Wikipedia talk:WikiProject Articles for creation

Main pageTalk pageSubmissions
Category — List (sorting)ShowcaseParticipants
Apply — By subjectReviewing instructions
Help deskBacklog
drives

Welcome—discuss matters concerning this project!

AfC submissions Random submission

4+ months

2,594 pending submissions to update

• Are you in the right place?
  • If you want to ask a question about your draft submission, use the AfC Help desk.
  • For questions on how to use or edit Wikipedia, use the Teahouse.
  • Create an article using Article wizard or request an article at requested articles.
• Put new text under old text. Start a new topic.
• In addition to this page, you can give feedback about the AFCH helper script by creating a new ticket on GitHub.
• New to Wikipedia? Welcome! Ask questions, get answers.

• Top of page
• Table of Contents
• Bottom of page

Problems with talking about inline citations

Hi,

At MfD, I note a continuing problem with reviewers making review comments, when declining submissions, referring to inline citations. I think this is probably almost always no-productive to counter-productive. Inline citations are not required for a start class article, so I don't think it appropriate to frequently ask for it. They are only required for content

I note that Wikipedia:WikiProject_Articles_for_creation/Reviewing_instructions#General_standards_and_invalid_reasons_for_declining_a_submission contains appropriate advice that some reviewers are not following.

I think that for difficult cases, where the topic is unsuitable but it is hard to explain, the reviewer is tempted to start speaking to citations. The problem is that no addition of inline citations can help a draft of an unsuitable topic. It can be hard to give an objective statement as to why a draft will never be suitable, but “inline citation” is definitely not close to a meaningful explanation.

A solution i think is like my suggestion at Wikipedia_talk:Notability/Archive_60#Paid_editing.2C_Advertorials.2C_and_Reference_bombing. Ask the author to state which 2 or 3 sources best meet our inclusion criteria of reliable third party secondary source coverage directly discussing the topic. “More sources”, inline or not, do not help and just make it harder to review. --SmokeyJoe (talk) 23:39, 23 October 2017 (UTC)

The only issue I have with your comment (and the guideline) is that per WP:BLPSOURCES there must be inline citations in a BLP article. For companies, parks, cars, electronics? You're right, refs alone are fine. Really the "ilc" decline notice should only be used for BLP drafts. Primefac (talk) 23:43, 23 October 2017 (UTC)

I declined a draft earlier today for exactly that reason. The draft itself is a promotional effort, which had been full of COPYVIO material. There's no claim of notability but I was happy to decline for something like WP:IC instead of just refusing for notability reasons. The era of "general references" is over, anyway. I'm not approving a draft in that state. Chris Troutman (talk) 23:54, 23 October 2017 (UTC)

• Yes, inline sources are for AfC demanded for BLPs. That's fine, although I note in passing that an new article with sources that aren't inline will not be deleted or draftified at AfD, but fixed, and so AfC is reviewing to a higher standard. I think, Chris Troutman, it is completely unhelpful to decline per WP:IC something that should be declined for serious reasons, like copyvio, promotion, and woefully-failing notability. Mentioning IC seems to cause the authors to reference bomb their draft with more worse sources, and leads to what is called "tendentious resubmissions", I think I observe. Draft:Ferenc Moldoványi may be a BLP, but attending to inline sourcing is not what the author needs to be told. Can you comment on my suggestion that the author should be asked to nominate 2-3 sources that he submits best meet WP:42 (or any of the many sourcing/notability tests)? I think it will stop the author from wasting his own time, and it will make it so much easier to review the draft topic's notability. --SmokeyJoe (talk) 00:58, 24 October 2017 (UTC)
• Draft:Winners Chapel Munich is the draft at MfD that got me inspired to post here. Not a BLP. Not close to being acceptable, but ilc was consistently used in rejecting comments, and ilc completely fails to convey the required message. --SmokeyJoe (talk) 01:01, 24 October 2017 (UTC)

I agree we should not be requiring inline citations of brand new users, except for BLPs and extraordinary claims. Draft:Winners Chapel Munich has been correctly rejected for notability several times. Some of the rejections are for insufficient sources, which is another way to get at notability. Legacypac (talk) 07:06, 24 October 2017 (UTC)

• Taking the first example in the section immediately above shows a continuing very poor standard of AFC reviewing. The thoroughly (but imperfectly) referenced Draft:Malaya Akulukjuk was commented on saying that the inline references should be in the form of footnotes. The advice to consult the Teahouse was well meant but inadequate. The submission was later declined as "a BLP that does not meet minimum inline citation requirements" again requiring footnotes and suggesting wrongly that there were no inline references. Why are BLP requirements applied to a biography for someone with a claimed death of 1995? The second example Draft:Robert Kost (Artist) was again being wrongly treated as a BLP. Hope is now being raised for this draft but in this case also the page's creator has departed these shores long ago. In both cases the standard of reviewing has been shamefully low. On another matter raised earlier in this section, some editors have standards for article inclusion that are stricter than those of the hoi polloi. From XFD and DRV discussions they are well aware of consensus against them. Their judgement may be right and the rest of us may be wrong. However they should realise that they should not be taking unilateral AFC decisions – nor should they be allowed participation. Thincat (talk) 08:43, 24 October 2017 (UTC)

