Wikipedia talk:The future of NPP and AfC/Archive 1
![]() | 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. |
Archive 1 | Archive 2 | Archive 3 | → | Archive 5 |
2016 community wishlist
In November 2016 the Wikimedia Foundation will begin soliciting proposals for the next "community wishlist" at meta:2016 Community Wishlist Survey. The meta:2015 Community Wishlist Survey resulted in some great successes for the community - the projects selected by that survey are ones that the Wikimedia Foundation probably would not have done otherwise, but as a result of the survey, they did these things with community support.
I am unsure of the technical needs of NPP/AfC reform, but is there anyone here who could articulate a request which could be provided through Wikimedia Foundation staff development of software or other WMF support? If so, this upcoming wishlist survey could be an opportunity to both solicit community support and, with that, get a commitment for staff time from the WMF.
If this project is seeking ideas for getting started, making an early proposal in the community wishlist survey could be a good way to get attention and community feedback. An early proposal would be seen by more people sooner, and by the design of the survey, earlier proposals are more likely to get more votes of support. If the work group being organized here can come to agreement from the posting of the proposal to the wishlist, and for everyone to give support from the beginning, then that would likely draw even more attention to the idea.
If I can help anyone here make a proposal to the wishlist then ping me. I am not sure what kind of solution is best, but I do recognize the problem to address. Blue Rasberry (talk) 16:30, 27 September 2016 (UTC)
- To be honest, the chances of getting something up are low because there's 10 years of neglect, software rot and technical debt to catch up on -- the TODO list is way too long, and I suspect we will have to wait for a couple of years. I myself have at least five things to propose this round. Server configuration like that needed for ACTRIAL is explicitly outside the scope of the Community Tech team. That said, things to consider:
- A new landing page that explains what Wikipedia is and isn't and points newbies towards improving existing articles. Kudpung will be better able to explain this one.
- Is there anything about Special:NewPagesFeed that impedes your productivity? Are there any features that are obviously missing?
- Is there anything about MediaWiki in general that slows down patrolling and the necessary followup (e.g. of sockpuppets)? Are there any features that are obviously missing? (I can name several, such as not being able to search for deleted pages.)
- Do you have any ideas for new tools?
- Hope this helps. MER-C 04:04, 28 September 2016 (UTC)
- The WMF software department writes programs that it likes to write, so the chance of getting new NPP features is almost zero. I think that we should write the necessary software without waiting for the WMF to do so. — Esquivalience (talk) 11:14, 28 September 2016 (UTC)
- MER-C If you are thinking about the history then you are thinking of a time before 100 million dollars of annual investment. Contrary to popular Wikimedia community opinion, there is a difference between what can be accomplished with a budget of 0 versus what can happen with a budget of $$$$99999999. Playing the proposal game might not be right for you but someone else might have an idea, and I hope if someone else posts something then you could support.
- Esquivalience I encourage you to explore non-WMF options if you see fit. In 2015, there were a list of wild ideas that WMF staff would not have chosen, and most of them have now been granted. At the time of the survey I think the Wikimedia community had little expectation of anything coming from it but I have been very happy and surprised with the outcomes. It is true that WMF software engineers do what they want, and they have had some shockingly horrible ideas in the past, but I like the idea of community support for the wishlist as a community shakedown of the WMF in which the castle provides bread and circus to prevent another user revolt and defenestration like the last time. We just had the spectacle of a beheading and are still in a state of anarchy so I wish you could have a little excitement about the potential for a power grab while a feeding frenzy is still possible. This is a popularity contest for ideas and if anyone wants to play the game then the option is available.
- Give this a look - there might be something here. Blue Rasberry (talk) 14:07, 28 September 2016 (UTC)
- A "landing page" proposal was submitted by JohnCD to the 2015 wishlist survey, though not under that name. It didn't do very well and was withdrawn before the end, on advice that WMF involvement was not necessary to implement this: Noyster (talk), 14:54, 28 September 2016 (UTC)
- The problem is that in spite of its millions of dollars surplus, the WMF practically refuses to anything that they think they can get done for them for free by community volunteers. My argument is that we, the volunteers, did not found Wikipedia, and they should spend the money on salaried devs to do whatever is required to protect the quality and reputation of their cherished Wikimedia movement. At the end of the day, without the volunteer content creators and content maintenance workers, there would be no no Wikipedia to appeal for donations for to pay their bloated salaried contingent and (some) bloated salaries.
- The wishlist is at Wikipedia:Page Curation/Suggested improvements and a 17 page one was sent to the WMF at their behest over two months ago and not a peep yet of thanks, or even acknowledgement of receipt by those concerned. Kudpung กุดผึ้ง (talk) 09:37, 29 September 2016 (UTC)
- The WMF software department writes programs that it likes to write, so the chance of getting new NPP features is almost zero. I think that we should write the necessary software without waiting for the WMF to do so. — Esquivalience (talk) 11:14, 28 September 2016 (UTC)
- I would like to thank Bluerasberry for his detailed, thoughtful, and very accurate post. It does however illustrate how absolutely important it is to read the project page here and to follow up with the linked pages before commenting. Article Creation Work Flow (landing page project) was a Foundaton project which was the other half of Page Curation. To understand any of this, one also needs to understand what motivated a group of users to WP:ACTRIAL, which in hindsight, because it could in fact have been implemented locally with filters, should have been done by the community as a user revolt. After all, en.Wiki has rarely seen an RfC with so many participants and such an overwhelming consensus.
- What should be done now, IMHO, rather than re-invent the wheel, is to:
- take the excellent Landing page development that the WMF begun and simply complete it. And this is why I also don't consider the issues listed at Wikipedia:Page Curation/Suggested improvements to be simply part of anyone's wish list. I regard them as fundamental core issues that for some reason the Foundation refuses to address and complete in favour of their relentless crusade to spend funds on nice little gadgets, rather than the serious stuff that is the very fabric of Wikipedia content. Foundation initiated wishlist projects are simply their red herrings to draw our attention away from the real issues. The problems surrounding the control of new content are not be be confused with fancy gimmicks such as syntax highlighting, visual editor, image viewer, notations, or flow.
- Address he issues at Wikipedia:Page Curation/Suggested improvements - Quiddity (WMF) is suggesting a shortlist of the five or so most important ones but it's a bit late for that now.
- Merge AfC and NPP into one, solid, well 'staffed' new article control system.
