Jump to content

Wikipedia talk:The future of NPP and AfC/Archive 2

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia
Archive 1Archive 2Archive 3Archive 4Archive 5

A little bit of progress

The WMF Community Tech Team (which I manage) has completed the following tasks:

  • Determine the cause of the recently growing backlog (T146959).
  • Request #8: Unreviewed new articles should be marked as noindex (T147544). Note that this code still has to be reviewed, merged, and deployed which may take a couple weeks.
  • Request #9: Make sure articles moved to Main namespace from other namespaces are included at Special:NewPagesFeed. This was already fixed actually, but we tested it to make sure.
  • Request #23: Update Page Curation maintenance tag list to match Twinkle's (T147544). This is now live and can be configured by administrators with JavaScript knowledge: MediaWiki:PageTriageExternalTagsOptions.js.

In addition, MusikAnimal and myself (on our own time) built a prototype for a new article creation notice: User:MusikAnimal/New article edit notice, which the community is free to use (or not). I realize that these improvements are only a fraction of the changes requested. However, in order to tackle the more substantial requests, either Community Tech or the Collaboration Team (which ostensibly owns Page Curation) will need to set aside significant time to plan and execute this as a real development project. I have a meeting scheduled with Katherine late next week in which we will be discussing the NPP situation (among other things). I'll let everyone know if anything comes out of that meeting. Kaldari (talk) 08:09, 6 October 2016 (UTC)

Great news about articles being marked as non indexed until patrolled. This means the search engines will ignore rather than mirror them and will reduce the damage done by attack pages and reduce the incentive for spammers. But to make it effective we need a change to the newpage screens so that articles already tagged for deletion can be excluded or selected. Most patrollers will want to exclude them, but admins are likely to want to select them. I suggest marking them in a separate colour in the feed. Another group it would be nice to highlight are people creating an article where their previous article was deleted G10 or G3. These articles need to be prioritised for checking and probably for blocking the author. With vandals we have a progression of warnings that escalates to a block, but the current system lacks that for page creators. Would it be possible to automatically post an AIV report on anyone who had had two pages deleted per G3 & G10 or three deleted G11 in the last 7 days? ϢereSpielChequers 06:54, 14 October 2016 (UTC)

@Kaldari: I'm interested in more information about this noindex thing. Does this also apply to internal searches or is it just external indexing i.e. robots.txt? ~Kvng (talk) 16:43, 18 October 2016 (UTC)

@Kvng: Noindexing only affects external search engines (same as robots.txt). Our internal search engines don't look at meta tags, so there should be no effect for internal searches. Kaldari (talk) 18:20, 18 October 2016 (UTC)

New edit notice

@Kudpung, Noyster, JohnCD, DGG, The Blade of the Northern Lights, and MER-C: How does this look? Instead of posting at WP:VPR or another venue and having this turn into an unnecessary support/oppose !voting survey, I'm was hoping to attain some rough consensus from you all. This seems mostly uncontroversial, it being a mild distraction with a potentially profound impact on how new users approach article creation — MusikAnimal talk 01:59, 7 October 2016 (UTC)

