History Merge Question (again)
I have another question about history merge and tagging for history merge. I know that if I see what looks like a copy-paste from draft space to article space, but the author of the draft was also the person who did the copy-paste into article space, there is no need for a history merge. In that case, the history merge is not required, because there is no attribution problem. If I encounter a draft and an article that are almost identical, both the work of the same person, where there is a history merge tag, the tag has probably been applied in good-faith error, a misunderstanding of when history merge is needed. Should I remove the history merge tag, or just leave it to let the administrator decline the history merge? Robert McClenon (talk) 04:09, 20 April 2023 (UTC)
If you feel confident enough to remove the tag, then feel free to do so. If not, feel free to leave it for an admin, which will usually be me, so you can also ping me if you want a 2O (I try to get to them at least once a day anyway, but sometimes I forget). Primefac (talk) 07:38, 20 April 2023 (UTC)
As anyone patrolling the help desk knows, a large proportion of the threads are blanks, where the user doesn't even ask a question, they just open up a new thread (often on a rejected draft), prompting the routine "you don't ask a question, but yada yada", which usually gets no response. To plug this particular time sink, is there any way of configuring the template so that it won't allow publishing until something has been entered into the question field? (We might still get random gibberish, of course, but at least not complete blanks.) -- DoubleGrazing (talk) 16:04, 5 April 2023 (UTC)
Hmmm...the instructions are different depending on which link is used, with the link from the decline notice being less clear (here) than the "Click Here to ask a question" on the Help Desk page (here). I think we should shore up the instructions and see if that helps. S0091 (talk) 17:02, 5 April 2023 (UTC)
@Novem Linguae not sure if you are the correct person to ask, but any clue about the reason for the pink boxes (in either example above) with the "If you see two empty white boxes below..." then do blah, if something else then do blah. It's quite confusing and I wonder if its needed or if it an be improved somehow. S0091 (talk) 19:48, 7 April 2023 (UTC)
Or... ask someone in the project who is template savvy...? Primefac (talk) 20:10, 7 April 2023 (UTC)
Lol, I was just thinking about pinging you. Is all of that instruction necessary? If I were a new editor seeing that, I would just give up or hit publish and cross my fingers. S0091 (talk) 20:17, 7 April 2023 (UTC)
I apparently missed the original posting of this thread, and I'm headed off in a mo', but I'll check in the morning and give my slightly-more-awake thoughts. Primefac (talk) 20:36, 7 April 2023 (UTC)
Okay, so we have two different preloads (Wikipedia:WikiProject Articles for creation/Help desk/preload and {{AfC decline/HD preload}}) which are somewhat doing two different things; the latter (since it's coming from the decline notice) preloads the page name, which does make for slightly-less-confusing editing when you look at the examples given above. There could probably be a bit of standardising there, but since there's a difference in what is being passed to the templates we'd still need different messages.
As far as the edit notice goes - it has two sections because one could theoretically edit the AFCHD directly, and thus get the standard editing window (meaning they would need to fill out the section header instead of it being preloaded). I think that's where the majority of the wordiness comes from in the edit notice. If we don't ever have people editing AFCHD directly to ask new questions, we could just remove most of that, but I don't work that desk so wouldn't be able to say.
Just because of how many things we need to pass to the preload pages, I'm honestly not sure how much we can simplify - if they include nothing, you'll either subst an error message (which was my initial thought last night) or subst nothing (which is what we're trying to fix). I do suppose a "you didn't add a question, please do so or this will be removed" notice would catch people's eyes (and give us an excuse to just remove the question immediately). Primefac (talk) 07:38, 8 April 2023 (UTC)
I think there is another preload for rejected drafts, see this and I think this is the one that is causing the biggest issue because most of the ones with no question has decline=Draft:XXXX and largely they are rejected drafts. Also, looking at others they all seem to be a preload rather than AFCHD directly. Can we try changing the edit notice to flip instructions 1 and 2 because most likely there is a preload so no need to read through the manual instructions? Then also in the preloads, change the type to all caps for the portion of the instructions about asking a question. For example, change "Below this line, tell us why you are requesting a re-review. Take as many lines as you need." to all caps and maybe add that if they do not leave comments, it will be ignored or something like that. @DoubleGrazing (or others) what do you think? S0091 (talk) 14:39, 8 April 2023 (UTC)
I don't really understand such things, so can't comment. If you think it would reduce the number of blanks coming through, that works for me!
Maybe even small cosmetic changes could help, such as replacing that bright blue 'Ask for advice' button in the rejection tag (which is almost demanding that you click on it!) with just a plain text link? -- DoubleGrazing (talk) 14:59, 8 April 2023 (UTC)
I don't understand templates either but I can follow bits of the code. I also agree we should change the button to text, like is used in the rejection notice that is placed on the user's talk page: If you have further questions, you can ask at the Articles for creation help desk, which launches {{AfC decline/HD preload}} (i.e. the Ask for Advice button in the reject message on the draft uses a different link than the reject message on the user's talk, see Draft:Recipes in Odia vs. User talk:SUNANDA.ODISHA). If we switch to using the same message/link that is used on the user's talk page, I think we could eventually retire {{AfC submission/declined/HD preload}} (maybe?).
@S0091, @Novem Linguae I agree that the instructions saying "If you see two empty white boxes below..." are confusing. And preventing the "no question was asked" submissions would be great. It might make the submitter focus on what they want to know, and they might even read the linked keywords in the decline notice.
I think it's not completely obvious that the blue words (in a decline notice, or in a help desk answer) are clickable links. I speculate that some new editors might think the color is for emphasis. That's why I sometimes use something like "reliable (click here)" to make it blindingly obvious.
Can the submit template for afchelp say "did you click the colored words and read the linked information?". Many of the questions imply that no decline explanation was provided, which is mostly true if they don't realize they can click. And, yay for @DoubleGrazing starting this discussion. David10244 (talk) 07:42, 12 April 2023 (UTC)
Just noting for the record that there are about four different overlapping issues here, each of which will require a different solution with varying difficulties. I'm going to split them into separate threads to deal with each one, but if I've missed one please feel free to add it as a new subsection. Primefac (talk) 09:53, 12 April 2023 (UTC)
@Primefac Thanks, a lot of great progress in a short time. Will be great to see reduction in bkank submissions. David10244 (talk) 19:18, 13 April 2023 (UTC)
Changing the edit notice
The edit notice for the AFC Help Desk is as follows:
This page is for questions about the Articles for creation process. Please consider asking this question at the Wikipedia:Help desk. - This is where editors will try to answer any question regarding how to use Wikipedia. Just follow the link, select the relevant section, and ask away. You could always try searching Wikipedia for any help related to the topic you want to know more about. I hope this helps!
This page is for questions about the Articles for creation process. Please consider asking this question at the Wikipedia:Reference desk. They specialize in knowledge questions and will try to answer any question in the universe (except how to use Wikipedia, since that is what the Help Desk is for). Just follow the link, select the relevant section, and ask away. You could always try searching Wikipedia for an article related to the topic you want to know more about. I hope this helps.
This page is for questions about the Articles for creation process. If you would like to start writing a new article, please use the Article wizard. If you have an idea for a new article, but would like to request that someone else write it, please see: Wikipedia:Requested articles. I hope this helps.
Hello. It appears that your submission to Articles for Creation was declined because it lacked reliable sources. Please note that Wikipedia requires third-party, independent sources for an article to be considered notable enough for inclusion in the encyclopedia. If you need further help on what sources could be considered reliable, please visit the help desk. Thank you. ~~~~
If you review and reply to a question on the help desk, consider notifying the submitter on their talk page. You can use {{subst:AFCHD/u|Name of submission|Name of section in the help desk}} to accomplish this
Welcome to the Articles for creation help desk!
If you see two empty white boxes below, one small and one big, please complete the following, otherwise skip to step #2 below:
First, copy and paste the following code in the 'Subject/headline' bar:
{{subst:#time:H:i:s, j F Y}} review of submission by {{SUBST:REVISIONUSER}}
Then, copy and paste this code in the big input box below that:
{{Lafc|username={{SUBST:REVISIONUSER}}|ts={{SUBST:#time:H:i:s, j F Y}}|link=
<!-- [[Wikipedia talk:Articles for creation/Your submission name here]] OR [[Draft:Your submission name here]] -->
}}
Below that, explain why you are requesting assistance, take as many lines as you need and make sure to remember to sign your post with ~~~~
If you see a single big input box below, which starts with == {{SAFESUBST:#time:H:i:s, j F Y}} review of draft by {{SAFESUBST:REVISIONUSER}} ==, follow the instructions in the box.
Finally, scroll the whole page to the bottom and click Publish changes.
This step is required or your request will be lost.
If you are waiting for a reviewer to review your draft, please be patient, as drafts are reviewed in no specific order. The reviewing process currently has a 2+ months backlog, with 2,442 pending submissions waiting for review. Please do not ask for a reviewer to tell you the status of your draft or to review it faster.
Please discuss below the ways we could improve it for clarity. Primefac (talk) 09:53, 12 April 2023 (UTC)
@Primefac Is this covered in ths next section, with the scripts? David10244 (talk) 19:19, 13 April 2023 (UTC)
No, that section is for the actual "put stuff on the page" links. This is for the pink edit box that one sees when they edit AFCHD itself. Primefac (talk) 19:33, 13 April 2023 (UTC)
OK, thanks. David10244 (talk) 08:09, 14 April 2023 (UTC)
AFC HD Preload templates
There are currently four preload templates for the AFC HD. Hatted below are what the user is shown when they click on the link that preloads the text:
== {{SAFESUBST:#time:H:i:s, j F Y}} review of submission by {{SAFESUBST:REVISIONUSER}} ==
{{Lafc|username={{SAFESUBST:REVISIONUSER}}|ts={{SAFESUBST:#time:H:i:s, j F Y}}|page={{SUBST:Void|<!--
FIRST ENTER THE PAGENAME FOR THE DRAFT YOU HAVE QUESTIONS ABOUT ON THE LINE BELOW. It's good to omit the https://en.wikipedia.org/wiki/ part -->}}
}}{{SAFESUBST:Void|<!--Please enter the pagename for the draft in question on the line above.
THEN TELL US WHY YOU ARE REQUESTING ASSISTANCE BELOW THIS LINE. Take as many lines as you need. -->}}
~~~~{{SAFESUBST:Void|<!--
FINALLY, MAKE SURE TO CLICK THE "Publish changes" BUTTON BELOW OR YOUR REQUEST WILL BE LOST!-->}}
== Request on {{SAFESUBST:#time:H:i:s, j F Y}} for assistance on [[Wikipedia:Articles for creation|AfC]] submission by {{SAFESUBST:REVISIONUSER}} ==
{{anchor|{{SAFESUBST:#time:H:i:s, j F Y}} review of submission by {{SAFESUBST:REVISIONUSER}}}}
{{Lafc|username={{SAFESUBST:REVISIONUSER}}|ts={{SAFESUBST:#time:H:i:s, j F Y}}|declinedtalk= <draft name is preloaded from source link>}}
{{SAFESUBST:Void|
<!-- First, tell us why you are requesting assistance. Take as many lines as you need. --> }}
<!-- Start of message -->
<!-- End of message -->~~~~{{SAFESUBST:Void|<!--
Finally, make sure to click the "Publish changes" button below or your request will be lost!-->}}
== {{SAFESUBST:#time:H:i:s, j F Y}} review of submission by {{SAFESUBST:REVISIONUSER}} ==
{{Lafc|username={{SAFESUBST:REVISIONUSER}}|ts={{SAFESUBST:#time:H:i:s, j F Y}}|declined=<draft name is pre-loaded from the source link>}}{{SAFESUBST:Void|
Below this line, tell us why you are requesting a re-review. Take as many lines as you need.-->}}
~~~~
{{SAFESUBST:Void|<!--
When you have finished, click the "Publish changes" button or your request will not be posted!!-->}}
== {{SAFESUBST:#time:H:i:s, j F Y}} review of draft by {{SAFESUBST:REVISIONUSER}} ==
{{Lafc|username={{SAFESUBST:REVISIONUSER}}|ts={{SAFESUBST:#time:H:i:s, j F Y}}|draft=<draft name is preloaded by the source link>}}
{{SAFESUBST:Void|
FIRST TELL US WHY YOU ARE REQUESTING HELP ON THE LINE BELOW THIS LINE. Take as many lines as you need. -->}}
~~~~{{SAFESUBST:Void|<!--
FINALLY, MAKE SURE TO CLICK THE "Publish changes" BUTTON BELOW OR YOUR REQUEST WILL BE LOST!!!-->}}
Let's discuss how we want to codify the messages shown by the templates and/or make it as clear as possible that information needs to be added before the page is saved. Technical solutions are also possible, e.g. we could have it save nothing if no text is provided. Also, if anyone knows where Template:AfC decline/HD preload is actually used, please update the note above; I think it's entirely unused so we could probably just redirect it to one of the other variants if that's the case. Primefac (talk) 09:53, 12 April 2023 (UTC)
Another alternative is having a script (like the submission wizard) for making posts to the help desk. This has the added benefit of being able to provide more direct and simpler instructions, as well as preventing users from submitting blank posts. I'm happy to build a mock-up if anyone thinks this is a good idea. — Ingenuity (talk • contribs) 12:12, 12 April 2023 (UTC)
I think this could be a really good idea. The instructions above ask the user to copy paste complex wikicode, which is probably confusing and unintuitive to non-programmers. If we can hide this code by having Javascript take care of it, that'd be a great solution. –Novem Linguae (talk) 12:16, 12 April 2023 (UTC)
Looks great. I think we can remove the explanatory text under each textbox though. For #1, it will be obvious from the textbox's text if it is prefilled or not. For #2, we can hopefully write it into the JavaScript that the form will prevent submission and show a warning if the textbox is blank.