- What should be done now, IMHO, rather than re-invent the wheel, is to:
- By all means let's explore non-WMF options, but we should be able to expect some return for the donations that our content generates that pays the WMF their salaries. We have a lot to do at our end first though and sadly, either that revolt may be lingering on the horizon, or Wikipedia is going to lose more newbies and even more of its established users. Kudpung กุดผึ้ง (talk) 14:01, 30 September 2016 (UTC)
The WMF software department writes programs that it likes to write, so the chance of getting new NPP features is almost zero
. I can't say anything "official", but I can say this is simply untrue. Please speak your mind at the survey — MusikAnimal talk 06:26, 1 October 2016 (UTC)- But this is absolutely true, MusikAnimal: ...the WMF practically refuses to do anything that they think they can get done for them for free by community volunteers, and whatever the internal squabbles were at the time, the WMF today has absolutey no excuse for not completing Article Creation Work Flow (landing page project) now. Right now. If they say again that funds are not available, it is a monstrous red herring. Their own reputation for quality encyclpedias depends on good policing of content and with a 15,900 NPP backlog up from 7,000 a few weeks ago, something has to be done. It's in their own interest to deploy ACTRIAL and the landing page. One problem is that they won't tell us who is in charge. Kudpung กุดผึ้ง (talk) 08:29, 1 October 2016 (UTC)
- @Kudpung: I wasn't around for that, so I don't know... but I beleive in 2012 there wasn't a team devoted to making things the community wants. There is now, and the wishlist is the place to show what there is a demand for. If something on the wishlist is already being worked on by a volunteer, Community Tech will try to work with them to get it completed, but otherwise the wish will not be ignored under the suspicion that a volunteer might at some point do it. For this landing page project, several things I see we can do without any coding at all using interface pages. If you can find me the assets (images) we can create this fairly quickly. The main landing page we could still do with an interface page, but it's a bit hacky, and wouldn't have the hover states. If you want to do the talking, I'm willing to step in as a volunteer developer for this as it seems quite simple... These just look like splash pages that provide simple information. Am I missing anything else? — MusikAnimal talk 15:54, 1 October 2016 (UTC)
- Okay, I see that there is some logic involved, in particular bypassing this process for experienced editors. We could do this with some simple JS, but that would require it be executed on every page... not that ideal. Just mind you again there's MediaWiki:Search-nonefound and MediaWiki:Newarticletext. This appears to be where the landing page info is shown, and we can freely put whatever we want there right now. But at the same time, it sounds like the extension was almost finished before being lost to higher-priority work... so I think this is an excellent candidate for the wishlist, and I encourage you to push for it! — MusikAnimal talk 17:46, 1 October 2016 (UTC)
- I don't think that following the wishlist process should be necessary to finish the last few percent of the extension. I am skeptical of the wishlist because, as evidenced by the abandoning of half-baked software, the WMF likely cannot finish anything major in a year along with nine other projects (even minor ones for convenience), and they haven't proved this wrong since 2011. Quality control is not only high-priority for an organization that maintains the most widely-used reference work, it's absolutely critical work, unlike other technologies that the WMF worked on during this period. For instance, readers can live without Media Viewer, but without proper quality control, they will question the quality of the encyclopedia and where their donations are going, and probably use Wikipedia less. Imagine if for example, the maintainers of Encyclopedia Britannica or World Book were 90% shiny software and 10% content and quality. They'd become a running joke pretty quickly! — Esquivalience (talk) 21:50, 1 October 2016 (UTC)
- Okay, I see that there is some logic involved, in particular bypassing this process for experienced editors. We could do this with some simple JS, but that would require it be executed on every page... not that ideal. Just mind you again there's MediaWiki:Search-nonefound and MediaWiki:Newarticletext. This appears to be where the landing page info is shown, and we can freely put whatever we want there right now. But at the same time, it sounds like the extension was almost finished before being lost to higher-priority work... so I think this is an excellent candidate for the wishlist, and I encourage you to push for it! — MusikAnimal talk 17:46, 1 October 2016 (UTC)
- @Kudpung: I wasn't around for that, so I don't know... but I beleive in 2012 there wasn't a team devoted to making things the community wants. There is now, and the wishlist is the place to show what there is a demand for. If something on the wishlist is already being worked on by a volunteer, Community Tech will try to work with them to get it completed, but otherwise the wish will not be ignored under the suspicion that a volunteer might at some point do it. For this landing page project, several things I see we can do without any coding at all using interface pages. If you can find me the assets (images) we can create this fairly quickly. The main landing page we could still do with an interface page, but it's a bit hacky, and wouldn't have the hover states. If you want to do the talking, I'm willing to step in as a volunteer developer for this as it seems quite simple... These just look like splash pages that provide simple information. Am I missing anything else? — MusikAnimal talk 15:54, 1 October 2016 (UTC)
- As the person who started the discussion that ended up in ACTRIAL, I can only say I support it more strongly than before. I've now had experience patrolling from both the non-admin and admin side, and I really never understood the reason for shutting it down if there was no subsequent implementation of anything meaningful. Tools are helpful, but the true core of the NPP problem is that it's so chaotic that even those of us who know what we're doing have a hard time keeping track of everything; until that's addressed in earnest, it will remain the mess it is now. The Blade of the Northern Lights (話して下さい) 15:06, 1 October 2016 (UTC)
- But this is absolutely true, MusikAnimal: ...the WMF practically refuses to do anything that they think they can get done for them for free by community volunteers, and whatever the internal squabbles were at the time, the WMF today has absolutey no excuse for not completing Article Creation Work Flow (landing page project) now. Right now. If they say again that funds are not available, it is a monstrous red herring. Their own reputation for quality encyclpedias depends on good policing of content and with a 15,900 NPP backlog up from 7,000 a few weeks ago, something has to be done. It's in their own interest to deploy ACTRIAL and the landing page. One problem is that they won't tell us who is in charge. Kudpung กุดผึ้ง (talk) 08:29, 1 October 2016 (UTC)
- The Blade of the Northern Lights, Esquivalience, MusikAnimal, JohnCD, vQuakr, Bluerasberry. I'm going to be frank here - I know I'm known for not minicing my words, but now it's time for some home truths:
- Wish lists are about as serious and effective as a child's letter to Santa Claus at the North Pole. Also, individual users will ask fro some feature without even noticing that there are community groups desperately trying for years to get the same thing done. I have extremely good and well founded reason to have very little confidence that the WMF will actually provide any serious software that is needed unless the community actually threatens and perhaps even carries out a revolt. The Foundation employees are totally ensconced in their world of spending thousands of dollars on gimmicks that are absolutely non essential and which Wikipedia can live without, and which they dream up to keep themselves in a salaried position.
- I am sick of being the only user who is apparently actively doing anything. about NPP - a lot of people join in the discussions but nobody actually does anything. I went specifically to Italy in June to address these issues only to find that a WMF staffer has usurped the meeting I reserved to use it under a similar abstract just to try and convince us again that they are listening to our needs. I've cobbled this project together but still not one of the signatories to the Work Group appears to have read the reading list - if they had they wouldn't be here asking the questions. I'm beginning to feel ashamed of the encyclopedia I've dedicated literally thousands of hours to and I'm sure that Blade and others who have been around a long time feel the same.
- There was nothing 'half-baked' about Article Creation Flow. Brandon Harris is an excellent programmer, understood exactly what I was describing as needed, but he was taken off the project due to some internal squabble at he WMF - probably what led staff like Oliver Keys and Howie Fung to be so rude and demeaning to had-working volunteers during the Wikimania conference in DC in 2012.
- WMF wishlists are just red herrings. They are like the constant research projects, all contrived to make us believe 'something is being done'. What can we do about it? Nothing - even the Board of Trustees is rotten to the core and after all these years there is still absolutely no bridge over the 'us vs them' gap between the communities and the Foundation.
- The issue surrounding the control of new content, whether it be through NPP or AfC, is so critical, that the WMF CEO should be brought into play. It's time for action, It's time for the Foundation to spend some money on essential software, and I'm sorry MusikAnimal if my words sound harsh, you are on the right track, you are wonderfully engaged, but you might possibly be relatively new to all this and you are too trusting of what you hear on the WMF grapevine. Where s your name here?What we are asking for here is not something the en.Wki wants, it's something the entire Wkimedia movement needs and if they don't provide it, it will be their own demise. Already the news media in the UK is awash with snide comments about the quality of Wikioedia. It's become a joke on the busses, the subway, and in the pubs; TV script writers are are using it in their serious TV crime dramas (Lewis, etc) I have absolute, irrefutable hard proof that when the WMF runs a survey, 1). it's often anther red herring. 2). They will totally ignore the real results and publish instead what they want us to hear. What they don't hear is what goes on down here on the factory floor - most of them have never even edited the encyclopedia.
- Even Quiddity (WMF) in spite of his attempts to be helpful is not really concerned with these issues because he's tied up with the WMF's cherished Flow project. I am of the teams of dozens of volunteers who have had to clean up some of the most monstrous and costly (tens of thousands of dollars and junkets) blunders the WMF has made, even after them having been told in no uncertain terms what's about to happen. We were just laughed at and I won't mention the senior staff names here, but email me for the facts if you will. Kudpung กุดผึ้ง (talk) 01:43, 2 October 2016 (UTC)
- Imagine a world where all the Wikipedia volunteer maintenance workers went on strike
- - that moment is closer than the WMF and most of us realise - new page patrolling has already dwindled to a near stop, nobody can be botheted to close the NPP RfC which expired already nearly a week ago, AfC is chaos, and the NPP backlog is increasing exponentially and is nearly at the level that made Blade and I engineer ACTRIAL.