It looks fine to me; communicates the message in a polite and pithy manner. I leave any stylistic comments to others. The Blade of the Northern Lights (話して下さい) 02:43, 7 October 2016 (UTC)
Good, but the description text is too squished. Change letter-spacing to 0.2-0.4px, or change the font-weight to 300 and the font-size to 85%. — Esquivalience (talk) 03:21, 7 October 2016 (UTC)
Thanks for the ping MusikAnimal - I hope you meant This seems mostly uncontroversial ... Fine by me, I just wonder if the red button, instead of directing to the main page, could take the new editor to somewhere offering help with whatever they were wanting to do, like Help:Contents: Noyster (talk), 12:04, 7 October 2016 (UTC)
Looks great- I'm delighted that we really seem to be getting somewhere with this. Perhaps there could some button for help- an IRC link perhaps? jcc (tea and biscuits) 16:08, 7 October 2016 (UTC)
Definitely meant uncontroversial, thanks Noyster! I have implemented all of your recommendations, including a new bullet point with links to get help. Going to deploy this now. Thanks, all! — MusikAnimal talk 17:33, 7 October 2016 (UTC)
 Done! :D — MusikAnimal talk 17:45, 7 October 2016 (UTC)
It's too in-your-face. KATMAKROFAN (talk) 18:48, 7 October 2016 (UTC)
It's for new users, so it's supposed to be a bit in-your-face. You can simply scroll down so that you don't see it, if you find it distracting. Or add #newarticletext { display: none } to your personal CSS — MusikAnimal talk 18:51, 7 October 2016 (UTC)
"It's for new users"? I've only been here for a few months and I find the new version WP:BITEy. It doesn't need an image, the text is too big, and links are used instead of buttons in most editnotices. KATMAKROFAN (talk) 19:43, 7 October 2016 (UTC)
It may be worthwhile to put something to make people think twice about autobiographies/resumes if it can be done without cluttering it up. The other thing is just a personal aesthetic preference. I find the colour scheme is too bright/loud particularly on a bright white background but I prefer duller colours overall so...
Great job and thank you! JbhTalk 19:10, 7 October 2016 (UTC)
Thanks! It's the "cluttering up" part that I'm worried about. Wikipedia:Your first article gives info about autobios, as does the Article Wizard. That's about as far as we can go. The only truly super important thing is to stress the importance of references for BLPs. Beyond that we kind of have to let them figure it out, or else we're back at TL;DR. The loudness I think is also favourable if we want to ensure it is read. I realize experienced Wikipedians aren't going to be too pleased, but hopefully they'll understand the audience it is benefiting — MusikAnimal talk 19:19, 7 October 2016 (UTC)
Yeah... I can see how bright would be good. With everything going through AFC My First Article is good enough I guess. I was hoping there could be a way to catch all of the "me" "me" impulse/vanity creations since there are a lot of those. I can not think of a non-cluttering way to do it without saying Wikipedia is not the place to write about yourself which is not really the friendly vibe we are looking for JbhTalk 19:49, 7 October 2016 (UTC)
  • I think the version I made was probably less cluttered and the buttons less imposing, but that's just my personal choice as a graphic designer. @KATMAKROFAN:, I see absolutely nothing bitey about it at all. It will become clearer when you have addressed the issues with your own editing (and if you need any help with that, don't hesitate to ask me). Kudpung กุดผึ้ง (talk) 23:11, 7 October 2016 (UTC)
    Somebody at Village pump (policy) doesn't like it. I like it. (I see that I can bypass it by entering data in the empty form, also.) I wasn't actually trying to create a new article, but to view the deletion status of a red link. I succeeded in doing that and in seeing it. Robert McClenon (talk) 23:28, 7 October 2016 (UTC)
    The message is now only shown to new users, and for confirmed users the old message is shown :) I don't know why I didn't think of that beforehand, there is user-group specific CSS. This is excellent because now we can look into adding User:MusikAnimal/Article search with no result to MediaWiki:Search-nonefound (once we tweak it to our liking). That message is very in-your-face, intentionally, but in that case too in-your-face for experienced users — MusikAnimal talk 02:56, 8 October 2016 (UTC)
  • Looks good, though I think stuff about being an encyclopedia should be mentioned somewhere. Is anyone going to obtain metrics to see whether this has the desired effect? MER-C 05:42, 8 October 2016 (UTC)
    I don't think there's a way to do metrics without significant effort (mw:Extension:EventLogging), no. But I can tell you that from my work with abuse filters, big flashy messages definitely help with newbies over simple plain black and white text. Whether they actually follow the advice is anyone's guess, but if we see a spike in AfC submissions it's probably because of the new edit notice.
    I've created Template:Newarticletext-unconfirmed and am now transcluding it in MediaWiki:Newarticletext (also Template:Newarticletext-confirmed), that way we can work on it without risk of message up the core interface message. Do note the bold red comment in the documentation to keep the message brief. I would argue a sixth bullet point is overdoing it. We're already pointing them to WP:YFA, which explains pretty clearly the do's and don'ts, so I personally feel what we have will suffice — MusikAnimal talk 23:03, 9 October 2016 (UTC)