@Thincat: Yes, I agree with your comments in general. With particular focus on "[the page] creator has departed these shores long ago" I would be wary before assuming that two bad reviews are what caused the page creator to stop editing. I often find when reviewing old pages that the talk page notification of "your draft has been declined because of XYZ" tends to cause said page creator to come back and improve the article, particularly when they have email notifications switched on- otherwise with the backlog and wait at present no reasonable page creator could be expected to wait that long. Additionally another thing I have seen recently is G12 notifications alerting the page creator to come back and have another stab at improving their draft. We are reasonably active at removing poor reviewers with a few recent examples in particular; in other cases you'll often see that someone has left a quiet word of advice for the reviewer in the question on their talk page. Currently, a resubmitted draft tends to be reviewed by a new, different reviewer so I'd like to think that cancels out any prior poor reviewing; I know that I've accepted pages previously declined for I think were spurious reasons. jcc (tea and biscuits) 20:22, 24 October 2017 (UTC)

• The need for inline references only applies to BLPs and to controversial or disputed information. It should not be used in other cases. And in any case, it is very rarely the key problem which is more likely to be lack of sufficient substantial reliable sources for notability (very often accompanied by promotionalsm). The purppse of AfC review is to prevent articles from being entered that will inevitable fail AFD. Its easy to move references inline, if that's the only problem, and it can be done in mainspace--articles do not generally get deleted at AfD for this problem. Not having references to show notability is something much more fundamental. It's important to give the right assistance to newcomers.

The only way to give the right assistance is to not limit oneself to the prebuilt templates. I almost always use custom and explain what the specific problem is. Sometimes I will use a general template and to indicate what the special problem is--the problem here is thatthe additional remarks do not show up on the user's talk page, so if I think there's a real chance of improvement, or a need to stop the user from fruitless resubmission I will modify the information that the template has placed on the talk page. I'll only use a general template by itself if the problem is so clear & the draft so hopeless that they say enough--which is about 1/3 of the time/ DGG ( talk ) 08:50, 25 October 2017 (UTC)

User:Thincat if you don’t already have the AfC tools you should get them and use them. Check out the 800+ pages in declined for lacking inline citations/footnotes [1] Legacypac (talk) 11:40, 25 October 2017 (UTC)

I would soon suffer the ignominy of being removed as a reviewer! I know that some articles must have inline citations but I believe that footnotes are never required. My understanding is too aberrant to be an AFC reviewer. More seriously, years ago pre-AFC when I was allowed to patrol new pages, I was disheartened by the amount of dross but when I came to a worthwhile article I was taking longer to check it out than to write something from scratch. Thincat (talk) 12:01, 25 October 2017 (UTC)

I think you do know that footnotes are never required-- any method of referencing is ok for all purposes, (though for blps the it must be able to show specifically where each individual matter is referenced. I think the templates aay "footnotes" just as shorthand, because its by far the most common method. DGG ( talk ) 16:04, 25 October 2017 (UTC)

I was looking through the AfC submissions declined as needing footnotes and spotted a few of my own declines there - I use that reason sparingly and on some occasions perhaps wrongly, so this discussion has been useful. I looked at Draft:Edith Kraft as an example - in this case, there are a bunch of claims made about the living person that are not supported by the one reference, so it looks like the author was relying on their own personal knowledge of them. I was hoping that there were indeed other sources that could be used which would solve both the issue of a single source and also the places where citations were clearly needed. I struggled to find such sources when I Googled. Curb Safe Charmer (talk) 16:42, 25 October 2017 (UTC)

Draft:Lee F. Jackson has been declined for not having inline citations, but in the comments, Caorongjin gives additional reasons - notability, and the article being too short. This is a four times elected Dallas County Judge who served 10 yeas in the Texas House of Representatives, and spent 15 years as chancellor of the North Texas university system. There are articles in four major newspapers cited, so I don't see a notability issue. Curb Safe Charmer (talk) 15:38, 29 October 2017 (UTC)

Curb Safe Charmer, it is fair enough that I was too quick to label it as not a notable subject – happy to be corrected on this point. The main reason for my declining it was for its absence of inline citations. This was true – it had a list of general references but no inline citations. This seems to be Primefac's point above, that "per WP:BLPSOURCES there must be inline citations in a BLP article." It would be worth clarifying if this is true. Is it that BLP must have inline citations, or BLP must have inline citations if it refers to "contentious material about living persons" (per WP:BLPSOURCES). Caorongjin (talk) 22:58, 29 October 2017 (UTC)

Layout

The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

---

I'm wondering if declines based on layout are valid. DGG left a message on my talk page a few moments ago and I think that in some cases it can be a valid reason to decline. We should tell newbies how to fix problems but not do it for them. Learning WP:MOS basics is important in making constructive contributors so I see no reason why an AFC decline on major structural issues should be a problem. Consensus?

Pinging @Nick, Primefac, TonyBallioni, Legacypac, and Kudpung: as potentially interested parties.

DrStrauss talk 21:38, 28 October 2017 (UTC)

• MOS is not a valid reason to decline as it is optional and to be taken with common sense. If there are major structural issues they are normally pretty easy fixes (headings, realist, etc.) so declining based on that doesn't make sense in my mind. TonyBallioni (talk) 21:53, 28 October 2017 (UTC)
• Also, I find The only reason for declining an AfC is that it would be deleted in mainspace, either by speedy for such critical problems as copyvio, or at AfD for notability. concerning, DGG. We're not here to make bog-standard articles, we're here to try and make a good encyclopedia. TonyBallioni, I know that per the guidelines it isn't but I'm asking the question, should it be? DrStrauss talk 22:04, 28 October 2017 (UTC)
• Formatting is not a content issue. Formatting can certainly hide content issues (or in some cases, hide good content), but it's not a reason in and of itself to decline. While I agree with you that we should be guiding new users to the MOS (or preferably, the CHEATSHEET), if the only thing standing in the way of an article being acceptable is layout / formatting / headers, just fix it. There is no good reason why a perfectly valid draft-cum-article should be deleted and have to slog through the month-long review process simply because someone forgot to put their headers inside ==s.