- Please everyone, just do us all a favour, read the pages I have created here, read the links I have provided, and act now. Please start here: Wikipedia:The future of NPP and AfC/To do. Kudpung กุดผึ้ง (talk) 01:43, 2 October 2016 (UTC)
- I understand your frustration and from an engineering perspective I am eager to help. Please review meta:2015 Community Wishlist Survey/Results. All of the top 10 are either finished or are in development, with exception of one that was declined with good reason (probably saying too much but for the record it wasn't Community Tech that declined it). Another five wishes in the bottom 40 or so have been completed, and another 12 are in development. There's only been one year of the wish list, and it went pretty well, I think. Obviously not all 100+ wishes are to be granted in one year's time, but the top 10 are given priority and all are not ignored. You'll probably see many of the same wishes for 2016, which is completely OK, and they'll be re-prioritized based on rank, feasibility, and engineering investment (e.g. some of the lower ones were easy and quick wins). This is a model the German community has used for some time also with success stories. If you don't ask for Article Creation Workflow to be completed, someone else might, but it seems you'd be the best spokesman for it. I have profound respect for you (and can't wait to meet you in Montreal), but I hope you realize the wish list is not a "red herring", and unlike the past N number of years, it is an opportunity to take advantage of engineering resources dedicated to helping volunteers like you. Would it hurt to try? :) I honestly feel like it would one of the quick wins given how close it was to completion — MusikAnimal talk 02:28, 2 October 2016 (UTC)
Please remain seated with your computers switched on until Wikipedia comes to a complete stop
MusikAnimal, I understand your arguments, really I do, but there are a couple of things that everyone still fails to understand:
- I'm not doing this for me or to help me, I'm doing it for Wikipedia because I don't want to see the tens of thousands of hours (and dollars) I've spent ln Wikipedia going to waste.
- I have 100% proof that the WMF is so poorly organised that they polarise and no one ends up doing anything that needs a bit more professionalism.
- I've been nudging the Foundation on and off for years to get this done. I spoke with Quiddity (WMF) in Italy and began a dialog on his MedWiki talk page after having been given to believe that he was somehow involved in this (because he usurped my meeting in Italy, and based it obliquely on my own abstract). I was then called on Skype by another WMF department for an hour nearly three months ago. AS mentioned above, there has been absolutely no further rection from the Foundation.
- Your wishlist suggestion makes sense, but again, why does it always have to be me? A bit of collective pressure from the community would work better. It's not just Artcle Creation Flow that we need completed, it's the rest of the issues on the Page Curation software, someone to close the expired RfC for NPP qualifications, and someone to start the ball rolling on the merging NPP and AfC. One of the best voices would be DGG but he is of course tied up with something else until the end of the year. Kudpung กุดผึ้ง (talk) 03:11, 2 October 2016 (UTC)
(Not directed at anyone in particular, but this has to go somewhere) I'm willing to accept the wish list has had some success with assisting the community, and to that end I'm glad it's here. The problem is that NPP poses an existential problem and needs action now, and the community made it clear what it wanted in 2011 and has yet to see it materialize. Being an American it's a bit like what the Affordable Care Act is to healthcare costs, the responses from the WMF have done some nice things around the fringes but ultimately gave more powerful tools to a large group of people incapable of properly using them; instead of diminishing the problem it's exacerbated it. It's extremely deflating to see, especially when I remember all the work I put in during 2011 (at the tremendous expense of other things in my life at that time) and know that if it materialized it would have prevented all of this. To the extent that I can I want to ratchet up the pressure to get this fixed, since I've also spent many thousands of hours (again, to the detriment of other things) trying to make this a better place. The Blade of the Northern Lights (話して下さい) 03:24, 2 October 2016 (UTC)
- Man oh man. Ten thousand barnstars for you both! :D I've read through the the Article Creation Flow talk pages and see you both are were just as frustrated then as you are now. I can't imagine... you are to new articles reform as I am to vandalism studies, I suppose... Only difference is nothing is holding me back. You're patiently waiting for help from the people who are capable of doing it. This is why I just wanted to shed some light that if NPP-related improvements were to get on the wish list it's bound to become a reality :) And Kudpung it doesn't have to be you who requests it, I just thought you'd be interested! The whole wish list process means it's still a long wait, sure. I think ACTRIAL would be a good first step, and that doesn't take time other than convincing, since it's just a configuration change (I think?). I'm not going to pretend to know what I'm talking about, I wasn't around when that was last proposed. I hope I've been the slightest bit of help in showing there are people on the WMF side who care... because I'm one of them :) And I'm going to look more into the landing page. We may not be able to fully replicate what the extension does, but some of the basic things we can do ourselves pretty easily, and I'm willing to put in the work. Let me track down the image of the cute little Wikipedia guy, create some draft interface pages, and I'll get back to you — MusikAnimal talk 04:30, 2 October 2016 (UTC)
- Following MusikAnimal's suggestion, Blade, I've gone ahead (again, why does it always have to be me?) and pre-registered the two major software issues at 2016 Community Wishlist. Having done that however, I reiterate that I have very little faith in that 'Letter to Santa Claus' project because at the end of the day, the WMF still only takes their survey as a basis and chooses what they actually want to work on.
- Also, a quick look at the items on the 2015 list, AFAICS, none of them are of the urgency, gravity, and importance attached to the Wkipedia/Wikimedia global reputation for accuracy and quality. They are all gimmicks and gadgets that might, but not neccessariy, slightly enhance the workflow of editors. What Ryan Kaldari, who has interjected a couple of times with the occasional helpful comment, apparently fails to understand is that the issues we are discussing and trying to motivate here are above and beyond the scope of his project, and need prioritising from the top down and not from the community up. This is something that Katherine Maher should be made aware of because chances are that she isn't - not that she's likely to even take an interest, it's the kind of thing a CEO waves his hand at a minion and says 'Look into his, would you" and promptly deletes it from his mind. In previous times, I at least had the ears of Vice-CEO Erik Möller but all that changed when Tretikov came on board.
- All I can suggest is that we unilaterally enact the huge consensus we had for ACTRIAL, doing it with a local script or filter. The WMF could hardly 'fire' any of us for it. I wouldn't bother though with all the niceties and templates that Blade, Scott, and I spend hours developing for it. Simply restrict page creation in mainspace to all accounts that have less than 10 edits and are less than 96 hours old. That would cut out most of the totally inappropriate new pages. If I knew how to do it I would do it without any hesitation. Perhaps someone here with engineering skills could come up with the necessary local code.