Compromise proposal

The Foundation and a minority of editors including myself want to keep open editing and a route into the community for the 25% of newbies who start by creating a new article. The majority of the community want some sort of restriction on newbies creating articles. Rather than the ACTRIAL proposal, what do people think of throttling account creation so that if you create one page you can't create a second until your account has been confirmed or your first page has been marked as patrolled? (Once we deploy the new Noindex until patrolled feature we can stop patrolling articles that are tagged for deletion). ϢereSpielChequers 11:34, 14 October 2016 (UTC)

That does pretty much nothing. A huge amount of articles are done by 'one and done' editors. Puttering around enough to become autoconfirmed is not that huge a burden. Maybe it will even give them time/motivation to read our content guidelines. The number of editors lost by not being able to throw something straight into the encyclopedia will be counterbalanced by the number of editors who do not leave because their first article was quickly deleted. There will be the added benefit of addressing a real world problem which the community has been complaining about for years with a solution overwhelmingly approved by the community.

Ideology/Idealism is well and good but there is a practical problem that is drowning Wikipedia in crap that must be solved. ACTRIAL was put off for years and the problem only became worse. It is past time to try the solution proposed and approved by the community. If it fails, we try to figure out something else but continuing to do nothing or mess about with half measures is, in my very, very strong opinion, neither wise nor sustainable for the project. JbhTalk 11:54, 14 October 2016 (UTC)

THe NOINDEX is not a new feature. It was introduced years ago on consensus but nobody knew it was broke until I asked the devs to fix it. My fault for assuming someone else would report it.
Many people are saying now that the ACTRIAL criterion is not strict enough by today's problems. At least the consensus is even stronger than it was then with even some of the former opposers crossing the floor. Let's just not forget that ACTRIAL wasn't just about preventing non autoconfirmed users from immediately creating in mainspace, it actually channeled them through the Article Wizard instead. There are now enough various local technical solutions to deploy ACTRIAL without going through Phabricator, and that's probably what will happen - but not just yet. Let's see if we can get some new page patrollers of the right calibre first. Kudpung กุดผึ้ง (talk) 12:54, 14 October 2016 (UTC)
@Jbhunley, I'd dispute you on the nothing bit, but I don't dispute it alone would do less than Actrial. However in life sometimes you can achieve more by seeking compromises and solutions that work for both sides. Another step would be to do what the German Wikipedia does and prompt editors for a reference. ϢereSpielChequers 18:26, 14 October 2016 (UTC)
Maybe not nothing but likely indistinguishable from it. In general I do not see nor do I believe that there are many newbies creating multiple articles in their first 4 days. Rather there are new editors who come in to make an article on themselves, their company, favorite band, company etc. - a single article. That does not include the UPEs who come in and throw up promotional articles in one edit.

Yes, I would love to see new articles required to have a sources. I would go so far as to extend BLPPROD to everything and say it could be validly applied to any new article without an RS supporting a specific claim. This is because UPEs are figuring out that they can put in some crappy ELs and be able to duck a BLPPROD. I really doubt we would get consensus for that but sure, in that case, I would have no problem with immediate article creation. Notwithstanding how unlikely getting consensus on that is, I think it would raise the chance of new editors being discouraged because a huge portion of them would get tagged with an Extended-PROD for lack of RS. JbhTalk 18:25, 18 October 2016 (UTC)

@WereSpielChequers: without dragging ourselves back to what went down with ACTRIAL, what exactly are we trying to negotiate? Is it as Jbhunley says a tradeoff between ability for new users to create articles immediately on the one hand and the amount of crap this puts into the encyclopedia on the other? ~Kvng (talk) 16:39, 18 October 2016 (UTC)