Now, if there are other reasons on top of terrible formatting, then don't feel that you "must" fix them (and in that situation it's perfectly fine to spell out what needs doing), but the decline itself should be for the content-based reason. Primefac (talk) 22:30, 28 October 2017 (UTC)

• No, personally I don't use it as a decline rationale and I don't think it should be a decline rationale. Most of the time, when I see layout issues (e.g. incorrectly formatted headings) I fix them myself under the reasoning that'll take me a minute or two at most- this applies to referencing as well (just today for example). Looking at the example in particular I have to agree with DGG; I would almost definitely fix it myself. jcc (tea and biscuits) 22:35, 28 October 2017 (UTC)
• (edit conflict) This issue seems to be generating more heat than light. And I think it's because we're getting tangled up in two false dichotomies. First, and despite what all of the guidance pages tell us, the appearance of an article is not completely separate from the question of whether it will be deleted. The surest way to avoid being deleted at AfD is to not get nominated in the first place, and a poorly-structured, poorly-formatted article is much more likely to attract the attention of a deletionist. By minimizing the chance of attracting a deletionist's attention, a well-written and well-formatted article does indeed improve its chances of survival. The other false dichotomy is accept/decline. There is a middle course. When I see an article that looks to be appropriate for Main space, except that it has more formatting/style problems than I care to fix myself, I'll start a discussion on the Talk page. In that discussion (to which I'll ping the draft's creator), I'll identify the problem and inform the creator that I'm prepared to accept the article as soon as the problem is addressed. When appropriate (such as in the case of bare URLs), I'll go ahead and fix a few of them and let the new editor use my edits as examples for doing the rest. I've never come across a new editor who insisted on poor formatting. Instead, they usually fix the problems within a day or two and thank me for making them aware of techniques that they hadn't yet learned. The only disadvantage of my approach is that the draft remains in draft space for an extra two or three days. The benefit, of course, is an accepted article that is better looking and more likely to avoid being deleted. Do the others here see this as a useful approach? NewYorkActuary (talk) 22:41, 28 October 2017 (UTC)

  I find it to be a delightfully appropriate way of handling the situation. I watch almost every active AFCH user's talk page, and I have seen plenty of situations where that exact situation has occurred. Primefac (talk) 23:08, 28 October 2017 (UTC)

Then it needs to be made clear that it is wrong. It is utterly febrile to assert that structure is more important than content. scope_creep (talk) 23:28, 28 October 2017 (UTC)

• If an article is appropriate, notable, readable, and adequately sourced, tone and layout can never be a reason not to accept. We don't refuse such articles at NPP, and Wikipedia has an entire catalogue of templates for tagging articles for various imperfections - see Wikipedia:Template messages/Maintenance. AfC reviewers, on releasing a draft, are free to apply them. Kudpung กุดผึ้ง (talk) 01:24, 29 October 2017 (UTC)
• There are many cases where an inappropriate style indicates copyvio or undeclared paid editing. But in general the place where an article is most easily fixed is in mainspace, where the hundreds of excellent WP:wikignomes make there very important contributions. Certainly we want a high quality encyclopedia, but articles almost always achieve that quality gradually.

The standard is not just passing NPP, but whether it will passing AfD. which most of interpret as not just a mere 50:50 chance, but a reasonable likelihood--ideally, that it will not be seriously challenged in good faith. DGG ( talk ) 04:14, 29 October 2017 (UTC)

• DrStrauss it sounds like your understanding of AfC has strayed quite a bit from the community consensus, and it might be good idea to refresh yourself on the guidelines. DGG is almost quoting them verbatim when he says that the only reason to decline is major problems that will lead to deletion. AfC is not about 'training' newbies or incubating high quality articles. It literally is just churning out bog standard ones, on behalf of editors who don't have the technical ability to do so. Anything else can be dealt with through cleanup tags or, like NYA says, comments and discussion. – Joe (talk) 08:20, 29 October 2017 (UTC)