- As for the Landing Page, during the run up to ACTRIAL I had started to completely rebuild the Article Wizard in php but Brandon said leave it to him and he came up with the Article Creation Work Flow (landing page project) which I think is pretty darned good - there was nothing 'half-baked' about it. AS I said above, Brandon is an excellent programmer, understood exactly what I was describing as needed, but he was taken off the project due to some internal squabble at he WMF - probably what led staff like Oliver Keys and Howie Fung to be so rude and demeaning to had-working volunteers during the Wikimania conference in DC in 2012. IMO it shouold simply be continued. Kaldari would tell you where the code is. However, I still think it's totally unreasonable to expect the volunteers to do this kind of thing for free. Chris Schilling (WMF) once suggested to me when I met him in London (and again in Italy) hat one could apply for an individual grant to do it, but that's another story (::sigh::) Kudpung กุดผึ้ง (talk) 05:04, 2 October 2016 (UTC)
- Well, I've already tried to explain the reality of the wish list, but anyway... I'm going to look at the Article Creation Workflow code and see what I can do with it. At the very least, we can create this notice right now, with very little effort. The only problem I foresee is the experienced users might complain about it, and also it probably won't look that great in VisualEditor. I say we give it a try, though. It's not super in-your-face, and hopefully experienced users would understand why it's being used. What do you think? — MusikAnimal talk 05:07, 2 October 2016 (UTC)
- Looks god but I'm being called to lunch. I'll get back to you.Kudpung กุดผึ้ง (talk) 05:32, 2 October 2016 (UTC)
- Something like User:MusikAnimal/New article edit notice. Needs a few styling tweaks, and you might want to do some copy editing. The "I want to create a draft" button currently goes to the article wizard, but it could go directly to a draftspace page with same name — MusikAnimal talk 06:37, 2 October 2016 (UTC)
- And when they search for an article and it doesn't exist: User:MusikAnimal/Article search with no result — MusikAnimal talk 07:09, 2 October 2016 (UTC)
- As for the Landing Page, during the run up to ACTRIAL I had started to completely rebuild the Article Wizard in php but Brandon said leave it to him and he came up with the Article Creation Work Flow (landing page project) which I think is pretty darned good - there was nothing 'half-baked' about it. AS I said above, Brandon is an excellent programmer, understood exactly what I was describing as needed, but he was taken off the project due to some internal squabble at he WMF - probably what led staff like Oliver Keys and Howie Fung to be so rude and demeaning to had-working volunteers during the Wikimania conference in DC in 2012. IMO it shouold simply be continued. Kaldari would tell you where the code is. However, I still think it's totally unreasonable to expect the volunteers to do this kind of thing for free. Chris Schilling (WMF) once suggested to me when I met him in London (and again in Italy) hat one could apply for an individual grant to do it, but that's another story (::sigh::) Kudpung กุดผึ้ง (talk) 05:04, 2 October 2016 (UTC)
arbitrary break
I've had a good look and a think about it, MusikAnimal. You know there's an essay at WP:EI? I'm not sure, but it might raise a stink. You know that even if it does not really bother them at all, Wikipedia users just love to complain about anything they can. I think it would be only justifiable if we were to introduce ACTRIAL, but then for that we already have a list of minor interface changes to make - and Brandon's project would have been a seamless fit. There is too much text on it. Would the buttons be functional? I'm working on something. In the meantime could you please email me, I want to send you a 17-page text file. Nothing secret but far to big to put on Wikiedia.Kudpung กุดผึ้ง (talk) 07:17, 2 October 2016 (UTC)
- Hehe OK, will email you. About the interface pages (at least the edit notice), this is totally WP:IAR by any stretch of the imagination, and we're not even breaking any rules. Those who complain will quickly get over it, add some personal CSS to hide it, or the most simple solution: just scroll down the page to the edit box so you can't see the mildly distracting edit notice (I just changed the background to white). We're otherwise talking about an extra 50px or so in height from the current edit notice, and potentially a profound effect on how new users approach article creation. It's this or spend an incredible and painful amount of time getting the Article Creation Flow extension to work... whether ACTRIAL happens or not, this will help and can be implemented now, and ACTRIAL will be a month away at best, which by your calculations will add another ~7,500 articles to the new pages backlog. Will respond to your next comment in a moment (got an edit conflict) — MusikAnimal talk 07:35, 2 October 2016 (UTC)
While you were doing that, MusikAnimal, I was having my lunch and doing this: File:New_page_edit_warning.png. One of the problems with Wikipedia is that all our alerts and templates are too verbose, however hard we try. This is of course only a temporary measure and what I think would get the least negative comments. Note that the button goes to the Wizard. If you are not familuar with the Wizard, you may wish to take a tour through it - you'll see what I mean about 'verbose'. In fact it's so verbose that some new users give up with it and go back to the editing window.Kudpung กุดผึ้ง (talk) 07:30, 2 October 2016 (UTC)
- I like it, and I totally agree about the verboseness. I have spent many an hour trying to trim down Help pages, which should be the most simple of all. Myself and another editor worked on WP:INTREF (and one for VisualEditor), which is enormously more helpful for newbies than Wikipedia:Citing sources, Wikipedia:Identifying reliable sources or even Help:Footnotes :) It's past my bedtime here in New York, but I'll work on simplifying the interface pages soon — MusikAnimal talk 07:40, 2 October 2016 (UTC)
- @Kudpung: Finally caught up on reading some of the discussions. It looks like you're not sold on using the Wishlist Survey, which is fine. I do, however, disagree with your characterization of the projects that came out of last year's survey as "gimmicks and gadgets". Replacing all the dead links on Wikipedia is pretty central to the quality of the project, and creating a good tool to detect plagiarism (https://tools.wmflabs.org/copypatrol) is also fairly important, IMO. Of course, everyone has different opinions on what the most important projects are, which is why we run a survey. I understand your frustration with the NPP situation, and I would love (probably more than anyone else) to have ArticleCreationWorkflow finished. However, the Engineering teams at the WMF can't just unilaterally drop what they are working on and switch to a different project. It has to go through some sort of existing channel for planning and approval, which basically either means coming from the top down or the bottom up (the Wishlist Survey). In this case, I would suggest pushing from both ends. I'm already working on getting a meeting with Katherine to discuss the current NPP situation (among other things) and if someone decides to propose some of the outstanding requests through the Wishlist Survey I think that would be a good back-up plan. Let me know if there's anything else I can do to help in the meantime. Kaldari (talk) 18:38, 3 October 2016 (UTC)
- Thanks for your input Ryan, it really is very much appreciated, and I do know that in spite of my characterisation of the WMF as a whole, you are personally concerned with the situation regarding the Landing Page. If I can get directly involved with Katherine to quickly outline the situation, I would be happy to do so. I did not want to bother her with it in Esino because her position had only been confirmed during the conference. She is probably not even aware of how these issues are damaging the reputation of Wikipedia on the one hand, and leaving it open for the clever corporate spammers on the other. What you could do in the meantime if possible, is look into unfinished Landing Page with MusikAnimal together and come up with some code - I'll be happy to help with the GUI and texts and Beta testing, as I did with Page Curation, but I can't write the sort of code that's required, and anyway, as you will have gathered, we all here feel that the paid staff should be doing most of the work. Kudpung กุดผึ้ง (talk) 19:24, 3 October 2016 (UTC)
- I've created two tasks out of this discussion: phab:T147224 which includes some easy updates to the PageTriage extension that you requested a while back I think, and phab:T147225 regarding the New Article Flow landing page. The latter would be done with interface pages, as a workaround until we can find the time to work on the actual extension. I saw your recommended changes to trim down the language and will work on that, run it by some other folks, then I think we should seek some quick consensus at another venue. It seems very uncontroversial, as I explained below. This would only help the situation until other improvements or ACTRIAL are implemented — MusikAnimal talk 19:34, 3 October 2016 (UTC)
- I'm currently taking a deeper look at the data: T146959. Kaldari (talk) 23:39, 3 October 2016 (UTC)
- I've created two tasks out of this discussion: phab:T147224 which includes some easy updates to the PageTriage extension that you requested a while back I think, and phab:T147225 regarding the New Article Flow landing page. The latter would be done with interface pages, as a workaround until we can find the time to work on the actual extension. I saw your recommended changes to trim down the language and will work on that, run it by some other folks, then I think we should seek some quick consensus at another venue. It seems very uncontroversial, as I explained below. This would only help the situation until other improvements or ACTRIAL are implemented — MusikAnimal talk 19:34, 3 October 2016 (UTC)
- Thanks for your input Ryan, it really is very much appreciated, and I do know that in spite of my characterisation of the WMF as a whole, you are personally concerned with the situation regarding the Landing Page. If I can get directly involved with Katherine to quickly outline the situation, I would be happy to do so. I did not want to bother her with it in Esino because her position had only been confirmed during the conference. She is probably not even aware of how these issues are damaging the reputation of Wikipedia on the one hand, and leaving it open for the clever corporate spammers on the other. What you could do in the meantime if possible, is look into unfinished Landing Page with MusikAnimal together and come up with some code - I'll be happy to help with the GUI and texts and Beta testing, as I did with Page Curation, but I can't write the sort of code that's required, and anyway, as you will have gathered, we all here feel that the paid staff should be doing most of the work. Kudpung กุดผึ้ง (talk) 19:24, 3 October 2016 (UTC)