Have we yet identified all the different links/buttons that create a new section at the AFC help desk? Will it be easy to change all these to use a JavaScript helpdesk wizard? –Novem Linguae (talk) 13:19, 12 April 2023 (UTC)
Yes, we need to review if it's fit to handle being invoked from all those places (looks like the preload texts for different cases vary by name of the parameter used for passing page name in {{Lafc}}), and if it has any bugs. We can then link it from templates, much like WP:AFCSW. – SD0001 (talk) 15:36, 12 April 2023 (UTC)
I've fixed the bugs and made the name of {{Lafc}} param configurable in this edit request. Once that's done, we should be good to go. – SD0001 (talk) 16:57, 12 April 2023 (UTC)
Thanks @Ingenuity. Just so you know, ¶m= can be used in URL to control the param used in {{Lafc}} invocation when the user saves. The preload codes posted by Primefac above indicate different params like page, draft, declined or declinedtalk may need to be used, though I haven't checked if they actually make any difference in the rendering. – SD0001 (talk) 11:12, 23 April 2023 (UTC)
Sooo much better! We do need to make sure they add their signature, unless it does it automatically. S0091 (talk) 13:51, 12 April 2023 (UTC)
AfC Reject Options
There are two options under the "Reject" section on AfCH. The two options are: "Topic is not notable" and "Topic is contrary to the purpose of Wikipedia". I feel like this can be improved with more options, like "This article is clearly a joke or vandalism". UrbanVersis32KB ⚡ (talk / contribs) 23:10, 16 April 2023 (UTC)
Jokes and vandalism are contrary to the purpose of Wikipedia. I have seen other proposed reasons for rejection, but I don't remember them now. If you are saying that "Topic is contrary to the purpose of Wikipedia" is not clear, I agree. Another reason to reject a draft is that it is advertising. That is also contrary to the purpose of Wikipedia, but that may not be clear to some reviewers, and may not be clear to some editors. I think that a minor change might provide some clarity, and that is for the "contrary to the purpose", include a link to What Wikipedia Is Not. I agree that it may not be obvious that "contrary to the purpose" has a broad scope. Robert McClenon (talk) 02:01, 17 April 2023 (UTC)
Yes, I have had the problem of new users wondering what "contrary to the purpose of Wikipedia" means and why I rejected their drafts, and there should definitely be more options for new users who are confused why their drafts were rejected. UrbanVersis32KB ⚡ (talk / contribs) 22:13, 17 April 2023 (UTC)
Agree as well - what does the contrary to Wikipedia even include? Ridiculous; there should be some proper justifications available for it...! Mattdaviesfsic (talk) 04:53, 17 April 2023 (UTC)
I usually leave a comment specifying which one, such as "See WP:NOTPROMO". There are also decline messages available for jokes/hoaxes, vandalism in addition to NOT that I have used. S0091 (talk) 15:17, 17 April 2023 (UTC)
If it's a joke or vandalism wouldn't WP:G3 generally apply? Galobtter (talk) 04:20, 18 April 2023 (UTC)
User:S0091 states, correctly, that there are decline messages for jokes, hoaxes, vandalism, and NOT. Those are decline messages. User:Urban Versis 32 was referring to Reject messages. There are only two Reject messages, which is probably not enough. User:Galobtter says that G3 applies to vandalism, etc. Yes. That is a speedy deletion criterion. Sometimes the reviewer only wants to reject the draft and not to tag it for deletion, or wants to remove it from AFC submission while waiting for an admin to check the deletion request. Decline messages, reject messages, and speedy deletion codes are three ways to deal with drafts, and two combinations of them are permitted. I think that "contrary to the purpose" should be clarified or more options offered. Robert McClenon (talk) 05:58, 18 April 2023 (UTC)
Yeah I do agree there should be more rejection options. It was not obvious to me either until your comment above that "Topic is contrary to the purpose of Wikipedia" includes WP:NOTPROMO. Galobtter (talk) 06:01, 18 April 2023 (UTC)
Further agree for more options; they're badly needed! Mattdaviesfsic (talk) 06:23, 18 April 2023 (UTC)
I would encourage folk to read/skim the original discussion that led to the Reject option being formulated, which in general had the consensus that simpler is better, and to not have too many options for simplicity. If we need to update, reword, or otherwise update the links in the template, that's fine, but having just as many Reject reasons as Decline reasons is somewhat contrary to the point. Primefac (talk) 11:13, 18 April 2023 (UTC)
User:Galobtter, User:Mattdaviesfsic - Do you really mean that there need to be more options, or that they need to be better explained? I think that we have agreement that "contrary to the purpose" is not clear, and that it was and is intended to encompass a variety of issues, but do we really need more options, or to explain them better? Robert McClenon (talk) 14:48, 18 April 2023 (UTC)
I think that "contrary to the purpose" is meant to cover everything that is included in What Wikipedia Is Not, but that isn't clear to reviewers.
There hasn't been much discussion of "not notable", probably because it is clear, but I will add that I think it should be used for rejection in two cases. The first is persistent resubmission of drafts with no references or no reliable significant references. The second is persistent resubmission after an article has been deleted. "Not notable" should only be used for persistent resubmissions. Otherwise, decline and give the editor a chance. Robert McClenon (talk) 14:48, 18 April 2023 (UTC)
I would also only reject it for "Not notable" if it's obviously clear that the subject is not notable (for example an article on a YouTuber with very few videos and very little subscribers as those are unlikely to have any reliable sources on them, not that they are never notable as there can be circumstances in which someone with very little susbcribers is notable) ― Blaze WolfTalkBlaze Wolf#6545 16:38, 18 April 2023 (UTC)
I think we should have more options, since WP:NOT covers so many things. There's no reason not to have the reject list be a version of the decline list excluding reasons that aren't reject reasons. Even if reviewers understand what WP:NOT covers we need to better explain to submitters. Galobtter (talk) 17:07, 18 April 2023 (UTC)
I do agree. With WP:NOT as a reason it could mean a number of things, not a dictionary, not a newspaper, not a link hosting website, etc. ― Blaze WolfTalkBlaze Wolf#6545 ― Blaze WolfTalkBlaze Wolf#6545 17:10, 18 April 2023 (UTC)
I would argue that is why we should add comments to clarify the reason by pointing them to the relevant section(s). S0091 (talk) 19:19, 18 April 2023 (UTC)
I agree that the reviewer should use comments to explain clearly why it is rejected, but if a draft was repeatedly submitted, the previous decline comments and the repeated submissions will often be clear enough. I don't think that a draft should be rejected as "not notable" on the first submission, and usually not on the second. Robert McClenon (talk) 22:34, 18 April 2023 (UTC)
I am not actually sure we do need to better explain to submitters. Maybe I'm not seeing a large subcategory of rejects that are confusing, but the ones I have seen are usually handed a reject for the reasons Robert McClenon mentioned. And there's always the comment box for reviewers to use if they think more explanation is warranted. -- asilvering (talk) 17:01, 23 April 2023 (UTC)
I see your point, but compared to the wide swath of options for the "decline" section, the "reject" options could be a lot better. For example, if a user wrote an essay but didn't realize that Wikipedia is not a place for essays, the reviewers shouldn't have to put "Wikipedia is not a place for essays" every time. And what if someone wrote an article that is everything but an advertisement for a company (Not articles that are partially based on advertising, but have most of the article based on an advertisement and the article would require a complete rewrite to meet Wikipedia's guidelines)? Reviewers shouldn't have to type in every time, "this article is clearly an advertisement." UrbanVersis32KB ⚡ (talk / contribs) 21:24, 23 April 2023 (UTC)
For one, there is a decline for essays and we should not be rejecting if that is the only reason. For a reject, yes, we should use the comments especially if the draft has not been previously declined. Like @Robert McClenon stated above, its one thing if a draft has been declined several times before, thus a comment may not be necessary for a reject but it is an entirely different situation if we reject a draft that has not been previously declined and also does not meet CSD criteria. In instances where it does meet CSD criteria like what you describe (the article would require a complete rewrite to meet Wikipedia's guidelines), it is not uncommon for reviewers to reject without additional comments, then nominate for CSD but largely outside of those type of circumstances (exceptions always apply) reviewers should be providing a comment. S0091 (talk) 22:20, 23 April 2023 (UTC)
I will also note that, when I tag a draft for speedy deletion, I often comment on it, for two reasons. One is for any statement for the reviewing admin as to why it is a speedy delete, such as the use of the first person plural or second person, which are promotional and non-encyclopedic. The other reason is that the comments will be copied onto the submitter's talk page. Robert McClenon (talk) 01:46, 24 April 2023 (UTC)
New users would find it difficult to understand what actually caused their draft to be rejected, and the reviewers should have more options to make it clearer to the creators of the rejected drafts. Comments would work, but comments shouldn't be needed when we could easily add more options to streamline the process. UrbanVersis32KB ⚡ (talk / contribs) 02:04, 19 April 2023 (UTC)
Proper way to mark article as duplicate of an existing entry?
Hey there, I'm going through the AfC list for the Switzerland project and saw draft:Trogen (Switzerland), which is a (shorter) duplicate of the existing Trogen, Switzerland entry (I suspect both mostly translated from the German version).
The draft has been sitting around for a couple of months and simply adds to the backlog with no chance of ever being moved to mainspace. What would the most appropriate way to flag this and get it removed from the list of AfC awaiting review? Simply comment out the review box? Superboilles (talk) 05:37, 23 April 2023 (UTC)
@Superboilles decline the draft and select 'exists' in the pre-defined reasons dropdown. – robertsky (talk) 05:52, 23 April 2023 (UTC)
And yes, it seems that the draft author has copied from the draft to the article. – robertsky (talk) 05:53, 23 April 2023 (UTC)
Thanks. Looks like I need to be listed as participant first to use the script, made a request. Superboilles (talk) 07:22, 23 April 2023 (UTC)
Ah.. you might not meet the requirements yet. Will do the decline on my end. Looks like it has been acted on. – robertsky (talk) 08:16, 23 April 2023 (UTC)
Sorry, meant to say I'd declined it, but forgot. -- DoubleGrazing (talk) 08:49, 23 April 2023 (UTC)
No worries. :) – robertsky (talk) 17:51, 23 April 2023 (UTC)
If a draft appears to duplicate an existing article, then I suggest checking whether the draft contains any material that is not in the article. If the draft either is the same as the article or is a subset of the article, blank and redirect the draft to the article. If the draft contains material that is not in the article, tag the draft to be merged into the article. (Please do not nominate the draft for deletion at MFD; such nominations are common, and unnecessary, and waste the time of the regulars at MFD. But you probably knew that.) Robert McClenon (talk) 04:32, 24 April 2023 (UTC)
AFCH updates
Can we get some additional functionality added to AFCH? It would be helpful for the tool to call out potential issues during the Accept process. For instance, the article Cladocolea cupulata was recently accepted but with a short description of "I made an article on a mistletoe species." That's a totally unacceptable short description. Had the tool made the acceptor review and explicitly accept this, it may have gotten changed in the process. Perhaps put an "Accept?" checkbox next to each field that isn't a checkbox or dropdown, and all must be checked before the submission can be accepted? - UtherSRG(talk) 12:32, 21 April 2023 (UTC)
Another update I'd like is a title check. When the draft's title is dab'd, check to see if the non-dab title is available in the main article space and suggest to have the draft moved there instead of the current dab'd title. In addition, if the dab is required (because the non-dab'd title is occupied) have a small note below the New Title text box with potentially better dabs (lower case, fewer words, etc). Also, a note to suggest the reviewer update the non-dab article with a hatnote or update any disambiguation page would be great, too! - UtherSRG(talk) 12:50, 21 April 2023 (UTC)
Regarding the short description - the tool already checks the shortdesc and asks the reviewer for input - any problematic shortdescs are simply because the reviewer didn't do their due diligence. Honestly, the same goes for just about every aspect you're raising. Primefac (talk) 13:54, 21 April 2023 (UTC)
Indeed. So the question is, is it easier/better to poke on people individually when they don't do their due diligence, or is it easier/better to update the tool to give users a pause and more information to help them change their behaviors? I think when a tool exists that can help people do the right thing, they more often do the right thing, and when the tool can be updated to help them further in this, then this increases compliance. - UtherSRG(talk) 16:11, 21 April 2023 (UTC)
It's a numbers game. If this is the first time anyone has seen and/or noticed the issue (or it's otherwise a fairly recent thing) I would poke people. If it's something that's been going wrong for a year (or more) then a script change might be in order. No point in rebuilding the barn wall if all that is wrong is a loose nail. Primefac (talk) 17:22, 21 April 2023 (UTC)
I don't usually ask when it's the first I see something. But when I've had to fix more than a handful of recent accepts in the past week, I thought it was time to speak up. Would you like me to keep track of who and what and how often for some time and report back? - UtherSRG(talk) 18:11, 21 April 2023 (UTC)
You can if you want. Some of our AFC stalwarts will likely comment here in the next day or so; curious to see what their thoughts are. Primefac (talk) 18:15, 21 April 2023 (UTC)
Would be interesting to know whether a reviewer who makes that mistake (accepts a nonsensical shortdesc) gets the acceptance form more generally wrong (wrong or no projects, categories, non-optimal title, etc.), or whether they only ignore what's in the shortdesc box. That might give a hint as to what the best solution is.