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

I was pinged but the discussion was closed before I got a chance to comment. I've some random (very random) musings about this to add.
There's a fine line to walk between declining an article for layout issues and declining an article because it has more significant issues such as tone, it reads like an advertisement, is a direct copyright violation or is heavily paraphrased.
I'm not so sure I agree that accepting badly written articles without trying to work with the author to improve the layout/appearance is such a good idea. The vast majority of new editors I speak to in the IRC help channel want to write good content, they want it to have the same layout as existing articles and want to get things right. How often do we see people replicate infoboxes using complex HTML syntax, for example. There's quite often a lot of dedication to trying to do things right.
I would be interested to know how many of the really badly written but otherwise acceptable articles which are accepted through AfC are being brought up to an acceptable standard by the WikiGnomes, and if so, how long it's taking. I would suggest we try to measure this over a three month period by assessing how many new AfC articles get copyedit tags, have tone or advertisement tags added, or indeed, how many have notability tags added, and then measure how long it takes for those tags to be removed.
There's the potential here to try and help new editors write better content by understanding more clearly the issues they introduce and help them avoid making edits which introduce these issues. I say this quite a lot these days about various aspects of Wikipedia, but we're always approaching the issue as experienced editors who know from memory how to add infoboxes, format references and the like, whilst not always being at all helpful at explaining (or understanding) the issues new editors encounter. I really don't think it is helpful to take a badly written page from a new user and hand it off to someone else to improve, we should be doing much more to help users fix their own pages, rather than doing it for them. I think that would be good from an editor retention viewpoint (though I do accept that's a less practical proposition when someone is writing one article about their employer or a client).
It may also be possible to better structure the improvement of newly accepted but poorly written content in a more streamlined process by providing better maintenance categorisation so our WikiGnomes can more easily find content they may want to improve and which suits their areas of expertise. I do think, for anything we improve without the involvement of the new users, it would be nice to generate some sort of easy to understand report which could go on their talk page with a basic overview of the fixes undertaken. I wonder if a bot could auto-generate such a report after maintenance tags are removed. Nick (talk) 09:24, 29 October 2017 (UTC)

Another Prof

Per related discussion above and both WP:AFCR and WP:PROF guidelines; can we lose this draft Draft:Ahmad Milad Karimi for improper tone? It has not been edited since declining and likely to remain so until deleted. I decided to accept it but since I objected to reviewer unilaterally overriding another reviewer's judgment, I brought it here for advise. –Ammarpad (talk) 08:42, 29 October 2017 (UTC)

Hi Ammarpad, I think those double quotation marks means they quoting some text, probably a review of one of his books or an article. It is not way I do it, but it seems to be an accepted version. Here is the German version of the article in DE wikipedia:[2]. Here is is Munster page as prof: [3]. PHD in 2013. Google Scholar doesnt have a lot on him, although there is a name mentioned for the university of Kurdistan, but that link is not mentioned in his CV. scope_creep (talk) 10:28, 29 October 2017 (UTC)

@Ammarpad: Traditionally we don't "lose" (my reading is you want to delete) drafts after a single decline. If the page gets to 5/6 months unedited, it'll be swept up in the CSD:G13 (Stale AFC submissions) harvests. If the page is repeatedly resubmitted for review without making substantial improvement, then we would conduct a MFD to determine if the community needs to keep the content. Hasteur (talk) 13:20, 29 October 2017 (UTC)

The tone does need fixing--we normally put such praise a little less prominently. I intend to fix it and accept the article. GS is not particularly useful for German humanities scholars. And the notability standard at deWP is higher than ours'. DGG ( talk ) 18:48, 31 October 2017 (UTC)

Discussion at Wikipedia:Village pump (policy)#Should Wikipedians be allowed to use community granted tools in exchange for money?

You are invited to join the discussion at Wikipedia:Village pump (policy)#Should Wikipedians be allowed to use community granted tools in exchange for money?. Regards:) Winged Blades of Godric^{On leave} 07:53, 2 November 2017 (UTC)Template:Z48

Template update request

Can someone more tech savvy than me update Template:Ffu talk to include a level two header? I'm getting really tired of manually typing "Files for upload" every time I use this. GMG^talk 12:50, 2 November 2017 (UTC)

 Done. Primefac (talk) 12:54, 2 November 2017 (UTC)

Umm... Primefac... something isn't quite right there. See for example User talk:50.64.119.38, before I undid it, when you click edit on the IP's talk page it takes you to the edit for the template itself. GMG^talk 13:00, 2 November 2017 (UTC)

Bleh. I undid the header, and then realized that you probably can't very well see what the problem is without there being a problem to see. GMG^talk 13:01, 2 November 2017 (UTC)

 Fixed. The header doesn't show up on the template but it transcludes properly to the final location. Primefac (talk) 13:02, 2 November 2017 (UTC)GMG, the issue was that you weren't substing the template. Primefac (talk) 13:43, 2 November 2017 (UTC)

No payments for reviews

Under "How to get involved" I've boldly placed

• Review solely on a volunteer basis. Soliciting or accepting payment for a review is prohibited.

This shouldn't have to be said, but it appears that it must be, e.g. User:KDS4444 asked to join as a reviewer on October 2, 2017 at Wikipedia talk:WikiProject Articles for creation/Participants. This was promptly and correctly turned down here. KDS4444 has now been banned for doing somewhat similar work at OTRS and there is a discussion at Wikipedia:Village_pump_(policy)#Should_Wikipedians_be_allowed_to_use_community_granted_tools_in_exchange_for_money.3F

I'm not married to the wording or even the placement of this requirement, but it looks like we have to have this somewhere. Smallbones_(smalltalk) 18:34, 2 November 2017 (UTC)

Shame it has to be said, but I totally agree with adding it. – Joe (talk) 19:34, 2 November 2017 (UTC)

Measurement of conflict of interest

Hello. I am doing a preliminary survey requesting about 6 Wikipedia volunteers to answer a few questions. I am posting the survey to 3 boards, and I choose to post here because I think the project here is relevant. Thanks if a couple of people could respond below.

I just created an essay and presentation at Wikipedia:Measuring conflict of interest on Wikipedia. In that page, I say that the Wikipedia community does not have good information about how often COI editing occurs or the extent to which it is a problem or benefit. I argue that the Wikipedia community needs some data on this issue.