- @Kudpung: Finally caught up on reading some of the discussions. It looks like you're not sold on using the Wishlist Survey, which is fine. I do, however, disagree with your characterization of the projects that came out of last year's survey as "gimmicks and gadgets". Replacing all the dead links on Wikipedia is pretty central to the quality of the project, and creating a good tool to detect plagiarism (https://tools.wmflabs.org/copypatrol) is also fairly important, IMO. Of course, everyone has different opinions on what the most important projects are, which is why we run a survey. I understand your frustration with the NPP situation, and I would love (probably more than anyone else) to have ArticleCreationWorkflow finished. However, the Engineering teams at the WMF can't just unilaterally drop what they are working on and switch to a different project. It has to go through some sort of existing channel for planning and approval, which basically either means coming from the top down or the bottom up (the Wishlist Survey). In this case, I would suggest pushing from both ends. I'm already working on getting a meeting with Katherine to discuss the current NPP situation (among other things) and if someone decides to propose some of the outstanding requests through the Wishlist Survey I think that would be a good back-up plan. Let me know if there's anything else I can do to help in the meantime. Kaldari (talk) 18:38, 3 October 2016 (UTC)
Some ACTRIAL code
As requested, a simple script to implement ACTRIAL:
Please see code here
|
---|
function block_creation() {
if (!(mw.config.values.wgIsArticle && mw.config.values.wgIsProbablyEditable && mw.config.values.wgArticleId === 0 && mw.config.values.wgUserEditCount < 10 && mw.config.values.wgUserRegistration < 345600000 && mw.config.values.wgNamespaceNumber === 0)) {
return;
}
$("#ca-ve-edit").remove();
$("#ca-edit").remove();
$(".mbox-text li:first").html("Use the <a href=\"https://en.wikipedia.org/wiki/Wikipedia:Article_wizard\">Article Wizard</a> if you wish,\
or <a href=\"https://en.wikipedia.org/wiki/Wikipedia:Requested_articles\">add a request</a> for it.");
}
function block_creation_at_form() {
if (!(mw.config.values.wgArticleId === 0 && mw.config.values.wgUserEditCount < 10 && mw.config.values.wgUserRegistration < 345600000 && mw.config.values.wgNamespaceNumber === 0)) {
return;
}
$("#editform").html("<div style=\"font-size:1.3em; font-weight:500; text-align:center\">Only registered, <a href=\"https://en.wikipedia.org\
/wiki/Wikipedia:User_access_levels#Autoconfirmed_and_confirmed_users\">autoconfirmed users</a> with 10 edits and 4 days of editing history\
can directly create articles. Please use the <a href=\"https://en.wikipedia.org/wiki/Wikipedia:Article_wizard\">Article Wizard</a> to create\
a draft article and receive feedback on your article, or register (if you have not already done so), make 10 edits, and wait 4 days.</div>");
}
block_creation();
block_creation_at_form();
|
I tested the speed of the two functions and they appear to only take 2-4 milliseconds. — Esquivalience (talk) 15:23, 2 October 2016 (UTC)
- I think this should be in wiki format so any admin can edit it. We currently have a CSS class to show things only to admins, and we could have another for users who don't meet this 96 hour / 10 edit threshold. The notice could be transcluded on the interface page for nonexistent articles, hidden by default, and shown as needed. The other JS we'd still need so we can hide the create buttons. Finally we'd need an edit filter to disallow edits for those who disabled JavaScript, used a direct link to create a new article, or even used the API — MusikAnimal talk 16:00, 2 October 2016 (UTC)
- Wouldn't it be easier to use the title blacklist? Something like
(?! list of namespaces we want newbies to be able to create pages in).* <autoconfirmed|errmsg=ACTRIAL-message>
(Sorry, not down with the regex.) This should prevent article creations and offers the ability to show a message just to non-autoconfirmed users. BethNaught (talk) 17:23, 2 October 2016 (UTC)- One needs to test this, I've got a funny feeling this will block all account creation as well. MER-C 05:31, 3 October 2016 (UTC)
- Yeah... we could try on testwiki I guess, or the beta cluster. I do think if this proves working it is the most ideal solution. No JavaScript required. I tested trying to create a page currently on the title blacklist, and it doesn't even let me try – which is what we want. E.g. I'd hate for the user to write their article then hit save only to find out they can't. This is yet another advantage over the edit filter — MusikAnimal talk 16:59, 3 October 2016 (UTC)
- @MusikAnimal: Re "doesn't even let me try": sadly that is desktop only at the moment—see phab:T145304. At least in some situations the mobile interface won't notice the TB until it's at the point of saving 😞 But this is still better an an edit filter, and community JS can't change the mobile site... BethNaught (talk) 20:06, 3 October 2016 (UTC)
- Yeah... we could try on testwiki I guess, or the beta cluster. I do think if this proves working it is the most ideal solution. No JavaScript required. I tested trying to create a page currently on the title blacklist, and it doesn't even let me try – which is what we want. E.g. I'd hate for the user to write their article then hit save only to find out they can't. This is yet another advantage over the edit filter — MusikAnimal talk 16:59, 3 October 2016 (UTC)
- One needs to test this, I've got a funny feeling this will block all account creation as well. MER-C 05:31, 3 October 2016 (UTC)
- what the customer sees should read 4 days, rather than 96 hours. It's psychological - 4 does not sound as menacing as 96. Kudpung กุดผึ้ง (talk) 23:52, 2 October 2016 (UTC)
- Done — Esquivalience (talk) 00:24, 3 October 2016 (UTC)
- Just a reminder...
ACTRIAL wasn't only about restricting the creation of new articles in mainspace to autoconfirmed users. It was not our intention to hinder the spontaneity of writing an article, so our proposal for WP:ACTRIAL also came with a complete package of template messages and interface mods to offer them the possibility ('force' them) to use the Article Wizard. Kudpung กุดผึ้ง (talk) 20:21, 3 October 2016 (UTC)
Complete list of everything that is needed to conclude this project
I've sent MusikAnimal a copy of the 17-page report I was asked for by the WMF two months ago which still hasn't even been acknowledged. I've left a long(ish) message at m::User talk:DannyH (WMF) who replied on my talk page at Meta to my post on the pre-survey letter to Santa Claus, but obviously I'm not holding my breath. They won't even start their survey until November, and then it will take a year before they start working on anything. MusikAnimal is of course free to use or publish all or any parts of my report - there's nothing secret about it. What I do resent however, are the occasional snide remarks from some users that work groups operate as secret cabals.
Also, (@Bluerasberry:), in replying on Danny's talk page I cast an eye over the other messages he has received and frankly I am now even less reassured that our community can have much confidence his project for addressing major critical issues. Perhaps nevertheless he might just take the initiative on our behalf to inform whoever has overall responsibility for such things, but again, I'm not holding my breath there either - the atmosphere within the WMF is that most employees won't take any initiative outside their brief for fear of compromising their paid positions. --Kudpung กุดผึ้ง (talk) 02:16, 3 October 2016 (UTC)
I think I have finally discovered the confirmation I need of the claims I make about the lack of cooperation between WMF staff who are even supposed to be working on the same team. Ajraddatz has deleted the page we were advised to register our requirement on. Danny almost never responds to messages on his talk page. None of this instills confidence and I see the Steven Walling syndrome looming. I think that in spite of his kind help here, we need to ask MusikAnimal to clarify his position. Kudpung กุดผึ้ง (talk) 05:23, 3 October 2016 (UTC)
- I wouldn't be that pessimistic -- all that's needed is to approach the right person at the right time with a persuasive argument (and be prepared to wait). Again, Orangemoody was a lost opportunity. To wit: I bypassed the Community Tech wishlist regarding phab:5233 at m:Harassment consultation 2015; it's now being implemented. MER-C 05:41, 3 October 2016 (UTC)
- @Kudpung: I work for meta:Community Tech. I saw the comments regarding the wish list and wanted to offer some reassurance that it is a real thing that we as a team honour and go by – on my own accord. That's the extent of it... anything else you'd like to know? The landing page implementation I've done (that I think we should move forward with), is completely volunteer, as is anything under this account — MusikAnimal talk 16:45, 3 October 2016 (UTC)
- @MusikAnimal: Is Community Tech willing or planning to accept direct proposals? Voting seems to prioritize what editors want as a very late Christmas present, such as new nifty gadgets or convenience features (such as dead link fixing), while other important software get ignored by the vote. Perhaps the team should get readers involved into the process as well: maybe they want to see better math (equations such as look atrocious on many devices). — Esquivalience (talk) 00:04, 4 October 2016 (UTC)
- @Kudpung: I work for meta:Community Tech. I saw the comments regarding the wish list and wanted to offer some reassurance that it is a real thing that we as a team honour and go by – on my own accord. That's the extent of it... anything else you'd like to know? The landing page implementation I've done (that I think we should move forward with), is completely volunteer, as is anything under this account — MusikAnimal talk 16:45, 3 October 2016 (UTC)
- Esquivalience: Really, the best way to get the Community Tech team to work on something is to propose it in the Community Wishlist Survey. If the project that you're proposing is really important and affects a lot of people, then it should be relatively easy to get a lot of other people to support-vote it to the top of the list. I think saying "I won't bother to submit this important idea for the survey, because the survey never includes any important ideas" is potentially self-defeating.