Sometimes you do come across some pretty interesting ones. A recent one I saw said something like "all sources updated and corrections made". :) -- DoubleGrazing (talk) 08:13, 22 April 2023 (UTC)
So I've created a log of issues I find. I've started with the issues I found yesterday. It can be found at User:UtherSRG/Accepted_mod_log. And DG my friend... you have a minor entry there... Now, I'm hardly saying that I'm any better. I know I make mistakes as well, and I hope I'm always improving. So I offer this log with no intention of chastising but in hopes that we all improve, and maybe the tool can be improved to help us all improve. - UtherSRG(talk) 12:04, 22 April 2023 (UTC)
@UtherSRG: fair point, and thanks for making it. I for one try to make sure the draft I'm accepting is okay, but I don't do off-page stuff like dab other articles, create incoming links, etc. My thinking (alternative spelling for 'laziness') is that either they've already been done, or if not, then someone else will take care of them at some point. Will try to do better going forward. -- DoubleGrazing (talk) 12:22, 22 April 2023 (UTC)
Things like bolding the title and fixing the short description are not technically required to be done by AFC reviewers (it's not one of the AFC criteria), and can always be handled by gnomes. Although I agree it's good to do if one has time. –Novem Linguae (talk) 13:12, 22 April 2023 (UTC)
Understood. I've certainly conflated a few different kinds of issues here, and perhaps I should have lead with that from the start, but sometimes talking things out with you smart folks helps crystalize my thinking. Indeed, the minimal requirements for AFC are one thing, while doing what's best for the encyclopedia may not always be the same thing. Does it pass an AFD check vs does it actually look right vs does the new article's acceptance require other items to be updated. And hopefully as I document various issues we can say "Hey, you know, we should do better at ensuring an article also looks right when we accept it, or maybe we can have other tools to help us do this better, or maybe just make a note that the draft would pass the AFD test but I'll come back to it when I have time to make it look right before I accept it" kind of stuff. I think one of the best things we did to improve Wikipedia in general was to create the AFC process, and I'm super impressed with how hard the reviewers work at this. But we can still do better. - UtherSRG(talk) 13:52, 22 April 2023 (UTC)
I think adding a "I checked this" box next to each field in the accept dialog would be overkill. But I think adding a warning that a non-parenthetical title is available would be helpful. I created a GitHub ticket for that just now. –Novem Linguae (talk) 13:10, 22 April 2023 (UTC)
Thanks! We can revisit whether its overkill or not as time goes on. - UtherSRG(talk) 13:45, 22 April 2023 (UTC)
Oh, the non-parenthetical title warning would be very useful. I'm sure I must have forgotten to check one of those at least once. -- asilvering (talk) 16:54, 23 April 2023 (UTC)
I am in general inclined to think that the acceptance part of the tool should be as simple as possible. Adding the short description to the Accept screen sounds like a good idea. It can be captured from the draft and displayed for possible change. I don't fully understand the comments about disambiguated titles. Robert McClenon (talk) 04:37, 24 April 2023 (UTC)
Adding the short description to the Accept screen sounds like a good idea. It can be captured from the draft and displayed for possible change. I think AFCH already does both of these things. We reviewers just need to remember to double check the short description box, and fix it if it is pre-filled with something poor. –Novem Linguae (talk) 04:45, 24 April 2023 (UTC)
Indeed. My suggestion, Robert, was to add something more so that users will have a harder time missing looking at the text in the short description box in the tool. - UtherSRG(talk) 11:26, 24 April 2023 (UTC)
My current list of issues (gathered from my issues log) that potentially can be addressed by updates to the AFCH tool:
Short descriptions that are missing or not acceptable
Disambiguated title issues
Titles that should not be disambiguated
Titles that imply a need for a hatnote or other update to another (undisambiguated) article
Other title issues
Titles that have improper casing
Titles with inappropriate text (such as LLC, Ltd, Inc, etc.)
Various MOS issues within the body of the article
I think a "before you accept" phase of the tool might be helpful to poke the reviewer to check various MOS issues, or at least to remind them that after they've finished the acceptance they should go and clean up any MOS issues. However, any of the current items the tool already asks about should at least have an explainer on what is expected, and perhaps a link to the appropriate MOS (such as WP:SDESC and WP:NC).
While none of this is the primary purpose of an AFC review, I don't think we do ourselves many (or any!) favors by accepting articles that have these and other issues. It's certainly possible for a reviewer to simply leave a comment to the editor(s) to address the MOS issues they see; but another option is to update the tool to give a "pre-acceptance".... some indication that the content of the draft is acceptable for an article, but that there are one or more MOS issues that remain to be smoothed out before the draft is moved into main article space. This could additionally tag the article with a category so that a reviewer with a small amount of time could check these pre-accepted articles for style compliance (or update them accordingly) and quickly accept them. (Hell, I'd probably be one of those kinds of reviewers...) The draft doesn't have to be perfect to be accepted, but MOS cleanup and content review are two different mindsets, so I understand a reviewer not wanting to switch between them when diving into review territory. - UtherSRG(talk) 12:01, 24 April 2023 (UTC)
I'm really against the idea of formatting issues being used to keep drafts out of mainspace. Having a draft sit for months in AfC is discouraging enough. -- asilvering (talk) 21:57, 24 April 2023 (UTC)
I agree. Per WP:AFCPURPOSE formatting or the like should never be reason to decline a draft unless it is incomprehensible but those are the type of issues for which the "custom" decline exists and I have used "custom" for that reason. S0091 (talk) 22:07, 24 April 2023 (UTC)
I wasn't saying to not accept a draft that has MOS issues. However, the accepting reviewer needs to clean up those MOS when they accept the article. If the article has sat for months without the MOS issues being fixed while in draft, they aren't going to magically go away when the draft becomes an article. - UtherSRG(talk) 10:40, 25 April 2023 (UTC)
That's not true from my experience - I don't see many editors fixing general MOS issues in Drafts but most of the time when I accept an article it will be improved over the next few days. Although the amount of attention I find to be topic dependent. Saying that, I usually try to do basic fixups on any articles I accept. KylieTastic (talk) 12:46, 25 April 2023 (UTC)
I don't know how to read there are one or more MOS issues that remain to be smoothed out before the draft is moved into main article space in any way other than "drafts with formatting issues should not be moved to mainspace". And like KylieTastic, I frequently observe changes being made once drafts have been moved to article space. -- asilvering (talk) 18:09, 25 April 2023 (UTC)
Please read it as "either fix the MOS issues before accepting, immediately after, or tag the accepted article so that the issues can be easily pounced on by an eager MOS editor". - UtherSRG(talk) 19:31, 25 April 2023 (UTC)
Alright, I'm happy to read it that way. I will still decline to do that in most cases, I'm afraid, since I think gnomes who enjoy doing that sort of thing are generally pretty quick at it and in many cases get there before NPP does, saving the article from having to be tagged at all. Knowing this, I prefer not to tag in a lot of cases, since I think maintenance tags often come off as passive-aggressive, confusing, and/or frustrating for new editors, who will be able to learn from other editors' work on their articles anyway. (Plus, having someone else come in and make some changes gives a sense that wikipedia is actually collaborative, which is a nice feeling when you're starting out, especially if your article has been lingering in draft space for a while.) -- asilvering (talk) 04:22, 26 April 2023 (UTC)
It is my experience as well that MOS type issues are often fixed within a short period of time, at least somewhat. I have also tagged drafts for attention if I am not inclined to fix it up myself. S0091 (talk) 18:22, 25 April 2023 (UTC)
I was a remote (because interactions are all online with the professor) and distant (I didn't get to sit in the professor's class nor look at their teaching materials) observer to a recent class, via the Outreach Dashboard (not a US college, hence not via eduwiki), in which the professor was making use of Wikipedia as a teaching tool. The new articles that some students submitted are quite well written and would deserve at least a C or even a B grade in the article assessment, and I didn't feel like faulting them on making MOS errors. I had worked on most of them to fix these errors before/after accepting the drafts, but I cannot catch them all apparently, with another editor whom I known to be a better gnome than myself continuing to remove some more MOS errors from the now articles.
My take is that if the draft isn't too egregious in terms of the content, and has MOS issues, we should fix whatever that are apparent to us and push it out the mainspace and another experienced editor will come along and help to take care of the rest.
Also, since our review outcome/KPI is aiming at 51% keep outcome at AfD if the accepted drafts ever go to AfD, we should look at AfD deletion rationales as well. Seldom, if not none at all, I see that articles are being deleted because of formatting issues. – robertsky (talk) 04:46, 26 April 2023 (UTC)
Non-english decline sometimes won't input the language
Hello! I noticed that rarely when I Decline a draft as not being in English, when I add what language the draft is in, it seemingly will ignore that I added the draft's language and instead use the default non-english decline. So far this has only happened twice to me (once when I declined a draft written in Ukranian and again just recently when I declined a draft written in German). Given the seeming rarity of the issue I don't think its of major concern but I figured I'd bring it up here in case anyone knows about it. ― Blaze WolfTalkBlaze Wolf#6545 15:19, 26 April 2023 (UTC)
Hey there. Thanks for reporting this. The German bug has been reported before, and the ticket is here. Ukrainian probably isn't working because the way you spelled "Ukranian" is missing an "i". I tested it with "Ukrainian" (note the extra "i") and it worked. Hope this helps. Happy editing. –Novem Linguae (talk) 15:34, 26 April 2023 (UTC)
Oh, that might be why it didn't work. I never thought that "Ukrainian" was spelled like that. ― Blaze WolfTalkBlaze Wolf#6545 16:04, 26 April 2023 (UTC)
Submission wizard and WikiProject banners
I follow various pages pertaining to new content in a topic area and its associated WikiProject. It appears that editors making submissions have the ability to add WikiProject banners to the talk page of their submissions, often adding a half-dozen or more WP banners which have zero to do with the topic of their submission. Is this a good idea? The WP in question has been semi-active for years, meaning that there aren't people active in undoing these edits. I suspect this is also the case with many of the other WPs added through this mechanism. RadioKAOS / Talk to me, Billy / Transmissions 08:01, 27 April 2023 (UTC)
We've had this discussion on and off, both here and other venues, many times over the years, with no real resolution. There are some who feel that we should never place any banners for any project that is inactive, while others feel that placing such banners is important for if/when such projects become active again. My personal preference is to say "who cares" and let folks do what they feel is best (i.e. if you want to, do it, but if you don't, don't). Primefac (talk) 08:08, 27 April 2023 (UTC)
Since I've never used the submission wizard, it occurred to me that I should check it out after making the above comment. At the bottom, there's a drop-down menu with some explanatory text. In the cases I refer to above, it appears users are just picking random WPs from the top of the list instead of "the 1–4 most applicable WikiProjects". There's not even a wikilink to any page explaining what a WP is, so it's not at all surprising that this is happening. The text alone is probably not a sufficient enough explanation given the learning curve of its typical user. RadioKAOS / Talk to me, Billy / Transmissions 08:16, 27 April 2023 (UTC)
Oh... right, I misread your statement. So on the submit Wizard folks are choosing potentially unnecessary WikiProjects. I wonder if we could limit the script to only allow 4; at the very least it would cut down on the amount of unnecessary tags. Primefac (talk) 08:23, 27 April 2023 (UTC)
Women in Red May 2023
Women in Red May 2023, Vol 9, Iss 5, Nos 251, 252, 267, 268, 269, 270
--Lajmmoore (talk) 18:27, 27 April 2023 (UTC) via MassMessaging
Thoroughly unpleasant and inappropriate Wikpedia:Articles for creation threat
Hi, I notice that in the page [Wikipedia:Articles for creation] we have the following text:
Please note that getting a review can take several weeks, but that your draft will be reviewed eventually. Attempting to bypass the process by moving the page, or cutting and pasting it into a new mainspace article, may lead to the page being moved back into draftspace again, speedy-deleted or listed for AFD, and repeated attempts may even lead to you being temporarily or permanently blocked from editing Wikipedia due to disruption.