I am not a data scientist, and I do not know how to design a valid social survey. Since this is wiki, as an amateur I am collecting some initial community thoughts as preliminary research. I want this survey data to help guide initial conversation and also to aid in asking around if anyone already has data of this sort.

Survey instructions

• Note that a "conflict of interest editing event" is any Wikipedia activity where a Wikipedian interacts with someone doing conflict of interest editing. This could mean reviewing a new article submitted by a COI editor, or responding to a COI editor's request, or reviewing a COI editor's Wikipedia article edits
• "The project" here refers to "WikiProject Articles for Creation"
• To participate, you must self-identify as having experience interacting with conflict of interest editing events
• Make educated guesses without doing additional research.
• Finish the survey in 3 10 minutes. Try to give your current thoughts without investing much work in this experimental survey.

Please volunteer #1 edit here

1. What is your best guess for how often the volunteer team in this project oversees conflict of interest editing events?

   50% of all submissions have a potential COI

2. Imagine all the conflict of interest editing events you have seen in this project. Divide them into two groups, "acceptable or productive contributions" and "unacceptable or unproductive contributions". What is your guess for the percentage of conflict of interest editing events which are "acceptable or productive contributions"?

   50% of COI contributions are productive, so 25% of total contributions here

3. What is your best guess for how many hours Wikipedia community volunteers spend in this project addressing conflict of interest editing events?

   We spend roughly 50% of our time addressing unproductive contributions and since I've estimated that 50% of all contributions have potential COI, I'd estimate 25% of time is spent on issues that have a potential COI component.

4. Imagine that Wikipedians in this project with your level of experience provide 10 hours of support to other volunteer Wikipedians. What is your best guess of how many hours of "acceptable or productive contributions" Wikipedia is likely to get in return for that 10 hour investment?

   20 hours

5. Imagine that Wikipedians in this project with your level of experience provide 10 hours of support to conflict of interest editors. What is your best guess of how many hours of "acceptable or productive contributions" Wikipedia is likely to get in return for that 10 hour investment?

   2 hours

6. In 2017 English Wikipedia has about 40,000 active editors. In terms of Wikipedia experience, where do you think you rank among editors? Bottom 20%, second 20%, third 20%, fourth 20%, and or top 20%?

   Top 20%

7. Please sign your username.

   ~Kvng (talk) 20:25, 14 November 2017 (UTC)

Please volunteer #2 edit here

1. What is your best guess for how often the volunteer team in this project oversees conflict of interest editing events?

2. Imagine all the conflict of interest editing events you have seen in this project. Divide them into two groups, "acceptable or productive contributions" and "unacceptable or unproductive contributions". What is your guess for the percentage of conflict of interest editing events which are "acceptable or productive contributions"?

3. What is your best guess for how many hours Wikipedia community volunteers spend in this project addressing conflict of interest editing events?

4. Imagine that Wikipedians in this project with your level of experience provide 10 hours of support to other volunteer Wikipedians. What is your best guess of how many hours of "acceptable or productive contributions" Wikipedia is likely to get in return for that 10 hour investment?

5. Imagine that Wikipedians in this project with your level of experience provide 10 hours of support to conflict of interest editors. What is your best guess of how many hours of "acceptable or productive contributions" Wikipedia is likely to get in return for that 10 hour investment?

6. In 2017 English Wikipedia has about 40,000 active editors. In terms of Wikipedia experience, where do you think you rank among editors? Bottom 20%, second 20%, third 20%, fourth 20%, and or top 20%?

7. Please sign your username.

Comments?

Thanks, comments here please. Blue Rasberry (talk) 19:12, 2 November 2017 (UTC)

Finish the survey in 3 minutes? Really? I didn't realize this was "flash respond to questions". Six rather detailed questions will take most people more than three minutes to complete. Primefac (talk) 19:21, 2 November 2017 (UTC)

@Primefac: I changed it to 10 minutes. I hesitate to ask for serious thought rather than impressions. I do not want to consume many people's time, and also, I thought that I could compare flash responses to flash responses easier than I could calculations which different people might do in different ways. Blue Rasberry (talk) 19:42, 2 November 2017 (UTC)

October AFC statistics

Hey all, some of you may know that I've been keeping tabs on the active/inactive lists for the AFCH access, as well as monitoring the number of drafts being submitted and reviewed. With a marked (and steady) increase of unreviewed drafts lately, I thought I would post the October data for this project. This is in no way meant to be "shaming" or otherwise critical of the reviewers (most of which are pulled in sixteen different directions) but to potentially start a discussion about how we can tackle the backlog.

Detailed stats can be found at this page. All of the stats were compiled on 5 Nov and any +/- numbers are based on the data taken on 8 Oct.