- We've only done the survey once so far, so I don't know if we can draw any conclusions about what people tend to vote for. What do you think about using the next survey to find out if we could get people to vote for important software? -- DannyH (WMF) (talk) 00:29, 4 October 2016 (UTC)
- @DannyH (WMF): If you cannot see the importance of new page patrol, why should a bunch of random people hanging around meta find it interesting? Is having 15,000 unreviewed new articles at enwiki a problem? Or is it someone else's problem? Johnuniq (talk) 01:02, 4 October 2016 (UTC)
- To further this, my experience putting together ACTRIAL was that editors from places other than en.wiki blithely dismissed the idea because of some pre-conceived notion that it was unnecessary because no one else did it. To a large extent there was complete misunderstanding (which in some cases appeared willful) of the serious problem en.wiki faces, being that it's so much larger and faster moving than anything else, and entailed a series or disingenuous comments about "the community's will", because some people who never actually edit en.wiki are entitled to shut down our consensus while bloviating about their superior understanding of the "real meaning" of the founding principles. Or something. Opening this up to the same degree will get identical results, and result in a lot of wasted time and bytes of text. The Blade of the Northern Lights (話して下さい) 04:09, 4 October 2016 (UTC)
- Okay, it's up to you. You're welcome to participate, if you want. -- DannyH (WMF) (talk) 16:34, 4 October 2016 (UTC)
- I don't underestimate the value of your project Danny, but as we've mentioned now multiple times here and elsewhere, it's not a venue where highly critical and fundamental issues of this magnitude should be listed and their priority decided by the managers of such a wish list. Wish lists imply neither necessity nor importance nor urgency.
- The wish list project was almost certainly never intended to replace or block discussion at higher levels on the need to address major issues, so I am seeking a solution to clearly identify the line of responsibility within the WMF for such requirements. That said, Ryan and MusikAnimal are now working hard to address many of the points on our program and we are now feeling some sense of progress for the first time. With any luck, we'll be able to get this accomplished without the need to publish a moan an groan, gloom and doom article on Signpost.Kudpung กุดผึ้ง (talk) 00:04, 5 October 2016 (UTC)
- Oh, it's absolutely not a blocker; you can raise the issue any way that you like. I'm just saying that there is currently a way to get a share of WMF development resources for projects that contributors think are important. If you don't want to use that opportunity, then you don't have to. But as I said, you're welcome to, if you decide you want to. -- DannyH (WMF) (talk) 00:54, 5 October 2016 (UTC)
- Danny, we will, of course but the scope of the immediate requirements for NPP (few have seen the extent of the actual report) would take up your team's resources for a whole year even if they started now. We wouldn't want to hold up the many requests for convenience tweaks that you are already admirably working through, so we also need to find a way to get some of the rest of the engineers allocated to these serious fundamental issues, or to increase the size of your team. Funds are not an issue - or so I have been told. Kudpung กุดผึ้ง (talk) 02:14, 5 October 2016 (UTC)
- Oh, it's absolutely not a blocker; you can raise the issue any way that you like. I'm just saying that there is currently a way to get a share of WMF development resources for projects that contributors think are important. If you don't want to use that opportunity, then you don't have to. But as I said, you're welcome to, if you decide you want to. -- DannyH (WMF) (talk) 00:54, 5 October 2016 (UTC)
- Okay, it's up to you. You're welcome to participate, if you want. -- DannyH (WMF) (talk) 16:34, 4 October 2016 (UTC)
- To further this, my experience putting together ACTRIAL was that editors from places other than en.wiki blithely dismissed the idea because of some pre-conceived notion that it was unnecessary because no one else did it. To a large extent there was complete misunderstanding (which in some cases appeared willful) of the serious problem en.wiki faces, being that it's so much larger and faster moving than anything else, and entailed a series or disingenuous comments about "the community's will", because some people who never actually edit en.wiki are entitled to shut down our consensus while bloviating about their superior understanding of the "real meaning" of the founding principles. Or something. Opening this up to the same degree will get identical results, and result in a lot of wasted time and bytes of text. The Blade of the Northern Lights (話して下さい) 04:09, 4 October 2016 (UTC)
- @DannyH (WMF): If you cannot see the importance of new page patrol, why should a bunch of random people hanging around meta find it interesting? Is having 15,000 unreviewed new articles at enwiki a problem? Or is it someone else's problem? Johnuniq (talk) 01:02, 4 October 2016 (UTC)
Taking a closer look at the data
![](http://upload.wikimedia.org/wikipedia/commons/thumb/3/3d/Number_of_articles_reviewed_per_day_through_Page_Curation.png/350px-Number_of_articles_reviewed_per_day_through_Page_Curation.png)
![](http://upload.wikimedia.org/wikipedia/commons/thumb/f/fc/Number_of_articles_reviewed_per_day_by_SwisterTwister.png/350px-Number_of_articles_reviewed_per_day_by_SwisterTwister.png)
I took a look at the data we have for Page Curation, Page Patrolling, and new article creation. Over the past couple years the rates for page patrolling (via the 'mark as patrolled' link) and new article creation have remained relatively flat. The rate for reviewing articles through Page Curation (Special:NewPagesFeed), however, dropped suddenly on May 20, 2016 and has not recovered. I reviewed the commit logs for both the PageTriage extension and core MediaWiki in the week before May 20. The only changes to PageTriage were localization updates and I was not able to find any suspicious core changes during that time (although due to the large number of core commits, I was not able to review them in depth). I will now start looking at the Page Curation logging data in more detail to see if I can figure out what changed then. Any leads would be appreciated. Kaldari (talk) 01:30, 4 October 2016 (UTC)
- @Kudpung, Esquivalience, and BethNaught: ^ Kaldari (talk) 01:34, 4 October 2016 (UTC)
- Kaldari, I've made some suggestions/leads direct on the Phabricator case page. Kudpung กุดผึ้ง (talk) 06:42, 4 October 2016 (UTC)
- I think this has to do with SwisterTwister dramatically decreasing patrolling around that time. Prior to May 20, that user was making over 500 patrols per day. However, between May 20 and June 17 they made fewer than 500 total. See here. The graph shows a peak towards late July, and that user was more active in patrolling at the same time as the peak.~ ONUnicorn(Talk|Contribs)problem solving 15:17, 4 October 2016 (UTC)
- That would account for what looks like a drop of 400-500 on that graph. Has anybody asked SwisterTwister why they stopped patrolling as much? -- DannyH (WMF) (talk) 17:13, 4 October 2016 (UTC)
- Probably a result of this discussion, even though it was closed as no consensus (I had no involvement and make no specific comment on it). The Blade of the Northern Lights (話して下さい) 17:43, 4 October 2016 (UTC)
- I agree it was a result of that discussion, but generally it is not good when a vital process largely relies on one user.--Ymblanter (talk) 20:29, 4 October 2016 (UTC)
- As it was on me circa early 2011, when I originally came up with what became ACTRIAL. It's bad for the user and for the process. The Blade of the Northern Lights (話して下さい) 21:09, 4 October 2016 (UTC)
- I agree it was a result of that discussion, but generally it is not good when a vital process largely relies on one user.--Ymblanter (talk) 20:29, 4 October 2016 (UTC)
- Probably a result of this discussion, even though it was closed as no consensus (I had no involvement and make no specific comment on it). The Blade of the Northern Lights (話して下さい) 17:43, 4 October 2016 (UTC)
- That would account for what looks like a drop of 400-500 on that graph. Has anybody asked SwisterTwister why they stopped patrolling as much? -- DannyH (WMF) (talk) 17:13, 4 October 2016 (UTC)
- I think this has to do with SwisterTwister dramatically decreasing patrolling around that time. Prior to May 20, that user was making over 500 patrols per day. However, between May 20 and June 17 they made fewer than 500 total. See here. The graph shows a peak towards late July, and that user was more active in patrolling at the same time as the peak.~ ONUnicorn(Talk|Contribs)problem solving 15:17, 4 October 2016 (UTC)
- Kaldari, I've made some suggestions/leads direct on the Phabricator case page. Kudpung กุดผึ้ง (talk) 06:42, 4 October 2016 (UTC)
- Yep, looks like SwisterTwister was doing the majority of reviews (see 2nd chart). That would explain the backlog piling up since June. Kaldari (talk) 23:00, 4 October 2016 (UTC)
- It is of course inadmissible, Kaldari, that such an important activity such as NPP, our only firewall against unwanted content, should rely on the work of a mere handful of operators. Some of them have stopped or reduced their participation at NPP because they were told that their patrolling was substandard and causing more problems than if they were to leave the process alone (SwisterTwister).