I note that this was added by Bearcat about one year ago. To my mind, the concept that an editor in good standing can be permanently blocked for developing an article in draft space and moving it to main space is completely outrageous; it is at odds with Wikipedia's philosophy, at odds with the whole point of draft space, at odds with the origins of AfC.
There are only three ways to create articles: (1) in main-space directly, which is almost never satisfactory, as it means articles must either be written in a single edit, or must linger in half-completed form; (2) in sand-boxes, which is unsatisfactory in a collaborative project, as it's very hard for other editors to find articles that have been begun in another's personal sandbox, leading to wasted effort, conflicting and overlapping articles on nearly-the-same subjects; (3) in draft-space, which was made for the purpose. But AfC manifestly struggles to keep up with its drastic workload, so it makes total sense for editors to move uncontroversial articles to main-space themselves.
If articles can only be moved from draft to main by AfC reviewers then at best AfC has a hideous work-load, at worst AfC has assumed sole responsibility for determining what goes into WP, abolishing the roles of the new page patrol, AfD, and community consensus on notability. And it has threatened to use administrative tools to assert its role.
I think that's wrong.
Elemimele (talk) 09:03, 23 April 2023 (UTC)
@Elemimele: TBF, it doesn't say you will be blocked for moving a draft into the main space. It says "repeated attemptsmay even lead you to being [...] blocked"; 'repeated', as in you did it more than once (see WP:TENDENTIOUS); 'may', as in may or may not; and 'even', implying that this is unlikely to be the first response.
It also isn't true that "articles can only be moved from draft to main by AfC reviewers". Pretty much anyone with the necessary user rights, who isn't subject to COI/paid restrictions, can publish directly without having to go through AfC. -- DoubleGrazing (talk) 09:14, 23 April 2023 (UTC)
@DoubleGrazing: It's difficult to interpret "Attempting to bypass the process by moving the page...may even lead to you being temporarily or permanently blocked" as anything other than "You are not allowed to move the page yourself". Repeated moves shouldn't really arise: if the article was moved into main-space but its creator, it should never be re-draftified (that's the beginnings of a move-war), it should go to AfD to seek community consensus (which, to be fair, is usually what new page patrollers and AfC reviewers do). Elemimele (talk) 09:46, 23 April 2023 (UTC)
I think I'd prefer to see a less bitey, more collaborative text, for example:
Please note that getting a review may take a few weeks, but that your draft will eventually be reviewed. Do be patient. Although it is permissible to move your draft into main article space yourself, we strongly recommend that you do not do so unless you are already very experienced in writing articles. There are many things that can go wrong with a new article, some quite subtle (Wikipedia is complex; it is easy to misunderstand how it works). All new articles in main space are checked by the New Page Patrol before being listed in search-engines. If you move your article prematurely, it will not be accepted, and is likely to be listed for deletion at WP:AFD or even speedily-deleted if the problems are serious. AfC reviewers are experienced editors who recognise these problems, and who will help you iron them out, saving much wasted effort and anguish.
I vastly prefer this to the current wording, though I would drop the second sentence. -- asilvering (talk) 16:50, 23 April 2023 (UTC)
I do too -- maybe this wording is a little better?
Please note that getting a review may take a few weeks or months, but that your draft will eventually be reviewed. Although it is permissible for autoconfirmed editors to move their drafts into main article space themselves, we strongly recommend that you do not do so unless you are already experienced in writing articles. There are many things that can go wrong with a new article, some quite subtle (Wikipedia is complex; it is easy to misunderstand how it works). All new articles in main space are checked by New Pages Patrol before being listed in search engines. If you move your article prematurely, it is likely to be listed for deletion at articles for deletion or even speedily-deleted if it has serious problems. AfC reviewers are experienced editors who recognise these problems, and who will help you iron them out.
I agree the current wording is WP:Bitey but I am also concerned about being too wordy. Maybe:
Please note that getting a review may take some time, but the draft will eventually be reviewed. Do be patient. Although it is permissible for autoconfirmed editors to move drafts into main article space, doing so prematurely could lead to the article being deleted. If the draft is not yet ready, a reviewer will provide guidance.
I like that wording. Galobtter (talk) 21:34, 24 April 2023 (UTC)
I like this too. The other new proposals are far too long. ~Kvng (talk) 21:53, 24 April 2023 (UTC)
I'm sure that is the intent of the phrasing. The expectation is that experienced editors using draftspace aren't using AfC. -- asilvering (talk) 16:47, 23 April 2023 (UTC)
I wouldn't mind eliminating the second sentence quoted in the original post, as that seems to violate the spirit of WP:DRAFTOBJECT. –Novem Linguae (talk) 11:18, 23 April 2023 (UTC)
DRAFTOBJECT regards the process of draftification, not draftspace broadly. Curbon7 (talk) 18:59, 23 April 2023 (UTC)
I prefer status quo; the warning is supposed to deter those who (often regardless of the warning) bypass AfC, which leads almost inevitably to a deletion, subsequent bad behavior by the author, and their eventual block. The warning is there for a reason and I have no patience for a newcomer crying about their feelings being hurt. Chris Troutman (talk) 18:58, 23 April 2023 (UTC)
Novem Linguae summed it up much better than my extended whinge. @Curbon7:, I'm not sure I understand the difference between the process and the broader use of the space. Does this mean that in developing articles in draftspace and transferring them myself to main space, which I thought DRAFTOBJECT allowed me to do, I am genuinely doing something wrong? @Chris troutman: my observation at the tea-house is that ferocious warnings aren't very effective: the sort of mentality that drives someone to write their promotional autobiography or fan-cruft here is the same mentality as excludes their taking any notice of the warning. But viewed from the perspective of expert writers in more specialist subjects (who already experience the longest delay-times), the warning portrays AfC not as a helpful tool to get their article into main space, but as a bureaucratic obstacle, backed up by ban-wielding policemen. It's just not a great look. Elemimele (talk) 12:16, 24 April 2023 (UTC)
Experienced article creators can use their sandbox and/or other userspaces to make their drafts outside of the AFC process, and then move them into main article space when they feel it is time. This is, in part, why DRAFTOBJECT exists; using the draftspace makes the draft somewhat more publicly workable for collaboration. That said, if an article has been draftified because it is lacking in some significant way, and an author then (repeatedly) objects by undraftifying it, now we have an issue that may warrant blocking. It's this kind of action that the warning is intended to put the authors on notice about, not the former. There is, of course, a lot of middle ground between the former and the latter. The current notice is too bitey, but some bite should remain. Perhaps it might be time to codify why a draft is in draft space, with different draftification reasons given different kinds of warnings; an experienced editor looking for collaboration to flesh out an article before it is ready for main article space is a different kind of draft than a first-timer using the AFC Wizard. - UtherSRG(talk) 12:42, 24 April 2023 (UTC)
+1 to that idea, COI editors should be given different messaging than a non-COI editor etc. Galobtter (talk) 21:36, 24 April 2023 (UTC)
Most editors do not declare COI/Paid until they are instructed to do so which is generally after they have submitted a draft so I don't think in most cases this will be helpful and not sure even technically possible but the technical folks can speak to that. S0091 (talk) 21:44, 24 April 2023 (UTC)
The WP:WIZARD already asks about COI editing; I think it just needs to do something more different with those editors. Galobtter (talk) 10:45, 25 April 2023 (UTC)
It does, but it currently just instructs you copy the COI disclosure to your User page which I am assuming does not connect to the specific draft they are creating. There are editors whose COI is only to specific subjects so not applicable to every draft they create so going by the User page is not sufficient. I think a different flow will be needed such as if they affirm they are COI/Paid via the Wizard, a template is automatically add the the draft (or the talk page) and a different template is presented. However, that will only catch those who use the Wizard so not sure putting the in the effort to change the flow may not be worth it. S0091 (talk) 18:53, 27 April 2023 (UTC)
I think this text and a related proposal Wikipedia_talk:Criteria_for_speedy_deletion#Proposal:_CSD_for_copypaste_articles_of_drafts_declined_by_AFC also about "bypassing" AfC are symptoms of the fact that half of AfC is dealing with COI editors for which the warning is very much appropriate and the other half is for new editors for which only a note is needed. There are definitely cases where an editor starts out not autoconfirmed but then becomes autoconfirmed and should definitely be allowed to publish their own article (as consistent with general wiki philosophy as mentioned). This text definitely needs amending. Galobtter (talk) 21:34, 24 April 2023 (UTC)
Part of the problem is that many of us newer editors don't actually know whether draft-space is something that belongs exclusively to the AfC project, created when AfC began, and therefore should be used only in conjunction with AfC, or whether draft-space goes beyond AfC as a general space for any editor to start a new article. I honestly still don't know which it is?? Elemimele (talk) 12:10, 27 April 2023 (UTC)
When users submit their completed drafts for AfC review, it could be a good idea for a prompt that asks the user for 2-4 of the strongest sources in the draft that establish notability. A feature of this ilk will be particularly helpful for drafts that are ref bombed. See here for previous discussion @ WP:Village pump (idea lab)–––GMH MELBOURNETALK 02:30, 25 April 2023 (UTC)
I am thinking it would be a good idea to include as an option in decline notices. For instance, at the end of decline notices for notability, something that prompts them to use the talk page to point out the references. --CNMall41 (talk) 03:21, 25 April 2023 (UTC)
Regarding CNMall41's comment - WP:THREE exists, and as much as I personally do not use it there are a few folk who regularly do in their decline comments. I am not sure whether adding it to a decline notice is appropriate, though, since it is just a user essay.
Regarding the initial query - this will not help. Honestly. I say this as someone who has been working the IRCHELP desk for almost a decade. Even chatting with folks live, we say "what are your best three sources" and nine times out of ten we don't get them. All it will do is bloat the process and cause more work for the reviewers. Basically, the ones who ref-bomb often have no idea what actually constitutes a good source, and thus figure quantity is better than quality. I'm not saying we shouldn't ask for good sources (so that we can help them out if they're wrong) but prompting it unnecessarily will likely waste everyone's time more than necessary. Primefac (talk) 08:50, 26 April 2023 (UTC)
The place where it might help, I think, is in Wikipedia:Requested_articles. I'd be much more inclined to have a go at writing an article if whoever suggested it gave me a foot-hold into the sources. But realistically I accept the take-up is probably going to be very low anyway. Elemimele (talk) 12:23, 27 April 2023 (UTC)
I've noticed a lot of drafts recently about future events that haven't happened yet and are at least a year away. Is there any specific policy regarding events like these? For example there's Draft:Hungary at the 2024 Summer Olympics, a draft about an event that hasn't happened so there are no statistics. I usually am inclined to decline these drafts as being WP:TOOSOON, but is that really the right rationale here? ― Blaze WolfTalkBlaze Wolf#6545 13:34, 28 April 2023 (UTC)
Alright sounds good. Just wanted to check here to make sure I wasn't missing something. ― Blaze WolfTalkBlaze Wolf#6545 14:48, 28 April 2023 (UTC)
Just to give a second opinion (from a WP:OLY perspective) - if there is no country-specific information in the draft, it should be declined until such time that there is. Until then the article should just be a redirect to the main competition article (if it exists). Primefac (talk) 17:20, 28 April 2023 (UTC)
List: Copyvios
Please see below the detected copyvios which have a 5015%+ probability from Earwigs. Updated every 12 hours at midday and midnight UK time (and sometimes manually run by me). Each run a random 200 drafts are checked + the existing detected copyvios (to see if they have been fixed)
Copyvios In Drafts(Beware Of Mirrors!) - Do You Use This Table? Please Let My Operator Know :)
I'm going to use this opportunity to point out something important that I see a lot both as an AFC editor and an RD1-patrolling admin - a high percentage does not automatically mean that a draft should be deleted, nor even declined as a copyvio. For example, Draft:Avijit Das is currently listed at 83.4%, but those quotes are from the references themselves. Once you remove those and the lists you get to a point where there is a bit of copyvio but it can be removed (see Special:Permalink/1152325731 for what is left after I temporarily remove the previous and the section of copyright).