• Total drafts submitted: 8054 (259.8 per day)
• Total drafts reviewed or deleted: 7594 (245 per day)
• Pending submissions: 1760 (+460)
• Number of reviewers: 151 (+3)
• Out of 4826 undeleted reviews:
  • 124 reviewers with 1+ reviews
  • Maximum number of reviews: 1078
  • Reviews by the "top ten": 3024 (62.7%)

There has been a marked increase in the number of daily AFC submissions in the last two months, likely due to ACTRIAL (the numbers for earlier this year were ~180 submissions per day), but I don't think we have an impossible hurdle to jump. If 1/10 of the AFCH users reviewed just one more draft per day, we'd start decreasing the backlog. I think that's a manageable goal, and I'll do my best to start doing just that. Primefac (talk) 14:09, 6 November 2017 (UTC)

I think part of that is that the ones remaining take more time to review while the reviewed ones are quick fails. But even then there are quite a few easy ones in the very old queue, and it doesn't look like all that much. Also the very old queue seems to have fallen by ~40 since yesterday, which is good news. Galobtter (talk) 14:30, 6 November 2017 (UTC)

Oh, definitely. I found a very old a few weeks ago that was 100% cv, but it looked "impressive" so it's likely people just passed it by. However, I don't want to say that we must review from one end or the other; if we do that we'll lose reviewers (and last month was the first month where we didn't lose reviewers!). Primefac (talk) 14:45, 6 November 2017 (UTC)

Do we need a catchy slogan? How about "One a day keeps the backlog away"? NewYorkActuary (talk) 17:59, 6 November 2017 (UTC)

I like it! Primefac (talk) 18:04, 6 November 2017 (UTC)

I’m a willing reviewer but am inappropriately restricted from moves to mainspace. All to do with a stalker. I mark pages as likely acceptable when I find them and focus on deletion otherwise.

We could cut down the size of the backlog by CSD or MfDing hopeless pages. Cut down on the resubmissions and discourage rather than encourage users from wasting time on unacceptable topics. Legacypac (talk) 20:11, 6 November 2017 (UTC)

I'd say that's reasonable, and I know you've been doing a fairly good job of working towards that goal. Primefac (talk) 20:12, 6 November 2017 (UTC)

We now have a super huge backlog filled with 2410+ pending submissions according the AFC script, which most of them over a month ago. KGirl ^{(Wanna chat?)} 21:07, 6 November 2017 (UTC)

Um.... false. There are 2,594 pending submissions. Other than enterprisey's toollabs counter, where are you seeing this 2400 figure? Primefac (talk) 22:20, 6 November 2017 (UTC)

@Primefac: It appears when it's finished declining or accepting a submission using the script (e.g: Continue to next random submission (24xx)) but these figures do not appear on other templates. KGirl ^{(Wanna chat?)} 22:43, 6 November 2017 (UTC)

Ah, yes, the script is the only thing that's still counting the categories incorrectly. Primefac (talk) 22:53, 6 November 2017 (UTC)

Feedback page

I noticed that the message for acceptance has a message : "If you would like to help us improve this process, please consider leaving us some feedback." that links to Wikipedia:WikiProject_Articles_for_creation/feedback. That seems essentially a black hole because it isn't linked anywhere here as far as I can see. I do see some feedback that might be useful. I think it'd be good if this page was linked somewhere on the project page (maybe under help desk?) cos I didn't even know this was a thing until I saw that. Also that page reallly needs archival. Galobtter (talk) 07:13, 8 November 2017 (UTC)

Christ, that needs archiving. Really sad no one on the project side noticed. Good catch! ProgrammingGeek ^talktome 12:50, 8 November 2017 (UTC)

Before archiving, is there someone who would be interested in reading the feedback and summarizing anything useful here? ~Kvng (talk) 20:08, 14 November 2017 (UTC)

Reliable sources template

At the AfC help desk, we have some templates for common questions asked by users (ie help desk, reference desk, etc). I wrote a template that I think could be included, to help users whose submissions were not sourced properly to find the relevant guidelines. I'd like input as to whether I should add it it the editnotice. It's found at {{subst:User:ProgrammingGeek/afc-rs}}. Obviously, I'd move it to a subpage of WP:AFCHD if it's liked. Thanks.

Generic Template

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.

Template with user specified with {{subst:User:ProgrammingGeek/afc-rs|Username}}

Hello, Username. 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.

ProgrammingGeek ^talktome 14:44, 8 November 2017 (UTC)

Shouldn't the decline already have that? If so then the person probably needs a little bit more specific advice on the help desk or has some other problem. I don't see the point of boilerplate advice at the help desk. Galobtter (talk) 15:13, 8 November 2017 (UTC)

Looks good to me. If people are using it often, it makes sense to have it in an easily accessible location. I've shifted it to {{AFCHD/rs}}, but if anyone really thinks that it shouldn't be used it can always be moved back.

For what it's worth, we do have a half dozen similar boilerplates. Primefac (talk) 15:17, 8 November 2017 (UTC)

As far as I can see, those other boilerplates are about when the question is inappropriate for the help desk. Galobtter (talk) 15:21, 8 November 2017 (UTC)

True. However, I've created a few userspace templates because I got tired of writing stuff over and over. If someone was told "this needs better references" and they immediately turn to the AFCHD and ask "what's wrong with my draft?" then I see no reason for a reviewer to have to re-type the message. Obviously if there is more to say they should say it, and there's nothing say this is all that should be said. But not having to type it all out is kinda nice. Primefac (talk) 15:23, 8 November 2017 (UTC)

Yeah of course. Just have to keep in mind that too many boilerplate responses may not be really helpful. Galobtter (talk) 15:25, 8 November 2017 (UTC)

Potential time limits on resubmissions

I propose this to reduce the relatively constant backlog of submissions. Declined submissions are often immediately re-submitted with little to no changes, and this process repeats. The amount of articles that are being accepted is not enough to offset the number of new submissions (because declined ones often get re-submitted).

If we could time-stamp each denial, and disable the button on the AfC template if 7^ish? days have not passed yet.

Of course, it could be manually re-submitted, but few AfC users will understand this.