- Those who have read (should have) the other discussions I linked to, will have noted also that three reasons for disinterest in NPP are: 1) it is a boring, lonely job, 2) there is no vibrant group of 'members' supporting it in the same way of AfC (which has almost even become a social networking project), and 3). there is no headware for the hat collectors - which works as a magical magnet to maintenance tasks.
- I've calculated that we would need ten new patrollers working away at the backlog for 96 days just to get it down to a daily input - without taking into consideration the 850 new pages that are received every day. Kudpung กุดผึ้ง (talk) 23:23, 4 October 2016 (UTC)
- The WMF commissioned an external research a few years ago into NPP. It came up with all the wrong results, but I believe it was established that about 50 patrols, properly done, are about the maximum a patroller can do in one session. I manage about 20-30. Kudpung กุดผึ้ง (talk) 02:22, 5 October 2016 (UTC)
- I can hardly do more than 10 (not including Rfded redirects) without compromising on my other activities at Wikipedia and sister projects.--Ymblanter (talk) 05:56, 5 October 2016 (UTC)
Idea for technically preventing "blank" AFC submissions
Is it possible to make the AFC "Submit" button available only when the page has reached a specified minimum size? The boilerplate content that is automatically placed on a new draft page (an AFC submit template and a References section with a hidden explanatory note) runs to about 300 bytes, depending on the page title and creator's username. The Submit button could be inactivated until the page is at least 400 bytes long. I believe it's impossible to create an acceptable stub with fewer than 100 characters of actual text. (The previous sentence is 101 characters) Roger (Dodger67) (talk) 17:23, 8 October 2016 (UTC)
- Roger, It's possible to get a script made that if an editor attempts to submit a very short article, it automatically puts an alert:
- Sorry, your article is still too short to have sufficient value as an encyclopedia article. Please expand it, being sure to source the new content, then submit again. Thanks,
- how severe is the incidence of such pages as compared to the arrival of new submissions as a whole? --Kudpung กุดผึ้ง (talk) 04:45, 9 October 2016 (UTC)
- Kudpung "Blank" is consistently in the "top four" decline reasons I use. The AFCH script offers reviewers an automatically sorted "shortlist" of the four most used decline reasons - "blank" has never dropped out of my shortlist. At a rough thumbsuck I'd say they make up at least 10% of all new (GFOO) submissions. Roger (Dodger67) (talk) 06:32, 9 October 2016 (UTC)
- Roger, be brutal. Blank pages are from time wasters. Tag them for CSD A1 or A3 or if you are an admin, delete them summarily. Check with your project that a decline template exists that states: 'Declined and tagged for deletion because your submission had little or no content'. Kudpung กุดผึ้ง (talk) 06:44, 9 October 2016 (UTC)
- The A# speedy criteria are not applicable to Draft or User spaces. I already have a problem with admins refusing blatantly obvious G11 speedy nominations because "it's a draft it can be fixed". However, speedying blank submissions doesn't solve the "time sink" problem of them being submitted in the first place adding pointless "overburden" to the reviewing queue and burning up more reviewer time. As far as possible preventing obvious crap from even entering the AFC queue should be a priority. Take a good look at the "quick fail" criteria in the AFC workflow, trying to automate them as far as feasible would free up reviewers to consider the ones that need actual thought. Of these, the blanks seem to be the lowest hanging fruit, easy to prevent their submission. Roger (Dodger67) (talk) 11:31, 9 October 2016 (UTC)
- Unfortunately, I agree with Roger. It is rare for an AFC submission to qualify for speedy deletion. The A criteria are, as their name implies, for article space. I don't have blank in my top five, but it isn't rare, and it is a timesink. I won't waste my time and that of an admin by tagging blank submissions with A1 or A3, but it would help if automated code would stop the blank or nearly blank submissions. Also, I see blank submissions as good-faith error, at least at AFC, by editors who really don't know what they are doing. They are more good-faith than spam, and a filter would be useful. Robert McClenon (talk) 20:25, 9 October 2016 (UTC)
- The A# speedy criteria are not applicable to Draft or User spaces. I already have a problem with admins refusing blatantly obvious G11 speedy nominations because "it's a draft it can be fixed". However, speedying blank submissions doesn't solve the "time sink" problem of them being submitted in the first place adding pointless "overburden" to the reviewing queue and burning up more reviewer time. As far as possible preventing obvious crap from even entering the AFC queue should be a priority. Take a good look at the "quick fail" criteria in the AFC workflow, trying to automate them as far as feasible would free up reviewers to consider the ones that need actual thought. Of these, the blanks seem to be the lowest hanging fruit, easy to prevent their submission. Roger (Dodger67) (talk) 11:31, 9 October 2016 (UTC)
- Roger, be brutal. Blank pages are from time wasters. Tag them for CSD A1 or A3 or if you are an admin, delete them summarily. Check with your project that a decline template exists that states: 'Declined and tagged for deletion because your submission had little or no content'. Kudpung กุดผึ้ง (talk) 06:44, 9 October 2016 (UTC)
- Kudpung "Blank" is consistently in the "top four" decline reasons I use. The AFCH script offers reviewers an automatically sorted "shortlist" of the four most used decline reasons - "blank" has never dropped out of my shortlist. At a rough thumbsuck I'd say they make up at least 10% of all new (GFOO) submissions. Roger (Dodger67) (talk) 06:32, 9 October 2016 (UTC)
Maybe a bot that monitored new submissions and auto-declined blank ones (and perhaps other types of obviously malformed submissions) with a polite explanation would be more effective than a warning on submission? I have a feeling that if someone is at the point where they're submitting a blank draft to AfC they've already seen and ignored a hell of a lot of instructions. Joe Roe (talk) 01:07, 10 October 2016 (UTC)
- It is in fact ridiculously easy for a completely new user to submit a blank in three easy clicks: 1. Click on the redlink "Sandbox" link in the user menu at the top of the screen. 2. Accept the "offer" to create the page that pops up. 3. Notice the "Submit" button that is prominently displayed on this brand new shiny clean page and click it (just to see what happens). Congratulations newbie editor, you have just added another blank submission to the AFC queue - a bitey canned message (which you will not understand) will be delivered to your Talk page soon. We should really take a good hard look at how the Article Wizard functions, it misleads editors into thinking AFC is the only way to contribute new additions and makes it too easy to (intentionally or unintentionally) add crap to the AFC "machine". Roger (Dodger67) (talk) 09:32, 12 October 2016 (UTC)
- The only thing that's basically wrong with the Article Wizard is that like everything else Wikipedia, it's too wordy. We're soon going to make it harder for IPs, spammers, and troll to create pages, but we need to make it easier for good faith editors. The Foundation isn't going to do anything about Article Wizard or AfC though because they are local projects. That said, some years ago I started to redesign Article Wizard, but when the WMF announced they were working on the development of a new landing page that would channel users to it, I held off, I'm still holding off 5 years later waiting for the new Landing Page from the WMF.