In other words, sometimes with a high percentage you'll need to remove non-infringing text just to see if there is other (infringing) text to deal with. Primefac (talk) 15:19, 29 April 2023 (UTC)
Oh yea, it's not a exact science, it's just a prompt... there are other cases where stuff shows up as a high percentage but in fact, the source is Creative Commons licenced (in a suitable form). Earwigs doesn't know the difference, and hence neither does User:RichBot - RichT|C|E-Mail 15:20, 29 April 2023 (UTC)
Oh, totally, wasn't implying that this is a worthless list by any means. Just soapboxing ;-) Primefac (talk) 15:22, 29 April 2023 (UTC)
Thanks for this. Quick note: I think the Earwig tool has a limited number of allowed uses per day before we hit some kind of API limit and the tool turns off for everyone. You may want to check with The Earwig to make sure it's OK to run 200 automated queries per day.
My general strategy is anything over 15% is suspicious and I do further manual inspection. Under 15% when I've checked it has always been proper nouns or small quotes, and therefore non-infringing. Hope this helps. –Novem Linguae (talk) 15:23, 29 April 2023 (UTC)
I can adjust to 15%, but I figured 50 was a good bet...
Earwig already agreed, it's in my TP history... somewhere - RichT|C|E-Mail 15:27, 29 April 2023 (UTC)
Thanks! And feel free to put back to 50 if my 15 idea has too many false positives. –Novem Linguae (talk) 20:46, 29 April 2023 (UTC)
Oh, also a note, User:RichBot is NoBots compliant... so if a draft is showing as a copyvio and its really not, you can always force it to be ignored with {{bots|deny=RichBot}} - RichT|C|E-Mail 22:56, 29 April 2023 (UTC)
Newly published article to replace redirect
Hi, I've just moved the draft for Dawn Moore to mainspace and re-named it Dawn Moore (American philanthropist) because Dawn Moore already existed, and then noticed that the page Dawn Moore is actually a redirect to her husband. I've tried to fix it but now the page Dawn Moore redirects to Dawn Moore (American philanthropist) - but I think the correct fix is for there to be no redirect at all? Could someone please assist? TIA! MurielMary (talk) 12:09, 30 April 2023 (UTC)
I've done a round-robin page move, so the article is now located at Dawn Moore. — Ingenuity (talk • contribs) 12:22, 30 April 2023 (UTC)
MurielMary (and any other reviewers who don't yet know), for future reference you should use {{db-afc-move}} if there is a redirect in the way of an acceptance. If you accept the draft and then notice the redirect, the standard {{db-move}} will also work (and likely be faster than posting here). Primefac (talk) 14:32, 30 April 2023 (UTC)
Many thanks to both editors above for your help and advice, much appreciated! MurielMary (talk) 09:35, 1 May 2023 (UTC)
Draftifying during AfD
I don't know if this an AfC or AfD problem, but I'll start here. Yesterday, Draft:Fractured (Mini-series) was created, and in its first six hours draftified no fewer than three times, with the creating editor Tyamutz each time moving it back to the main space. This morning I took it to AfD, and immediately Tyamutz moved it back to drafts. (Which, BTW, should IMO be downright forbidden, but that's a separate point.) What should I do now – move it back to the main space and let the AfD run, or withdraw the AfD and treat it as an AfC draft as if none of this happened? -- DoubleGrazing (talk) 06:21, 2 May 2023 (UTC)
The author of the draft un-draftified it 3 times, then after all that draftified it? My my.
My understanding is that once the AFD tag goes on, the article should stay exactly where it is. There's even a bot that will edit war with people if the AFD tag gets removed. Personally I'd move it back. Looks like someone moved it back while I was writing this comment.
The closest thing I could find to a policy is at Wikipedia:Articles for deletion#Wikietiquette, which states While there is no prohibition against moving an article while an AfD or deletion review discussion is in progress, editors considering doing so should realize such a move can confuse the discussion greatly, can preempt a closing decision, can make the discussion difficult to track, and can lead to inconsistencies when using semi-automated closing scripts. Hope this helps. –Novem Linguae (talk) 08:06, 2 May 2023 (UTC)
Thanks. I think that "While there is no prohibition against moving an article while an AfD" bit is where the problem lies, and the policy should be changed to expressly prohibit that.
Editors who are keen to get their article published at all costs and ASAP, simply don't get that it's not actually in their interest to publish something that genuinely isn't ready for publication, and often this only dawns on them when deletion is requested or proposed.
I know this has been discussed here many times before, and I'm preaching to the converted, but I'm just venting... -- DoubleGrazing (talk) 08:38, 2 May 2023 (UTC)
Here's the most recent RfC on the subject. In the case of draftifications during AfD, you're well within your rights to just move it back to mainspace citing WP:DRAFTOBJECT. Extraordinary Writ (talk) 20:27, 2 May 2023 (UTC)
User:DoubleGrazing - It's an AFD problem. As Extraordinary Writ points out, I introduced an RFC about a year ago to prevent moving an article during AFD, and it appears that I should have written it more narrowly to prohibit draftifying it. I agree that an article should stay exactly where it is during the AFD, but some editors thought that there might be an emergency that required moving the article while the AFD was still running. Moving an article from article space to draft space during an AFD is a form of gaming the system by a stubborn would-be article creator. I would be in favor of explicitly prohibiting draftification during the AFD, but that is only my opinion, and I think that some of the participants in the previous RFC didn't understand. Robert McClenon (talk) 00:03, 3 May 2023 (UTC)
DoubleGrazing, why did you and Eejit43 both edit war the article back into Drafts after the creator moved it back out to mainspace the first time? The proper response at that point is to do an AfD, not to try and force the article into Drafts again. Also, I notice you not only moved it into Drafts, but you also added the AfC template, submitted it, and declined your own submission. Which seems inappropriate, bitey and generally uncivil in multiple ways. Not something someone involved in AfC should be doing in any fashion. SilverserenC 00:23, 3 May 2023 (UTC)
Whoops, apologies for that one move I made, I don't think the script warned me that it had been previously draftified- but I probably just missed it! ~ Eejit43 (talk) 01:02, 3 May 2023 (UTC)
@Silver seren: "why did [I] edit war?" seems a bit antagonistic way of putting it; I moved it back to drafts once (yes, admittedly for the second time). This is one of those situations where no matter what you do someone is going to take issue with it. Draftify, and someone complains you should have taken it to AfD. Take it to AfD, and someone accuses you of being bitey, saying you should have just sent it back to drafts. (Trust me, I've seen both.)
One could turn your question around, and ask what would have been gained by taking it straight away to AfD, if the creator was then going to draftify it anyway? And that's what my question was about; what to do when that happens during an active AfD?
And not that it's entirely the point here, but the reason why I submitted the draft was to encourage the creator to continue following the AfC process. And the reason why I then declined it was to show the reason why it wasn't ready to be published. But thank you for sharing your views on how I should, or should not, be behaving. -- DoubleGrazing (talk) 05:29, 3 May 2023 (UTC)
User pages in Subcategories
For some reason a large number of user pages are showing up in the subcats at PAFC. I've flagged this up at VPT, but if anyone knows of a resolution and/or a better place to report it to, let me know. -- DoubleGrazing (talk) 07:09, 3 May 2023 (UTC)
I am no longer seeing the incorrect items. The problem seems to have been corrected. Robert McClenon (talk) 18:53, 3 May 2023 (UTC)
Returning Member
After having been moved to the list of inactive reviewers, I applied for access to the AFC Helper Script and was approved. My name was subsequently removed from the list of inactive reviewers, but has not been added back to the list of active reviewers. Am I supposed to add it back myself, or was this a mistake? Thanks! Noahfgodard (talk) 21:01, 5 May 2023 (UTC)
Copy/paste fail... at least I got your name in the edit summary! Fixed. Primefac (talk) 21:06, 5 May 2023 (UTC)
Rewarding AfC participation regularly
Right, earlier last month (see here, lots of us made a discussion about the backlog and backlog drives, how to reduce it, how people work, etc. Out of that came a firm desire for more regular reviewing. With that in mind, I want to suggest a way of doing this which would a) help keep the backlog down potentially, and b) encourage new reviewers in their work. So, my idea(s) is/are:
Older reviews: An editor who has reviewed x articles which are over y weeks/months old would receive a barnstar (perhaps a bot to carry out the giving out of the barnstars?);
New reviewers: A new/probationary reviewer (and please, someone check the list at Wikipedia:WikiProject Articles for creation/Participants#Probationary members - it certainly needs updating!) would receive (for example) the Original Barnstar, after their first 10 reviews (with an encouragement to keep going, of course);
Regular reviewing: Like streaks, if an editor reviews a consistent number of articles per week, over the course of (say) a month, they would receive the AfC Barnstar or something.
Thoughts? Is this complete and utter rubbish, does it sound absolutely genius, or a potentially good idea with tweaking? Mattdaviesfsic (talk) 09:23, 1 May 2023 (UTC)
Just a minor point regarding #2, in that I do go through the list every few months; those that are still on that list from 2021 simply have not done enough reviewing to be recategorised but still do the occasional review and thus have not been removed. Primefac (talk) 09:47, 1 May 2023 (UTC)
Thanks for the clarification - I was wondering...! Mattdaviesfsic (talk) 10:10, 1 May 2023 (UTC)
I think some kind of "streaks" system would be a good way to reward consistent work without encouraging burnout. Certainly I tend to work in bursts (at one point I was the top reviewer for the month, last month I've reviewed, uh, exactly zero articles) and think my review quality is highest when I am working consistently (to keep guidelines fresh in my head) but not constantly, and I'd find recognition for that motivating. I'm imagining some kind of system with recognition for reviewing X articles a week for Y months and maybe also at least one article a day for Z weeks?
This is all to say that I like #3 a lot, and #2 sounds good though unlike #3 I'd want this to be done manually (I've seen some prolific new reviewers really not get what we're doing here, with all the best intentions, and having someone take at least a quick glance at their first 10 reviews seems like a good safety net here). Don't really have thoughts on #1, I think we usually Rusalkii (talk) 22:36, 2 May 2023 (UTC)
@Mattdaviesfsic Sorry for the late reply, but I definitely co-sign your idea, especially as an AfC beginner! : ) Oltrepier (talk) 19:58, 6 May 2023 (UTC)
It is not an uncommon situation for a reviewer to encounter a draft that has the same title as an existing article. I would like to review the different reasons why this can happen, and my advice as to what a reviewer should do:
The article and the draft are about different topics, such as different people with the same name. The draft should be disambiguated. When the draft is accepted, a hatnote should be added to the primary article.
The article and the draft are about different people with the same name, and there are also other articles on people with the same name. When the draft is accepted, a disambiguation page should be created. If there are already multiple articles and no disambiguation page, it might be a good idea to create the disambiguation page anyway, regardless of whether the draft is accepted.
The entry in article space is not an article but only a redirect, such as from an album title to the recording artist or from a song title to the album. If the draft is not ready for acceptance, then the redirect should be tagged {{R with possibilities}} if it is not already tagged.
The article and the draft are the same or almost the same, but have different authors. Check to see if the article is a copy-paste from the draft. If so, tag the article for history merge of the draft.
The article and the draft are the same or almost the same, and have the same author. Redirect the draft to the article.
The draft is a subset of the article, and does not have any content that is not in the article. Redirect the draft to the article.
The draft expands on the existing article, which may be a stub. The article is a subset of the draft. The draft should be tagged to be merged into the article, or the article should be tagged to be merged from the draft. Decline the draft with a message inviting the author of the draft to merge the draft into the article.