Yes, this particular compromise would be a tradeoff, though I'd rather talk of compromise and seeking alternatives. I actually believe that we can build a happy medium, one that keeps a route open for new editors who start by creating new articles and also reduces the flow of crap and or makes it easier to reject. The frustrating thing about the entrenched positions on this is that what the two sides want to achieve isn't diametrically opposed, it is just their preferred methods of achieving that. If the philosophies were opposed then a consensus position would at best be a compromise that both regarded as suboptimal. But I think it should be possible to achieve something that isn't a tradeoff and instead achieves both sides objectives. The tricky thing is that this would require the WMF not just to be willing to invest in software, they do that a bit, but to invest in software the design of which has been agreed with the community in advance. This is culturally challenging as it conflicts with their self image as the leaders and designers for the movement. ϢereSpielChequers 22:31, 18 October 2016 (UTC)
I think we're in a fairly compromised position :) with the way things are currently set up. As you say, different people want to solve this problem in different ways and we have multiple solutions active so what's the problem? New articles are either not indexed (NPP) or remain in draft space (AfC) until someone is able to review them. There is a low barrier to creating these. There is consensus that there needs to be review on new submissions. The problem is we just don't seem to consistently have enough qualified volunteer manpower to do these reviews at the rate they're coming in. I challenge the assumption that we need to eliminate the backlog. A long backlog will eventually serve to discourage new contributions and maybe encourage new editors to join the project and the system should reach equilibrium at some point. And if it doesn't, so be it. We'll review new submissions at whatever pace we can sustain. WP:NODEADLINE. ~Kvng (talk) 15:43, 19 October 2016 (UTC)
There is no need to look for alternatives - ACTRIAL was backed by an overwhelming consensus from one of the lsrgest RfC we have ever known. Those still looking for alteratives are those who are either WMF staff who constantly need to impress the world with a raw statistic for new articles, those who wrongly believe that it would reduce the number of new articles any more than the barrier did that was put in place in 2006, and those who haven't actually done any patrolling of late to appreciate how the profile of the same 1,000 articles per day has radically changed. I've been doing NPP on and off for many years and I can definitely say that there is very little in the New Page Feed these days that is not blatant corporate spam, blatant bio & autobio spam, soccer players, and plain rubbish. I hardly ever get round to actually patrolling a new article as OK because as I have a deletion button I'm deleting pages faster than I review the others in the feed. NPP today is totally boring and it's hardly surprising so few want to work there. Closing AfDs by contrast can be quite entertaining. Kudpung กุดผึ้ง (talk) 16:54, 19 October 2016 (UTC)

NPP backlog getting down

The current backlog is 13,578, which is a bit down from 16K we had two weeks ago. I am not sure though how sustainable it is, and if it is, what is the reason.--Ymblanter (talk) 18:00, 18 October 2016 (UTC)

I believe that one reason is all the new RfCs that are being made. They are being publicised in many places, which results in more people receiving interest. Dat GuyTalkContribs 18:10, 18 October 2016 (UTC)
  • A lot of it is because of poor, and very fast, reviewing by new editors like this one WebCite - Curration log. They do not seem to want to take the hint to slow down either, several complaints and attempts to get through to them are in their talk page history. JbhTalk 16:18, 19 October 2016 (UTC)
They made a specific request not to be "bothered" about their patrolling [1] and they have been typically unresponsive. I figured a ping from me is likely to only annoy - they already have lots of "unreviews" from me today. I guess it would be polite to ping them though @WebCite: since they are likely to end up being discussed rather than simply being noted as an example.

I'm just wading through their log as I have time trying to find the worst of the errors. I think they have speedie-reviewed more articles than they have made edits! JbhTalk 17:14, 19 October 2016 (UTC)

NOTE: For this kind of discussion which is not directly related to the technical reform project, a new talk page has specifically been created at Wikipedia talk:New pages patrol/Reviewers (the page at WT:NPP now has another purpose). We hope to see this same kind of vibrant discussion over there.