Please give thoughts. Thanks. ProgrammingGeek ^talktome 13:57, 9 November 2017 (UTC)

I don't know. It should be relatively easy to check if improvements have been done and quickly decline, and improvements can be done in one day. Galobtter (talk) 14:01, 9 November 2017 (UTC)

The issue is not really tenditious resubmissions with zero changes; those declines take maybe thirty seconds to perform. Yes, if you're talking about a hundred such pages that's almost an hour of your life, but that's unlikely given that there are only about 250 submissions per day. I would be more concerned with Galobtter's point (which I have seen) - no point in hamstringing well-meaning editors just because we have a few numpties about. Primefac (talk) 14:04, 9 November 2017 (UTC)

On the other hand, we did at one point discuss a bot that would revert a resubmission if it was literally the next submission after a decline, but people didn't like the idea of bots declining drafts. Primefac (talk) 14:06, 9 November 2017 (UTC)

No, I agree. Having a bot decline drafts is probably just a giant middle finger to potentially productive editors. That's news to me. I'm shocked it was even discussed. ProgrammingGeek ^talktome 14:10, 9 November 2017 (UTC)

One thing that would be useful if the AfC script automatically showed the changes since the last decline in a openable diff - or just had a link to the diff "declined vs cur". That'd make it easy to see if improvements have occurred. Going to the page history is pretty annoying. (of course, as Jcc points out, you should also check if you agree with the previous review) Galobtter (talk) 14:11, 9 November 2017 (UTC)

That's a good idea, but is the AfC script still actively being maintained? I went to their GitHub out of curiosity yesterday, but it didn't seem very active. ProgrammingGeek ^talktome 14:12, 9 November 2017 (UTC)

Maybe a script can be developed. Shouldn't be too difficult too search the history for the last "declined". Bugfixes are done, something about multiple decline reasons was being developed, though the script isn't too active. Galobtter (talk) 14:14, 9 November 2017 (UTC) Addendum: read "shouldn't be too difficult" as "I have no idea". Galobtter (talk) 14:19, 9 November 2017 (UTC)

(edit conflict) The short answer is "yes" - AFCH has kind of reached a "stable" point in its life, so there's not much that needs updating on a regular basis. However, when changes are needed the script is updated. Primefac (talk) 14:15, 9 November 2017 (UTC)

Some editors (quite rightly sometimes) just want a re-review, which doesn't take that long to complete given you either agree with the review given by the previous reviewer or don't, and leave a comment to that effect. I see no reason to impose a time limit. jcc (tea and biscuits) 14:07, 9 November 2017 (UTC)

• Re-submissions with no meaningful improvements have little effect on the intractability of the backlog; as Galobtter observed, they can be declined quickly if you concur with the earlier review. If the submitter hasn't made a good faith effort to improve the draft, then rather than promptly declining it, consider skipping it and working on the oldest submissions instead. Having to wait in line can discourage over-hasty re-submissions.

Another way to discourage ill-advised re-submissions is to leave a brief comment with each decline to drive home the point. The boilerplate bio-notability decline links to two policies, three guidelines, three essays, and two help pages. All good stuff, but they leave a lot of novices confused about what exactly is wrong with their draft. They may change one thing and re-submit in the hope that they've fixed everything. If sources 1-3 are not independent, source 4 is user-generated and thus not reliable, and sources 5-6 contain only passing mentions instead of significant coverage, tell the submitter that. --Worldbruce (talk) 15:19, 9 November 2017 (UTC)

Yeah since you anyway (usually) have to check the sources for whether it is passing mentions or whatever, you've done most of the work so I spend a minute or two to leave at least a sentence but usually a few on why it is not notable. Galobtter (talk) 15:22, 9 November 2017 (UTC)

These quick resubmissions will hurt our productivity more the faster we re-review them. If you want to limit the frequency which a given draft is resubmitted and improve our ability to address the real backlog, go review submissions that have been sitting longer awaiting review and let these resubmissions wait their turn. ~Kvng (talk) 20:07, 14 November 2017 (UTC)

Agree 100%. Rather than going for the quick declines, reviewing and giving feedback to the ones that actually could be an article is more important. Galobtter (talkó tuó mió) 13:48, 15 November 2017 (UTC)

Lint error

{{AFC submission}} uses the parameter |small=yes for re-rejections, but this parameter triggers a Lint error of type Missing end tag, leaving a <small> tag unclosed. For example, in declining Draft:Christopher Roach, this line was added:

• {{AFC submission|d|ilc|u=Jnice2k3|ns=118|decliner=Atlantic306|declinets=20171023225648|small=yes|ts=20171023142638}}

{{AFC submission}} should be modified so as not to leave behind an unclosed <small> tag. —Anomalocaris (talk) 01:03, 15 November 2017 (UTC)

I'll get right on this. Primefac (talk) 13:44, 15 November 2017 (UTC)

 Fixed. Primefac (talk) 15:20, 15 November 2017 (UTC)

Primefac: Thanks! —Anomalocaris (talk) 16:51, 15 November 2017 (UTC)

Remove the "backlog drives" tab

There's consensus against holding backlog drives, so I think that the backlog drives page should be removed from the tabs at the top of the page. ProgrammingGeek ^talktome 00:19, 22 November 2017 (UTC)

Can I have a link to a discussion(s) regarding consensus against backlog drives? I was thinking about planning one for New Page Patrol for January, but would like to review the opinions expressed about their efficacy and productiveness before making any decision. Cheers. — Insertcleverphrasehere ^{(or here)} 00:38, 22 November 2017 (UTC)