There is sometimes a special case of a draft that expands on a stub article. The draft has been developed as a class project to expand the article. This is sometimes the case with a stub entry for a biological species, and a more complete draft on the species. The problem, which I sometimes see, is that the instructor and the class are trying to use AFC to expand an existing stub article. That is a good-faith misunderstanding as to what AFC is for.
The draft and the article are by different authors, and have some overlapping content, but the draft and the article each also have their own content. The draft should be tagged to be merged into the article, or the article should be tagged to be merged from the draft. Decline the draft with a message inviting the author of the draft to merge the draft into the article.
That's a long list of different cases. I have seen all of them. Comments? Robert McClenon (talk) 17:04, 11 May 2023 (UTC)
Does #4 cover copypaste moves? (These drive me up the proverbial wall!) Is it just me, or are these getting more common recently (possibly because of the large backlog)? -- DoubleGrazing (talk) 17:19, 11 May 2023 (UTC)
Yes, #4 includes copy-paste moves by the same author. I agree that they are getting more common, because authors get tired of waiting for review, and copy the draft into article space. I redirect the draft to the article, but I usually decline it with {{twocopies}} before redirecting. That message is erased from the draft when the draft is redirected, but it also goes onto the author's talk page. I think that it may also be a good idea to tag the article with {{notability}} for the attention of the NPP reviewer. Robert McClenon (talk) 18:46, 11 May 2023 (UTC)
History-Merge Notice
On the subject of history merge, I will repeat a comment that I have made before. When a reviewer requests a history merge, the template on the surviving article advises putting a canned message on the talk page of the editor who did the copy-paste. I have said that the message is too wishy-washy. It implies that doing the copy-paste was all right, and that moving the article would have been even better. Doing the copy-paste was not all right. We want to discourage it. We don't want to be excessively bitey, because sometimes the copy-paster didn't know better, but we don't want to imply that it was all right. Robert McClenon (talk) 17:04, 11 May 2023 (UTC)
Article creation hypothesis
Hello
I'm Trizek, community relations specialist working with the Growth team.
The Growth team is exploring a project idea that aims to improve the experience of new editors by providing them with better guidance and structure in the article creation process. The hope being that by providing new editors with more structure around article creation, it will lead to newcomers creating fewer low-quality articles that create work for patrollers who check recent edits and mentors who review newcomers’ drafts.
In 2022, about 28% of newly registered users who completed the Welcome Survey indicated that they opened an account specifically to create a new article (all stats). These newcomers don't yet understand core Wikipedia principles and guidelines around notability, verifiability, conflict of interest, neutral point of view, etc. These newcomers need additional guidance or they end up frustrated and disappointed when their articles get deleted. Because they aren't receiving the proactive guidance they need, they end up creating additional work for content moderators (patrollers, admins, watchlisters…) who need to provide reactive guidance which is rarely well-received or well-understood.
We're committed to shaping the overall plan based on community feedback and needs, while adhering to the following requirements:
The feature will be Community configurable, enabling each community to customize it to meet their unique needs.
The feature will provide guidance and guardrails to help newcomers create higher-quality articles and improve their overall experience.
The feature will be designed to reduce the downstream workload for content moderators.
So, we would love to hear from you:
Do you think this project will help new page patrollers on English Wikipedia?
Do you have any suggestions for improving this idea?
Is there anything about this idea that you find concerning, or you want to ensure we avoid?
Or do you want the Growth team to consider a totally different idea? Keep in mind that the Moderator Tools team and two other teams are also working the shared “improve the experience of editors with extended rights” key result, so there will be other teams approaching this from a less new-editor centric perspective.
Thank you in advance for your replies. Trizek (WMF) (talk) 15:43, 16 May 2023 (UTC)
I think it would better to change the bullet points into a numbered list of participants. This would make it easier to see how many people are participating in the Articles for creation. Carpimapstalk to me! 13:03, 23 May 2023 (UTC)
Not done: That hypothetical number would not be accurate anyway, since administrators can use AFCH without being on the list, and soon NPRs will be able to as well. Primefac (talk) 13:17, 23 May 2023 (UTC)
Draft Question
If a draft is rejected, is that permanent, or can drafts be edited enough to where it is un-rejected? QuicoleJR (talk) 13:10, 21 May 2023 (UTC)
QuicoleJR Rejection is not permanent for all time. If something has fundamentally changed about a rejected draft, such as new sources that the reviewer did not consider that addresses the reason for rejection, a user should first appeal to the reviewer that rejected the draft directly. If that fails, or the reviewer is not available, the matter may be brought to WP:AFCHD. 331dot (talk) 13:18, 21 May 2023 (UTC)
Rejection usually means the draft should not be worked on further. It means a reviewer has determined that there is a fundamental, unsolvable problem with the draft such as notability or WP:NOT. Feel free to link the draft here for a second opinion, but in general, rejection is the end of the line. Hope that helps. –Novem Linguae (talk) 14:15, 21 May 2023 (UTC)
Here's the specifics. I'm not going to submit the draft now, since it's not mine, but it was rejected for being non-notable. It now has fourteen sources. QuicoleJR (talk) 16:30, 21 May 2023 (UTC)
Actually, I might as well link it, since it's ready for mainspace IMO and the author hasn't touched it in a month. It's this draft. QuicoleJR (talk) 16:32, 21 May 2023 (UTC)
@QuicoleJR: I agree it looks pretty good, I'll reverse the rejection and replace with the AFC pending template. It's up to the author, or you if you wish, to submit it. - RichT|C|E-Mail 17:37, 21 May 2023 (UTC)
I moved it to an article as it may be good enough for AFD survival. Graeme Bartlett (talk) 23:47, 23 May 2023 (UTC)
RfC: Rewards for AfC participation
I started this idea above with some degree of support, but wanted to actually develop it into a proposal. With the AfC backlog now at over 4 months (and 77 articles, as I write this, being left stale for over 4 months), I think it's high time this gained some proper traction. Consequently, my ideas are to reward the reviewing of older articles (such like those 77), rewarding new reviewers, and rewarding regular participation and reviewing. In more depth, this would look like:
Older reviews: An editor who has reviewed x articles which are over y weeks/months old would receive a barnstar (perhaps a bot to carry out the giving out of the barnstars?). Proposed number would be 5 articles which are over 3 months/12 weeks old, or 10 articles over 1 month/4 weeks old.
New reviewers: A new/probationary reviewer would receive (for example) the Original Barnstar, after their first reviews (with an encouragement to keep going, of course). Proposed number would be 5 reviews, regardless of the timescale.
Regular reviewing: Like streaks, I guess - if an editor reviews a consistent number of articles per week, over the course of (say) a month, they would receive the AfC Barnstar (for example). Proposed rate would be 60 per month (i.e., twice per day).
All discussion and comments welcome. I'm happy to answer questions about how this might work, etc. Mattdaviesfsic (talk)
With all due respect, I have disabled your RFC - this does not need a project-wide discussion, just enough consensus to implement. If you really, really feel strongly otherwise, you are welcome to restore the RFC tag. I will give thoughts on this when I have more time. Primefac (talk) 07:35, 18 May 2023 (UTC)
This is a nice idea for promoting editor retention and generally just showing appreciation for others. However, I don't think this process should be done by a bot; it would only be marginally be better than those impersonal "Thanks for your edits!" notification when a certain edit count milestone is met. I think having an actual person give barnstars would be more personal. I don't think a strict criteria is warranted; imo, barnstars should be from person to person with a meaningful appreciation for their work. A criteria would remove that message. Carpimaps (talk) 11:39, 19 May 2023 (UTC)
Thanks for your thoughts. I have no problems with your thoughts on bot use, because I am one of those people who like things done at a particular time! I wonder if a bot would be able to communicate the fact that X user has made Y reviews - I can't say I have the time to trawl through everyone's AfC logs! Mattdaviesfsic (talk) 15:15, 19 May 2023 (UTC)
One of the misaligned incentives in the AfC process is that when an article would probably survive a trip through AFD (=the long-standing goal) but doesn't "look pretty", then the reviewer has a dilemma: Do you decline an article on a notable subject, which is bad, or do you move it to the mainspace, and risk someone yelling at you?
It might be possible to use a reward system to slightly counter this, by recognizing people not for the speed with which they say "needs moar citations", but for the work needed to get articles into the mainspace. That is, only count work on articles that reach the mainspace.
I also like the idea of recognizing people who review the oldest pages. They're often the oldest ones because they require more effort.
As for the mechanism, a bot could create the list, and perhaps simply publishing the list is the key point. But if you wanted to see a different approach, then take a look at s:en:User:WhatamIdoing. That group used userbox-like awards to encourage participation. WhatamIdoing (talk) 20:59, 19 May 2023 (UTC)
I think we need a way to encourage reviewers, but I don't think giving out barnstars is the way to go. I think a "leaderboard" would help; for example, a review of a one-day-old submission would be worth 1 point, while 3+ month old drafts would be worth ~3, and it would be reset at the end of each month. Displaying the leaderboard on the AfC talk page, along with newsletters and potentially the userbox awards suggested by WhatamIdoing, would go a long way in increasing reviewer motivation. — Ingenuity (talk • contribs) 19:09, 20 May 2023 (UTC)
Thanks for your thoughts, nice ideas. Mattdaviesfsic (talk) 20:06, 20 May 2023 (UTC)
Ingenuity - any thoughts on the table I've put at User:Mattdaviesfsic/AfC awards? Too complicated to be updated weekly, for instance? I've grouped various time frames together as you'll see, whether 8 points is too much, I don't know. (It would certainly be an incentive though!)
Looks pretty good. I think that 8 points is too much; maybe it could be 1 for <7 days, 2 for 7 days to 1 month, 3 for 1-2 months, 4 for 2-3 months, and 5 for >3 months. I also don't think that we need re-reviews, since it's not a backlog drive. If other people like this idea, I can write the code for a bot to do this. — Ingenuity (talk • contribs) 21:28, 20 May 2023 (UTC)
I don't think that declining should get the same points as accepting or rejecting (=getting it out of AFC's queue). @Mattdaviesfsic, looking at your contributions on 29 April 2023, you were declining articles at a rate of about one per minute for a while. It's quick and easy, and it's sometimes necessary. But declining just kicks the can (backlog) down the road. It doesn't actually resolve anything. I'd suggest zero points for declining an article.
As for review times: we might want to incentivize getting articles through AFC quickly, in which case the points might have a U-shaped curve: extra points for prompt action, and extra points for tackling the difficult older ones, but few points for the usual average. For example: 2 points for the first week or two, 1 point for the next couple of months, and 4 points for old articles.
@Ingenuity, could you collect the basic statistics, without assigning points? A table that gives accept:decline:reject ratios, something like this:
Editor || New || Average || Old
Matt || 2:9:0 || 2:15:1 || 0:3:0
(for whatever time periods you want) would let us see how many we're talking about in each range, and how many of them resolve the article's status. WhatamIdoing (talk) 01:39, 21 May 2023 (UTC)
However, I agree with @WhatamIdoing on the fact that declining should not have the same value as accepting or rejecting: it's likely the most common action we perform while reviewing drafts (or at least, I often found myself doing so as of late), so giving it the same points as other contributions could risk to undermine the whole meaning of the rewarding system.
Having taken on board these comments, I've updated the system to be a bit less generous with the points, and not counting declines (but still counting rejects). I hope this looks a bit more incentivising now? With some sort of consensus, Ingenuity can get some code going and theoretically we could launch this for June... ¯\_(ツ)_/¯Mattdaviesfsic (talk) 08:42, 21 May 2023 (UTC)
I disagree with WhatamIdoing, declining does not just "kick the can down the road" and suggests/encourages an overuse of reject. Reject is WP:BITEY and was added just to terminate problem cases such as tedious resubmits and should be used sparingly. Many declines are also very terminal as followed with a speedy delete (G11 and G12 mostly). Historically, accepts run at about 20-25% of submissions - in the last month much worse 13% accept, 82% decline and 5% reject (not including deleted which are mostly declines). Many declines are not resubmitted so also an end, no can kicking. If a subject is notable but not yet shown then decline - they hopefully they improve, resubmit and we accept - without the decline it's just stagnation unless the reviewer improves then accepts. IMHO yes accepts should get more but decline/reject should be the same and I agree that more for the older categories is the correct motivation. Yes most declines happen day 0, but the majority of the old submits waiting are still not accepts (or rejects) and are important to get done, not de-incentivise. Ideally 0 day declines and speedy delete would also get more for removal of spam, copyvios, attack pages from the project, but that would need a bot with admin level to work out. KylieTastic (talk) 10:20, 21 May 2023 (UTC)
@KylieTastic, I don't think that's a full analysis. First, there shouldn't be any G12 declines; copyvios should be immediately deleted, not declined (and that ought to be recorded as a rejection, not as a decline).