Past RFCs

RfC to create a 'Special:NewDraftsFeed' system for example, maybe others — also essential reading? --Gryllida (talk) 20:05, 20 November 2016 (UTC)

Interesting reading perhaps, because it does reiterate the purpose of the Draft system. That old RfC has however been made largely defunct by later suggestions that point towards incorporating a Draft feed in the special:Newpagesfeed. --Kudpung กุดผึ้ง (talk) 22:41, 26 November 2016 (UTC)

Integrating NPP and AfC (a bit)

If you'll forgive the intrusion of a talk page lurker, I had a couple of ideas for hopefully quite easily implemented things that could start to bring the NPP and AfC processes more in line with each other:

  1. Since they have the same requirements, restrict the AfCHelper script to people with the new NPP right rather than maintaining a separate list. The benefit being that AfC reviewers have to be proactively vetted by an admin, rather than retroactively removed from the list if they didn't read the requirements properly. Existing AfC reviewers could be grandfathered in or asking to reapply of WP:PERM, which has the side benefit of making them all aware of NPP.
  2. Mark drafts that have been accepted through AfC as patrolled automatically. They've already had at least one pair of competent eyes look over them so we might as well ease the NPP backlog a bit. Not sure how technically feasible this is, though.

Then NPPers and AfC reviewers would be in the same 'pool' and there'd be a recognition that they're essentially doing the same task. Anyway, I thought this workgroup would be the best people to run the ideas by. Joe Roe (talk) 04:00, 26 November 2016 (UTC)

My quick thoughts is that right now at PERM, AfC participation is being used to demonstrate competency for the NPP right- but if this continues and people can "prove" themselves through AfC reviewing, that does presume that things going wrong at AfC with a bad reviewer are more easily spotted and not as important as a bad reviewer at NPP? Let me know if I'm not making any sense. jcc (tea and biscuits) 23:22, 26 November 2016 (UTC)
Jcc, AfC participation is not being used to demonstrate competency for the NPP right. Because both systems require similar, in-depth understanding of policies, guidelines, and above all, a mature manner of interaction, it is reasonable to assume that AfC reviewers who have clearly demonstrated skill and competency in those respects may well be suitable candidates for NR, and that's why they have been sent an invitation to apply for the NPR right - which does not pre-suppose they will get it.. Administrators never accord user rights on the basis of the minimum numerical requirements alone without also performing significant scrutiny.
New Page Patrolling is an essential function, nothing else can substitute it. AfC however, is not an essential service; it's a bonus service that policy does not require Wikipedia to provide, and without it, the quality of new mainspace articles would not be affected. That however, is not to belittle the excellent work that some its operatives actually do but ,compared to NPP, it only has a handful of new submissions to process every day. One of its problems us that like NPP, although a very large number of editors like to claim they are regular patrollers/reviewers, in reality the bulk of the work is done by only a tiny fraction of them, and that is what causes the backlogs that users like me and DGG are so acutely aware of and why we are looking for doable solutions.
The logical step therefore, as suggested by a significant number of users already, would be to combine the functions of the AfC Helper Script into the Page Curaton tool and have both systems running simultaneously from there. It therefore doesn't need AfC to be closed down, which is also what a lot of people want and it would be a win-win solution. Kudpung กุดผึ้ง (talk) 02:11, 27 November 2016 (UTC)
No need to apologize for posting here, Joe Roe - with accurate observations like these and the concern you feel, you should consider joining our work group.
Many people are already aware of #1 and all AfC reviewers have already been messaged a while back about New Page Reviewing, suggesting they apply for the new right (which does not however pre-suppose that they;ll get it automatically). On #2, I don't personally agree that this would be a good idea. For the same reasons that NPR is not accorded automatically to AfC reviewers. The standard of AfC reviewing still has a long way to go. I'm one who is monitoring it very quietly in the background and also watching all the additions to the user list. A lot of those are users who have been waiting until they have exactly 500 edits but have no other appropriate experience. Some of them are gaming the system in order to approve their own articles.Kudpung กุดผึ้ง (talk) 01:41, 27 November 2016 (UTC)
On the subject of marking drafts approved through AFC automatically as patrolled, I have a question. Yesterday I accepted several articles that had been in the AFC backlog into article space. I then looked at the New Page Feed to see if either I needed to mark them as patrolled or if they were already marked as patrolled. They did not show up. Am I correct that this means that they had already been marked as patrolled while they were in draft space? If so, I assume that they had not been indexed, and were not indexed by Google until I accepted them into article space. Also if so, am I correct that new pages in draft space will be on the New Page Feed, but that reviewers will only be validating that they do not need speedy deletion, e.g., as G10 attack pages or G11 spam? Robert McClenon (talk) 02:43, 27 November 2016 (UTC)
Hi Rob. To your first observation, the current situation is that all drafts that are published to mainspace by an AfC reviewer are supposed to appear in the New Pages Feed along with any pages moved from any namespece to mainspace, where they will be reviewed and which then releases them for indexing by Google. We either need confirmation from the devs, MusikAnimal and Kaldari that this has been checked, processed, and implemented.
To your second question, this is a more complex issue. It has been suggested that without deprecating the AfC system entirely, that newly created drafts should appear in the New Pages Feed as a reviewer selectable filtered list. At the same time, similar features to he AfC Helper Script would be added to the Page Curation tool. Kudpung กุดผึ้ง (talk) 04:09, 27 November 2016 (UTC)
Thanks for the responses. I guess my second suggestion is dependant on the first and can really be generalised into the observation that there's something illogical about the fact that if I (for example), as an AfC reviewer with the NPP right, move a draft to mainspace it's still added to queue for someone else to review. Similarly, I suppose that when an NPPer creates an article they're also marked as needing review, even though we trust them to review other people's contributions. So another approach might be to automatically give NPPers the autopatrolled right too? That way potentially problematic AfC reviews are still double-checked while reviews by vetted editors skip NPP. Admittedly this might be far too much effort for a quite small reduction in the NPP workload... Joe Roe (talk) 16:56, 28 November 2016 (UTC)
I support these incremental improvements to our reviewing policies. ~Kvng (talk) 11:57, 6 December 2016 (UTC)