One second. Also, I think that this was limited to AFC, so NPP would be fine ProgrammingGeek ^talktome 01:40, 22 November 2017 (UTC)

Got it. I vaguely remembered Dodger67's comment here, if he'd like to chime in that'd be good too. ProgrammingGeek ^talktome 01:47, 22 November 2017 (UTC)

(edit conflict)I realize that the comparisons are not identical, but the concerns (of quality over quantity) are very similar. There has been some talk of giving awards based on maintaining an ongoing 'streak' of reviewing i.e. reviewing 1/5/10/15/20 articles per day for X days. Though given that some of us are weekend warriors (or opposite) I think I would do the streaks as a 'per-week daily average'). I.e. awards that emphasize ongoing quality reviewing over a period of many weeks/months rather than bulk "as quick as you can get 'em done" reviewing. This approach might be very good for AfC as well. If we go ahead, I might let you know how it works out. — Insertcleverphrasehere ^{(or here)} 01:54, 22 November 2017 (UTC)

From past discussions I gather that the last backlog drive suffered from poor quality reviews which, even though they were the fault of only a small number of participants, soured many regulars on the idea of backlog drives. Surely, however, we can devise rules for participation and scoring that would avoid the problems of the past.

One thing to keep in mind is that we're in the middle of the six-month WP:ACTRIAL. One of the many things it is supposed to measure is the effect on AfC. Holding the first backlog drive in many years during the trial could distort those measurements. --Worldbruce (talk) 15:46, 22 November 2017 (UTC)

It should still probably be there, as a past record. Maybe something like "Backlog drives (inactive)"? We could have a system like GOCE has, where a small fraction of reviews are checked for quality, and if they're not up to par no points will be given. I personally have a number of reviews done userbox, which is enough for me. Galobtter (pingó mió) 15:53, 22 November 2017 (UTC)

• TLDR: We shouldn't get rid of the tab. I think if it's done right a backlog drive would be good, but we should wait a few months.

It's fairly clear, just from me visiting the page (incidentally for the first time ever) that the page is very unused. I don't think we have to worry about people suddenly expecting prizes based on something that last occurred in 2014.

That being said, the main reasons for a general dislike of backlog drives is that they have (historically) produced a lot of very poor acceptances. Past experience should not (always) be a reason for future activities, but in this case I am somewhat inclined to agree with that sentiment. There have been times I myself have wanted to just approve the entirety of "Very Old" just to clear it out.

And with that being said, I have been watching the NPR discussion, and I think (based on how that backlog drive goes) we might be able to implement something similar. However, I think we should wait until after that drive, and ACTRIAL "unofficially" ends (there's some predicting that it will never actually "end" because no one will remember it started in the first place), but while I am keeping fairly detailed statistics, I would like to get a few more months of data at "normal" rates/levels before we try to smash everything out at once. Primefac (talk) 16:18, 22 November 2017 (UTC)

There too there is discussion of reviewing the reviewers. Considering the only successful backlog drive system I know of is GOCE where that occurs - makes sense. Agree that we need some more data - and it's probably not the best idea to have NPR and AfC drives at the same time. Galobtter (pingó mió) 16:32, 22 November 2017 (UTC)

Agreed. I'm sure the good people at NPP would love us to multiply their workload and deny them that sense of accomplishment. As Primefac said, we can hold another one in a few months. ProgrammingGeek ^talktome 16:36, 22 November 2017 (UTC)

Just as a minor thought, we used to be regularly hovering around 3000 pages in the queue (back in 2014-15). Thanks to the hard work of a few individuals (a couple of which are still with us) the backlog dropped down to what it was a few months ago (between 600-1000).

The recent surge is undoubtedly due to ACTRIAL, but we have managed to drop big backlogs without drives. We're almost at a net neutral, and last I calculated it all it takes is for one reviewer in ten to perform one more review per day to start decreasing the log. Basically, this means every person should ideally do +1 review per week, which I think is reasonable. Primefac (talk) 16:40, 22 November 2017 (UTC)

Those 20 reviews could mean 20-30% more work though, as there are a lot of easy declines that get done but there are also ones that take longer time. But I've noticed that while the very old cat is now pretty huge - but a lot of it is stuff that would've been previously been caught on the first day but has slipped through because of the surge in submissions, so it isn't insurmountable. So hopefully we can get it down . Galobtter (pingó mió) 16:57, 22 November 2017 (UTC)

Or we could just approve stuff more quickly and be a little less fussy. We can't seem to get any streamlining between AfC and NPR passed so anything approved at AfC goes into NPR like any new article and gets reviewed/tagged/AfD'd by the much larger pool of editors. If I was a new editor - getting a message about serious backlog and a month plus delay I'd be unhappy. People expect instant gratification on the internet. Anything we can do to reduce the backlog should be considered. Legacypac (talk) 18:30, 22 November 2017 (UTC)

I could imagine that leading to things getting accepted and then draftified by NPP. Because my standards are pretty low - though others do have higher standards. Galobtter (pingó mió) 18:57, 22 November 2017 (UTC)

νew Υork City listed at Redirects for discussion

An editor has asked for a discussion to address the redirect Νew Υork City. Please participate in the redirect discussion if you have not already done so.

This mixed Greek/Latin redirect was requested via AFC, so I'd though I'd let you know about the RFD.

-- 70.51.45.76 (talk) 03:41, 29 November 2017 (UTC)