Here are the scenarios I'd like to avoid:
Quick decline with no response:
Article submitted. G13 timer starts ticking.
Reviewer glances at it, says "needs moar citations" or "Wikipedia is not for advertising" and declines. G13 timer resets.
Nothing happens.
Article deleted as G13.
Decline for hopeless article:
Article submitted.
Reviewer glances at it, sees that it's a non-notable subject (e.g., an autobiography from a teenager), says "needs moar citations" or "Wikipedia is not for advertising", and declines.
Editor changes something and re-submits.
Reviewer glances at it, clicks the same button in the AFCH script as the previous round, and declines again.
Repeat steps 3 and 4 until someone gives up.
Decline for notable subject:
Article submitted.
Reviewer glances at it, knows that it's notable, is convinced that it would pass AFD, ...and thinks about how unpleasant it is to have editors yell at you for permitting "sub-standard" or "embarrassing" articles, so instead of accepting it, says "needs moar citations" or "Wikipedia is not for advertising" and declines.
IMO we need to put our thumb on the scale to say that the first isn't helpful (it just delays the inevitable), the second should be rejected, and the third should be accepted.
What I'd like to see more of is:
Article submitted.
Reviewer spends 10 minutes looking it over in detail, checks a couple of the sources, determines that the subject is notable, is convinced that it would pass AFD ...and moves it to the mainspace, decorated with as many maintenance templates as seem appropriate, because the lousy article about a notable subject is statistically more likely to get improved if it's in the mainspace than if it's left in the draftspace.
Obviously, that can't happen for a lot of drafts, but IMO it could happen for more of them. WhatamIdoing (talk) 22:16, 21 May 2023 (UTC)
First, there shouldn't be any G12 declines - yes, there absolutely should be cv declines. Users need to know why their draft was declined, and it is not always the case where a G12-tagged draft is deleted as a copyvio, so tagging without declining just means the draft has to get re-reviewed anyway. Primefac (talk) 08:11, 22 May 2023 (UTC)
If you decline it as a copyvio, and it doesn't get deleted because it's not a copyvio, then it has to be re-reviewed anyway.
But more generally, I think that if the reason (should it prove to be valid) results in mandatory deletion, that should be considered instances of rejection, not merely declining the current version. Users can be told why their draft was rejected. WhatamIdoing (talk) 22:47, 22 May 2023 (UTC)
Thanks for your thoughts everyone (and KylieTastic) - with response to the thoughts on declines, I've updated the table at User:Mattdaviesfsic/AfC awards once again. Now the declines are worth half of accepts and rejects, and the maths is made easier (?) by adding together the outcome total and the time-based total. Is this too difficult, or is this much better? Please let me know! Mattdaviesfsic (talk) 10:55, 21 May 2023 (UTC)
Now the declines are worth half of accepts and rejects. I think incentivizing rejects over declines is not a good idea. Most non-accepts should be declines. For example I almost never reject. In fact, I think incentivizing accepts is also not a good idea. Accept, decline, and reject should have equal weight in any awards system, so that reviewers give a true and objective assessment, and are not nudged towards one or the other to get awards or points. Hope that makes sense. –Novem Linguae (talk) 11:16, 21 May 2023 (UTC)
Given it makes things more complicated to work out, and on the basis of good argument here, I have removed the different points for the outcome to focus wholly on the time of which the article has been waiting.
Any further thoughts again welcome. Mattdaviesfsic (talk) 11:54, 21 May 2023 (UTC)
@Novem Linguae Good point, too: I didn't think about that aspect. Oltrepier (talk) 13:32, 21 May 2023 (UTC)
I think that accepting is more work, and therefore deserves more points.
I also think we have existing incentives to avoid acceptance (if someone [anyone, not just someone who knows what they're talking about] things you are too generous, they will yell at you for permitting the pollution of the mainspace with notable subjects whose current version is "embarrassing"), and that therefore we need something to counterbalance those incentives. WhatamIdoing (talk) 22:01, 21 May 2023 (UTC)
I would argue against your first point; as I am going through my mental "checklist", whether the draft ticks the final "box" or not I have still done all the prior work. Filling in some additional fields for mainspace-moving is about as much effort as writing a thoughtful decline rationale (sometimes less). Putting in extra incentives to get a draft accepted will only skew the numbers towards acceptance as those who want to be at the top of the list will accept more than they should. Primefac (talk) 08:08, 22 May 2023 (UTC)
Do you really do the whole checklist each time? Let's say that you can tell at a glance that the page is a BLP that cites no sources. Do you then proceed through the full multi-point checklist anyway? Or just decline with "needs sources" and move on? WhatamIdoing (talk) 22:49, 22 May 2023 (UTC)
You missed my point. I was saying that if I am going to accept a draft, I have already done just as much work as if it was a narrow decline where I was "not checking the final box". Obviously, there are drafts that are quick-failed, but I tend to work at the back of the queue where it's not as easy to do that. As implied elsewhere by others, I think every one of my reviews takes at least 10-15 minutes. Primefac (talk) 07:28, 23 May 2023 (UTC)
My point is that if someone wanted to get a lot of points, they could go through the entire category and click decline them all with a request for more sources or because it feels promotional at a rate of a several drafts per minute. Editors won't do that for accepting articles, but they could do that for declining them. WhatamIdoing (talk) 14:57, 23 May 2023 (UTC)
WhatamIdoing - the premise is that editors will see it as fun, rather than a way to 'beat the system'. This situation could happen - I'm not doubting that (and editors like myself will admit to straight-out declines even without a backlog drive) - but (1) we must remind ourselves that the WP:ONUS is on the submitter to find sources, and (2) we're working on an assume good faith principle that editors will take as much time and care as usual over each draft they review. Mattdaviesfsic (talk) 15:12, 23 May 2023 (UTC)
I agree, @Mattdaviesfsic. Also, generally if a reviewer is mass declining without valid reasoning, we get questions at the help desk. We actually had something similar happen a couple months ago. Concerns about the reviewer were brought here and they were removed. We have also had instances where a reviewer is blocked for UPE so samplings of re-reviews were conducted. These are things that do not happen often though, at least with my AfC experience thus far. S0091 (talk) 15:30, 23 May 2023 (UTC)
Providing a "valid reason" for declining is not incompatible with quickly declining a large number of articles.
Here's what I'm seeing:
When I look through his contributions, Matt actually did decline a somewhat large articles at a rate of about one per minute. I believe that every one of those declines was deserved. My point isn't whether he was correct; my point is that it's a quick and easy activity.
But in this discussion, reviewers are claiming that they frequently spend a really long time looking at articles before declining them.
To put it another way:
In 20 minutes, you could decline 10 articles (90% quick declines and one that takes careful review first), or you could accept 2 (10 minutes each). But this project proposes to treat them all the same. WhatamIdoing (talk) 17:27, 23 May 2023 (UTC)
Third possibility - you spend 20 minutes reviewing and decline all of two articles. Which happens to me fairly regularly. The point of this venture is not to make it a competition, it is to provide some measure of feedback and thanks for those who are doing a lot of work. If people are gaming the system (hurray for cheating to win "internet points"?), we will deal with it in the ways described in this subthread. Primefac (talk) 07:58, 24 May 2023 (UTC)
Or you can spend a couple minutes and accept a couple articles. I am not sure why @WhatamIdoing thinks accepts always take longer. They don't. Just now I accepted Scream (2022 soundtrack) which took me all of a minute to review and accept. Its is difficult to tell this by looking at someone's logs because most are declines so you are generally not going to see a series of back-to-back accepts like you do declines. That is not to say there are not super quick declines; there are and those are most often in the very front of the feed but then starting even a day or so back, it begins to average out about the same in my experience. S0091 (talk) 14:40, 24 May 2023 (UTC)
Actually, take a look at @Bkissin (log) as they have a high acceptance rate with back-to-accepts and back-to-back declines so might make a good case study. S0091 (talk) 14:57, 24 May 2023 (UTC)
Back in the day if I was really motivated (and had time) I'd do 5-10 reviews all at once, then bomb through and decline/accept them all in succession (rather than do one at a time) so I can see how that's possible. Primefac (talk) 19:11, 24 May 2023 (UTC)
I strongly disagree that declining is less time consuming than accepting. Either outcome takes time to review. Some may be easy declines or accepts so only take a minute while others can take few minutes. I have certainly spent 20-30 minutes reviewing a draft I ended up declining. In addition to what Primefac notes, declines also lead to getting questions from the submitter which require thoughtful answers. I don't think I ever have gotten a question from a submitter whose draft I accepted. S0091 (talk) 13:27, 22 May 2023 (UTC)
I agree, let's not incentivize rejects. It looks like we can get consensus for equal weighting for any disposition. Also higher weighting for reviewing older stuff. Looks like that's where it has landed. ~Kvng (talk) 22:10, 23 May 2023 (UTC)
I've also created a userbox to use (e.g., for June), as seen on the linked page above. I don't do a lot of stuff with userboxes, so more than happy for someone to design something newer! Mattdaviesfsic (talk) 12:06, 21 May 2023 (UTC)
@Mattdaviesfsic It looks good to me, thank you for creating it!
Actually, you can forget what I previously wrote about deleting: @KylieTastic and @Novem Linguae did bring good points about the need to balance this new system properly. Oltrepier (talk) 13:46, 21 May 2023 (UTC)
Ingenuity - would you be able to write the code for your bot to update a leaderboard like the one at User:Mattdaviesfsic/AfC awards? (Obviously rejects would need to be counted before deletion.) I could write a message to send to AfC'ers (and possibly NPP'ers, depending on when their permissions are added here) and we could potentially get started on one for June? I would suggest a leaderboard be hosted at Wikipedia:WikiProject Articles for creation/Leaderboards/2023/June and then so on monthly and annually, as it would be easier to keep track of each month. I could then see who to award userboxes (etc) to at then end of each month. Could the bot update the leaderboard, say, every 2 or 3 hours, do you think? I've no experience with bot running so not my area of expertise! Mattdaviesfsic (talk) 16:08, 23 May 2023 (UTC)
I'll try to write the code sometime in the next few days. I can use a lot of the code from the stuff I wrote for the backlog drive, but it'll need some changes to do this. — Ingenuity (talk • contribs) 17:06, 23 May 2023 (UTC)
I have a question, maybe for User:Primefac, about whether a history merge is required. I reviewed Draft:Sajjad Jani about a week ago. A stub had been created at Sajjad Jani two months after the draft was created. I tagged the draft to be merged into the article. I considered whether to tag the article for a history merge as a copy-paste from the draft, but concluded that the wording of the article was different from the wording of the draft, so that it was not a copy-paste, although it looked like an "almost copy-paste". The originator of the draft, User:HMGelani, has asked for another look. It appears to me that the originator of the stub just barely managed to get credit for the stub, and that HMGelani should merge their additional content into the article. Comments? Robert McClenon (talk) 16:26, 24 May 2023 (UTC)
IMO, the published article would probably not be accepted through the AfC process. The refs on the published article can be added to the draft (as external links, say), the published article deleted, and the authors of both the draft and the published article can work on the draft.
Not a nice view I know, but purely my thoughts. Mattdaviesfsic (talk) 18:45, 24 May 2023 (UTC)
@Robert McClenon: Looking at this page comparison, the amount of verbatim text that is reused suggests is was copy/pasted and then expanded and slightly copy-edited before publishing to main space. Seems like something a history merge would be used for (notability questions aside). -2pou (talk) 19:17, 24 May 2023 (UTC)
Yes, it does appear to be a copy/paste. I have histmerged. Primefac (talk) 19:22, 24 May 2023 (UTC)
I stumbled across this draft while reviewing today. I fairly quickly rejected it, because a quick background search pulls up 0 google results for "Skill Broughtner" in quotes and the mentioned twitter account the article is focused on has 2 followers, but it also looks like it was written by AI. I can't tell if this really qualifies under any CSD criteria; can this be nominated under anything, or should it just be left there until G13?