RFC on hasty tagging at VP

There is a pending RFC at the Village Pump (proposals) to require a 30 minute delay after article creation before tagging an article for deletion under A7 and perhaps other criteria. Individuals interested in that topic should opine there. Regards, TransporterMan (TALK) 22:57, 2 January 2017 (UTC)

RfC on the scrapping of the AfC process

There is an RfC at WT:Drafts asking if the AfC process should be scrapped altogether, which participants of this project may be interested in. Best wishes, jcc (tea and biscuits) 20:20, 22 January 2017 (UTC)

Backlog drive!

I've seen other projects (AfC for example) use backlog drives to increase participation for a defined period of time - do you think the WMF would make available some funds for some prizes? Maybe a couple of Amazon vouchers and a WMF t-shirt? Hell, I'd throw £10 in towards an Amazon voucher for the "winner". Of course, we'd have to make it clear that quality of patrol > number of patrols. Yes we're all volunteers and we do this for free - but prizes are great motivators.. Thoughts? -- Samtar talk · contribs 10:56, 6 February 2017 (UTC)

Asked some initial questions (just seeing if this would be under grant scope) to the rapid grant guys - need to know if this is something we'd a) want to do in the near future and b) be willing to coordinate together -- Samtar talk · contribs 11:06, 6 February 2017 (UTC)
AfC is no longer doing backlog drives. The drives were causing more problems than they were solving. Specifically, the quality of the reviews took a dive during the drives. ~Kvng (talk) 14:19, 9 February 2017 (UTC)

Proposal to unpatrol moved pages

There is currently a proposal at the village pump that bears directly on new page patrolling:

– Joe (talk) 15:41, 11 February 2017 (UTC)

WMF Report

For those who do not follow the NPR talk page, the Wikimedia Foundation has just issued a report with recommendations as to the future of new pages patrol at Wikipedia:New pages patrol/Analysis and proposal TonyBallioni (talk) 23:18, 31 May 2017 (UTC)

Moving forward

Now that the WMF has issued its report and proposal for the future of New Pages Patrol and editors have had time to review and respond directly to the report, I would like to ask the question in a more general sense of what the next steps are for what I think everyone agrees is a process in need of reform. In particular I see there being three questions critical to this conversation:

  1. What are the practical measures that are relatively uncontroversial that can be taken now to improve the NPP process?
  2. What are the people's thoughts of what I perceive as the core proposal of the WMF report: having a end date for the new pages feed when pages would fall of?
  3. Is there still support to attempt to implement ACTRIAL, and if so, what steps should be taken in regards to it?

I personally think that there have been some good suggestions to the first question on the report talk page and think the answer to number two is that the backlog won't be fixed by pretending it doesn't exist. I also still support ACTRIAL and think that the WMF has made a poor argument against it, especially when their own numbers show that it would stop approximately the amount of page creations that the backlog was growing by (see T149021.) I've posted this here rather than the report talk page so that the conversation can be both about it, but also discuss things not covered by it as needed. TonyBallioni (talk) 00:43, 7 June 2017 (UTC)

My thoughts on ACTRIAL have been tempered somewhat. Some people write good content with their first registered edit. This does not mean that good content is encouraged by encouraging newcomers at start writing new content immediately. I have three ongoing thoughts:
(1) Auto-{{welcome}} all new registrants. Possibly give them an opt-out of auto-welcoming in the registration process, but note that registration is public, the new registrant is listed as Special:Listusers. The welcome templates give newcomers a better ordered list of helpful reading and suggestions. Old arguments of the data burden of these seem very outdated, and answer is WP:PERFORMANCE. Auto-welcoming
(2) Redesign the presentation of the main WP:AfC page. It encourages newcomers to start an article without doing any reading. Go to the page. In your face is "Welcome to Articles for Creation" in really big letters. Then there is a blur of not easy to read find text. Then there is a big blue box with big white text that says "Click here to create an article now!" Its size, and prominence, deman attention and imply that you don't need to read the text above it. When you click on the big blue button, the interface brings up some back-tracking instructions (are you really ready?), but the newcomer has already been put into the wrong mindset.
The best redesign, in my opinion, is that matching Template:Welcome, where the link to new page creation is at the fifth dot point, and it is not made to look more prominant than the dot points above it. I think Template:Welcome is beautiful in its simplicity, and AfC is over-decorated.
(3) Similar to my comments here: treat new article draft writers more like adult Wikipedians, and less like children newcomers. Talk to them the normal way, per WP:Talk. On the talk page, new signed comments go below the earlier comments. Simple, un-templated, minimal wikimarkup. I guess the current AfC habits derive from when new articles were on subpages of WT:AFC. That was appropriately changed, but changed incompletely. Anyone ready to create an article knows about talk pages.
--SmokeyJoe (talk) 01:26, 7 June 2017 (UTC)
the WMF report is insightful and nails exactly why I don't want to accept many pages in NPP. The expected standand is so high, and if you miss the standard you get dragged to ANi and lose the NPP user right. The correct standard should be "Is it elegable for CSD, need to go to AfD or really belongs in Draft space?" If not, let the mass of editors improve the topic already. Legacypac (talk) 01:36, 7 June 2017 (UTC)
Perhaps we should give their idea a chance. I support a trial of the WMFs proposal by means of the addition of two new features to the page curation tool. Allow reviewers to hide unreviewed articles older than n days and another filter out articles that they have already skipped. Mduvekot (talk) 02:13, 7 June 2017 (UTC)
I don't think those would be bad features. I would oppose a full scale trial by actually making them disappear, but allowing individual reviewers to control the timeframe they are reviewing more easily by means of filters would be positive. The other filter that would be very helpful in my mind is allowing to filter for pages without citations like we can for orphans and categories. TonyBallioni (talk) 02:19, 7 June 2017 (UTC)
I would very much like to see a filter for articles without citations. I have no idea how hard it would be to implement, but we already have a note in the feed saying No citations in bold red. That suggests they can get that info. Mduvekot (talk) 03:41, 7 June 2017 (UTC)
  • I love the idea of being able to filter the feed by date. I am strongly against entries dropping off after XX days. I still think ACTRIAL is fundamentally a good idea for reasons I don't think the WMF addressed in its report. I've been following the discussion on the report page, but haven't commented. I'm out of town right now, but will probably comment more next week.~ ONUnicorn(Talk|Contribs)problem solving 02:33, 7 June 2017 (UTC)
Can we modify the tool to exclude what we skip and filter out over X days? Would we need WMF to do that? Seems like WMF is telling us the solution - lower the approval bar to get out of NPP list and age out stuff that's not been deleted. Anything else is not going to solve the que. Legacypac (talk) 02:55, 7 June 2017 (UTC)
That assumes the backlog is a problem in itself and not a symptom of the actual problem of not having people checking pages for quality, which is something needed if we want to align with what the WMF has as one of their strategic goals for the next 15 years. I think hiding skipped pages would be a net negative, since what you skip I might find easy to review and vice versa. As for a fall-off date: we already have a page that does that if someone prefers to do pages 30 days or younger: Special:NewPages. TonyBallioni (talk) 13:40, 7 June 2017 (UTC)
I use Special:NewPages if I want to get to 30 days, but try getting to anything between 8 May 2017 and 14 February 2006 and you're scrolling for a really long time. Those pages in the tail of the log are not sitting there because they're too hard to review, as the WMF seems to think, they're sitting there because they're inaccessible. I just tried scrolling for 5 minutes from the back of the backlog and I'm only at 9 January. Mduvekot (talk) 15:43, 7 June 2017 (UTC)
Yes, I agree with that. Some sort of filtering by date would be a huge plus. TonyBallioni (talk),
I assume the suggestion is to hide what I skip from me and what you skip from you, not hide what you skip from me or I skip from you. Do you skip me yet? Legacypac (talk) 17:42, 7 June 2017 (UTC)