(I vaguely remember rejecting a draft on first submission was discouraged at some point as well, but this falls under the criteria at WP:AFC/RI, so I figured it was fine; is it not fine, and/or is there anything else I should be aware of when rejecting submissions?) Skarmory(talk •contribs) 03:24, 25 May 2023 (UTC)
I went ahead and tagged it {{AI-generated}}. If you want to CSD it, G3 blatant hoax is usually the one that applies to AI. But letting it sit until G13 also sounds reasonable. Hope this helps. –Novem Linguae (talk) 05:34, 25 May 2023 (UTC)
This isn't a hoax, necessarily, since the twitter account mentioned does exist and there are no falsified sources (can't have falsified sources if your article is unsourced!); I don't know how admins are treating G3 when it's exclusively because it looks like it was written by AI. I think it's best to just let it G13 because there's no real harm to doing that, and the {{AI-generated}} tag will let anyone more involved with the LLM stuff find it if it needs to be deleted sooner for any reason. Thanks! Skarmory(talk •contribs) 07:57, 25 May 2023 (UTC)
Yet another Articles for creation helper script
I propose we rename "Yet another Articles for creation helper script" to "Articles for creation helper script". The "Yet another" part is a bit verbose, and in practice I don't think many people call it that. Any objections? cc Enterprisey –Novem Linguae (talk) 20:03, 25 May 2023 (UTC)
I like the quirky nostalgia of it, but I am not opposed if there are boring people who are fun haters that want it to more accurately reflect its current status as the One True ScriptTM. Primefac (talk) 20:19, 25 May 2023 (UTC)
Hahahaha. Sorry, that tickled me...not a comment about the suggestion although I do appreciate the "Yet another...". S0091 (talk) 20:25, 25 May 2023 (UTC)
(I miss the old scripts, broken in all their glory...) Jonatan Svensson Glad (talk) 22:17, 25 May 2023 (UTC)
There has been a question raised as to whether we should have the "Automatically open the review panel on AfC submissions" option enabled by default (i.e. you have to opt-out of having it show up) or keep it as the current status where it is in the "More" dropdown by default (opt-in to auto-open). Personally, since we have just had a script update that allows NPR to use AFCH (assuming they have the gadget turned on) I would think that not having it automatically display would be a better option (since it's just one more thing to pop up if one is randomly viewing drafts). Thoughts and comments appreciated. Primefac (talk) 10:36, 26 May 2023 (UTC)
I don't really care if it's on by default, but personally I would turn it off again and that should still be an option. One issue could be I know i missed the "preferences" option and I know others have so maybe that needs a change - however I don't know how as once I noticed it seemed obvious! Also if on by default maybe make it a tad smaller vertically it seems unnecessarily big. KylieTastic (talk) 11:10, 26 May 2023 (UTC)
Hopefully, the change can be made in such a way that if a reviewer has ever enabled or disabled that preference, then they should see no difference from before. It should affect only reviewers who have recently enabled the gadget. -MPGuy2824 (talk) 11:22, 26 May 2023 (UTC)
I think the point is more that one shouldn't need to turn it off if they are not expecting it. But, that's why we discuss these sorts of things - if folks think it would be more beneficial to have it always on by default, then that's what we'll do. Primefac (talk) 13:20, 26 May 2023 (UTC)
Wait, there's a way to make AFCH display by default? How do I enable that? – bradv 13:23, 26 May 2023 (UTC)
It's in the preferences (see the image to the right).
Thanks! I was looking for a little gear icon in the corner or something, and missed the link right in the middle of the page. – bradv 13:45, 26 May 2023 (UTC)
Again, another example of how the how the preference hides in plain sight! Maybe it's because the primary green and red with huge font just demand attention but it's still a very odd mental trick. KylieTastic (talk) 14:07, 26 May 2023 (UTC)
Another additional option is if the user has not saved any preferences (i.e. no "userjs-afch-preferences" exists in the account data) the preference form is shown. Ideally it would be tweaked to say "This is your first use of AFCH tool please check your preferences" and the "Cancel" changes to "Accept defaults" and saved, so it only ever happens the first time. KylieTastic (talk) 14:29, 26 May 2023 (UTC)
I like this option. Have it display by default when viewing a draft, but make the preference form pop up at first visit so people like me can't miss it. – bradv 20:02, 26 May 2023 (UTC)
Why are articles for creation categories so strange?
I've notices they are quite strange and arbitrary. For example we have Politics of the United Kingdom and History of Poland but no other individual country politics or history tags. This makes it hard to add the appropriate projects to a lot of AFC submissions. Can we fix them to make every wikiproject available or similar? Immanuelle ❤️💚💙 (talk to the cutest Wikipedian) 21:45, 26 May 2023 (UTC)
I'm pretty sure you don't mean categories, but WikiProject templates that are to be added to the new article's talk page. In that case, these would be restricted to WikiProjects that have separate templates; many don't. Many (most?) wikiprojects are added via their parent project's template params. -MPGuy2824 (talk) 02:07, 27 May 2023 (UTC)
Women in Red - June 2023
Women in Red June 2023, Vol 9, Iss 6, Nos 251, 252, 271, 272, 273
The Wikipedia Library: #1Lib1Ref - May 15th to June 5th
Tip of the month:
Looking for new red links? Keep an eye out for interesting and notable friends, family, or associates of your last article subject, and re-examine group photos for other women who may still need an article.
--Lajmmoore (talk) 09:14, 28 May 2023 (UTC) via MassMessaging
Moving drafts into article namespace without reviewer approval
As I've been thinking, The Nervous Set (which I created as a draft) was moved to mainspace several years ago by non-reviewer Moxy soon after Curb Safe Charmer rejected one of draft revisions. I didn't feel like bringing it up at the time because I created the draft and I just wanted to feel pleased to see the draft being in the mainspace. Two more examples are Policy Man, moved into mainspace by PigeonChickenFish after S0091's rejection, and Eleazar (painter), moved into mainspace by Netherzone without AFC review.
No offense to those who moved those drafts, but I just have concerns about such un-reviewed moves. I know that AFC has backlog issues, but are they excuses for those users to move them without awaiting an AFC reviewer? I mean, I wonder whether such practice is okay. I honestly don't have much confidence with using my notability skills for articles that I created, so I have had to use the AFC process, despite its agonizingly slow process. George Ho (talk) 05:47, 30 May 2023 (UTC)
@George Ho, You are welcome to move Eleazar (painter) back to draft space. I was helping out with a project to improve and save some drafts from a deceased Wikipedian.[4] I didn't realize it should go thru AfC since I did not create the original draft. Sorry if I made a mistake in doing so. Netherzone (talk) 06:31, 30 May 2023 (UTC)
Unless there is an editing restriction in place, it is not mandatory for a draft to go through the AFC process. Just because a page declined (or even rejected) does not mean that it cannot be moved by someone else to the article space; clearly the 2017 move of The Nervous Set was acceptable, since it was never subsequently nominated for deletion (by any means). Moving a draft back to the draft space purely because it was not accepted through AFC is disruptive more than anything. Primefac (talk) 06:57, 30 May 2023 (UTC)
@George Ho I moved the Policy Man article after clean up effort and the addition of new RS sources, as per the AFC rejection. Understanding WP notability is a skill set for all of us here (not just people moving articles). I am sorry if this article move offended you, or someone else, as it wasn’t personal. Would you like me to do something specific? PigeonChickenFish (talk) 07:24, 30 May 2023 (UTC)
From what I can tell from your edit summaries those moves back to draft space were because you felt the pages were not acceptable. Primefac (talk) 09:17, 30 May 2023 (UTC)
@Netherzone and PigeonChickenFish: No need. I'll leave the articles as-is for now. I just wanna bring one matter up and used those articles as just examples rather than an attempt to move back. George Ho (talk) 08:48, 30 May 2023 (UTC)
@George Ho a couple other things to keep in mind is articles are still reviewed by NPP unless the editor is autopatrolled so if deemed inappropriate for mainspace NPP will handle it (or the community at large). If the editor is autopatrolled like @PigeonChickenFish that generally means they are experienced with creating unproblematic articles. Also, AfC reviewers do make mistakes or may not have all the relevant information as usually determinations are made based on what is presented, not what is available. In my view, using Policy Man as an example, PigeonChickenFish rectified my error/uninformed decision. S0091 (talk) 15:30, 30 May 2023 (UTC)
@S0091 Thank you for being understanding. I want to note that the previous non-notable decision during AfC made sense at the time of the edits, especially given the state of the article and the provided sources. I only deemed it passing notability after adding new RS sources, removed non-RS sources, and general article clean up. I am not an AFC reviewer, which is why I specifically chose not to use that system to accept the article. But like you said, I am autopatrolled. I wasn’t trying to step on any toes, merely working alongside others to help move articles to the next level whenever possible, which in this case was to the article space. If someone disagrees with notability after, typically they would bring it up on the talk page of the article, or bring it to AfD? PigeonChickenFish (talk) 20:51, 30 May 2023 (UTC)
@George Ho, in the case of Eleazar (painter), there was a discussion between several editors what to do about DGG's drafts in user space as you can see HERE after he passed away. I was pinged by admin Star Mississippi on 11 April 2023 to see if I could help with these, since we had helped to clean up another deceased editor User:Possibly's drafts before his drafts expired, which I did (help out with).
If you look at the Eleazar (painter) article history, you will see that it was originally created as an unacceptable autobiography by editor User:Eleazar1954 in 2009. There was back and forth re: the AfD tag and DGG userified it to his own user space. I didn't know the draft existed until I was pinged, and I offered to help with the draft since I am quite familiar with notability criteria for visual artists. I then made improvements to bring it up to WP standards. According to article history, you tagged the draft as a user space draft on 13 May 2023, then User:CNMall41 moved it from User:DGG/Eleazar (painter) to Draft:Eleazar (painter) where I continued to work on it. On 30 May 2023 I moved it to article space (knowing that several other editors and admins were aware of the draft and it's longer history).
I did this in good faith to help out with David's drafts and did not think I did anything inappropriate by moving it to article space; moving it didn't seem like a "big deal" to me. I don't believe David (DGG) would have userfied the draft in the first place if he had not seen its potential, given his vast experience. I did not mean to step on your toes, and if I did, again, I apologize for that. Netherzone (talk) 15:57, 30 May 2023 (UTC)
Oh, no. You didn't step in my toes, honestly. Well... we've lost our beloved admin, no doubt. George Ho (talk) 18:45, 30 May 2023 (UTC)
Thanks for the ping @Netherzone. I'm just back from a few weeks offline and haven't dug into this. @George Ho, what Netherzone recounts is what happened as far as Eleazar and from my experience with their edits, I see no need for them to use AfC unless they feel they'd like another set of eyes for a specific reason. It may be out of process but when I draftify a deleted article for an established editor and the script adds AfC headers I explicitly say to ignore AfC and restore it if/when they're ready/have resolved G4 or other issues. Happy to return to this when I've dug out. StarMississippi 03:08, 31 May 2023 (UTC)
Contested Draftification
There is a situation in which one should look carefully at an article that was moved from draft space into article space other than by a reviewer. That is if the article was previously draftified. In that case, a previous editor, probably an NPP reviewer, thought that it was not ready for article space. If so, it is a good idea to check whether the problem that the previous reviewer identified has been addressed. Sometimes it will have been addressed, and sometimes it will not have been addressed. If an article has already been moved to draft space once because it was not ready for article space, and is back in article space, but is still not ready for article space, then, if you are willing to state that it is still not ready for article space, an AFD nomination is in order. An alternative then is to put a {{notability}} tag on the article, and maybe another editor will write an AFD. Moving the article back to draft space a second or third time is move-warring and is disruptive. AFD is sometimes necessary. Robert McClenon (talk) 01:01, 31 May 2023 (UTC)
A Different Copy-Paste Question
I have a different copy-paste question. I just reviewed a draft that is identical to a section in an article, because the originator copied it without attribution from the article. What I normally see is that an article was copy-pasted from a draft. In this case the draft was copy-pasted from an article. I have declined the draft. Is there anything further that I should do, such as tagging the draft? By the way, it is Draft:Tav-Prasad Savaiye. Robert McClenon (talk) 01:00, 1 June 2023 (UTC)
Nothing much. Unless the editor intended to expand the article further? – robertsky (talk) 03:37, 1 June 2023 (UTC)
The redirect Military style has been listed at redirects for discussion to determine whether its use and function meets the redirect guidelines. Readers of this page are welcome to comment on this redirect at Wikipedia:Redirects for discussion/Log/2023 April 8 § Military style until a consensus is reached. 64.229.90.172 (talk) 03:27, 8 April 2023 (UTC)
Fixed. Primefac (talk) 21:06, 5 May 2023 (UTC)
Mattdaviesfsic (talk) 10:10, 1 May 2023 (UTC)
Not done: That hypothetical number would not be accurate anyway, since administrators can use AFCH without being on the list, and soon NPRs will be able to as well. Primefac (talk) 13:17, 23 May 2023 (UTC)