My responses to the three questions posed by TonyBallioni:

  1. NPP should focus on deleting obviously inappropriate material. This will not eliminate the backlog but I beleive it is our best bang for the buck towards keeping the crap out of the encyclopedia.
  2. Regardless of how I personally feel about it, I don't think it will be possible to get a consensus on the WMF proposal. We need a new proposal.
  3. I don't support bringing up ACTRIAL again in this context because the report has (despite some flaws in the data) made a reasonable case that ACTRIAL would not fix the backlog issue. Also it's just not gonna happen and further harping on it is divisive and has no benefit.

As I've said on the other page, I am in favor of ignoring the backlog. Some have argued that this is ignoring, not solving, the problem and so it is a non-solution and we are looking for a solution. For those who insist on a solution, we must either improve our ability to patrol and/or reduce the flow of new articles. The WMF has made it clear they're not interested in entertaining the latter. There is no deadline for implementing NPP improvements however the most recent improvements have decreased our reviewing capacity and if we continue with that type of improvement, we will never have a solution. ~Kvng (talk) 18:47, 7 June 2017 (UTC)

We can't really reduce the inflow. WMF is essentially saying "lower your standards" amd instead of 1500 editors bearing the load, let the entire pool of editors work on it. Legacypac (talk) 21:44, 7 June 2017 (UTC)
We can reduce the inflow any time we like by locally implementing ACTRIAL. The WMF might not like it but there's nothing they could do about it, The WMF however, sees ACTRIAL as a simply preventing non confirmed accounts from creating articles, but in fact ACTRIAL as planned had a more holistic approach and catered for the good faith creators who had simply not read the instructions, while at the same time the Foundation was supposed to be creating a proper landing page for all newly created accounts. It's quite clear for example, that plenty of accounts get created for the sole purpose of vandalism or deliberteltely creating other types of junk pages - I know, because I've blocked hundreds of them. Kudpung กุดผึ้ง (talk) 15:08, 8 June 2017 (UTC)
Retrieved from "https://en.wikipedia.org/w/index.php?title=Wikipedia_talk:The_future_of_NPP_and_AfC/Archive_2&oldid=1142662734"