- It should be fairly easy to build a filter to do whatever you want with blank pages. Perhaps not so easy though to find someone to do it. Ideally, one could start an RfC at WT:CSD for a new CSD criterion to cover it, but with only around 100 or so articles arriving at AfC every day, 10%, i.e. 10 pages, would be arguably manually processable. Kudpung กุดผึ้ง (talk) 10:46, 12 October 2016 (UTC)
- I see tons of blank AfC pages in userspace anything to reduce the creation would be very helpful ~~,~
Blank submissions continue to be a problem, as do very short ones. There is no way to accept something less than 4 sentences, and I don't think your average new user even realizes they are submitting a blank or very short effort for someone else to review. I bet a lot of people think Submit = Save. There are over 300 blank declined submissions in User/Sandboxes from the last 6 months. I already CSD G2 all the Blank Draft space ones but sandboxes are G2 exempt. Sometimes I just blank the sandbox with edit message "blank test and AfC submission" but prevention of creatition would be a lot friendlier than getting Decline, then 6 months later Delete messages. Legacypac (talk)
New Pages Feed - access
MusikAnimal. Now that we have the consensus to restrict access to the functions of NPP, could you or one of your colleagues please look at this script User:Ryan Vesey/sidebar.js, and make it only accessible to holders of the new user right? The author Ryan Vesey. Ryan went suddenly AWOL a long time ago and while we sincerely hope he is well, we have no news and it doesn't look as if he's going to be back any time soon. --Kudpung กุดผึ้ง (talk) 04:54, 9 October 2016 (UTC)
- @Kudpung: There are but 10 people who have that script in their personal JS, and apparently you, two sysops and Sam Sailor are the only ones using it. Altering it isn't going to help anything, I believe, and all the script does is add a link to Special:NewPagesFeed in the sidebar. Most people simply go directly to Special:NewPagesFeed, perhaps have it bookmarked. We should wait for the new right to be deployed, and we'll update the Page Curation code itself — MusikAnimal talk 23:09, 9 October 2016 (UTC)
- Hmm wait a minute, there isn't a new user right, correct? We're just removing the "patrol" permission from autoconfirmed, confirmed, and reviewer groups, and moving it to a dedicated user group. It appears Page Curation is already looking for this permission, so no changes are actually necessary — MusikAnimal talk 23:15, 9 October 2016 (UTC)
- MusikAnimal, I think xaosflux in his brave attempts to be helpful that started around User talk:Kudpung/Archive Aug 2016#WP:NPP may have caused some confusion. It certainly confused me. Most people consider 'Reviewer', 'Rollbacker', 'Autopatrolled', and 'Page mover' (have I missed any?) as 'user rights'. Indeed the Special:UserRights/Kudpung page where the are controlled is even called User rights management. Therefore, for us ignorant hard working volunteers the distinction between User Right and User Group not only escapes us, but is not even strictly necessary for us to comprehend. All we want is for a user right/group that prevents unqualified users from accessing the New Pages Feed and the Curation tool, and for whichthe permission requests will be handled through the page at Special:UserRights. Kudpung กุดผึ้ง (talk) 21:12, 10 October 2016 (UTC)
- @Kudpung: Yep, so let me explain: There are two things, user groups and rights. Each user group has a set of rights associated with it, which are listed at Special:UserGroupRights. What you see at Special:UserRights/Kudpung are user groups. The thing that makes patrolling pages possible is a right called patrol, which currently is in the "(auto)confirmed", "pending changes reviewer" and "administrator" groups. So all we need to do is create a new user group, "new page patroller" that has the right patrol, then remove patrol from "(auto)confirmed" and "pending changes reviewer". That leaves us with two user groups that can patrol pages: "new page patroller" and "administrators". This is all done with a simple config change, no code changes to Page Curation are necessary — MusikAnimal talk 21:20, 10 October 2016 (UTC)
- Thanks MusikAnimal. As long as it does what we need I don't think most of the editors driving this NPP reform projet are particularly concerned with the engineering. I'm just surprised (as a very adept administrator of highly granular groups/rights in some other web sites) that people make it all sound so complicated. All it achieves of course is putting logs through the spokes of RfCs and some 'oppose' votes at RfC fthen come from people who are made skeptical and who would have supported if the RfCs were kept on the KISS principle. There is this ridiculous notion held by some people that every Wikipedia editor has to be an IT geek. Kudpung กุดผึ้ง (talk) 21:33, 10 October 2016 (UTC)
- I've updated the title for Special:UserRights to match all of the other recent updates (notice the side bar has changed to "change user groups" recently as well. The terminology can indeed be confusing. — xaosflux Talk 21:37, 10 October 2016 (UTC)
- Nice! If only we could get the proper name Special:UserGroups in place of UserRights. It's true it is confusing... and the core software even conflicts with itself — MusikAnimal talk 21:40, 10 October 2016 (UTC)
- Are you sure you want to remove the patrolling right from Pending changes reviewers? That's a manually assigned user group and seems to align pretty closely with the qualifications you would want from a patroller. I would suggest only removing the right from the (auto)confirmed user group (if any). Otherwise, you're going to start with a very small pool of potential patrollers. Kaldari (talk) 05:01, 12 October 2016 (UTC)
- @Kaldari:Yes, quite sure. PC Reviewer has the lowest threshold of entry of all the Wikipedia user rights - and is supposed to be relatively easy to obtain. It requires (theorectically) so little clue that it was accorded to thousands of users by a bot. Some users have actually opposed our RfC because they feel 90 days/500 mainspace edits is not severe enough. New Page Patroller (or to give it its new name, New Page Reviewer) actually needs a near admin level knowledge of notability and deletion criteria.
- Are you sure you want to remove the patrolling right from Pending changes reviewers? That's a manually assigned user group and seems to align pretty closely with the qualifications you would want from a patroller. I would suggest only removing the right from the (auto)confirmed user group (if any). Otherwise, you're going to start with a very small pool of potential patrollers. Kaldari (talk) 05:01, 12 October 2016 (UTC)
- Nice! If only we could get the proper name Special:UserGroups in place of UserRights. It's true it is confusing... and the core software even conflicts with itself — MusikAnimal talk 21:40, 10 October 2016 (UTC)
- @Kudpung: Yep, so let me explain: There are two things, user groups and rights. Each user group has a set of rights associated with it, which are listed at Special:UserGroupRights. What you see at Special:UserRights/Kudpung are user groups. The thing that makes patrolling pages possible is a right called patrol, which currently is in the "(auto)confirmed", "pending changes reviewer" and "administrator" groups. So all we need to do is create a new user group, "new page patroller" that has the right patrol, then remove patrol from "(auto)confirmed" and "pending changes reviewer". That leaves us with two user groups that can patrol pages: "new page patroller" and "administrators". This is all done with a simple config change, no code changes to Page Curation are necessary — MusikAnimal talk 21:20, 10 October 2016 (UTC)
- MusikAnimal, I think xaosflux in his brave attempts to be helpful that started around User talk:Kudpung/Archive Aug 2016#WP:NPP may have caused some confusion. It certainly confused me. Most people consider 'Reviewer', 'Rollbacker', 'Autopatrolled', and 'Page mover' (have I missed any?) as 'user rights'. Indeed the Special:UserRights/Kudpung page where the are controlled is even called User rights management. Therefore, for us ignorant hard working volunteers the distinction between User Right and User Group not only escapes us, but is not even strictly necessary for us to comprehend. All we want is for a user right/group that prevents unqualified users from accessing the New Pages Feed and the Curation tool, and for whichthe permission requests will be handled through the page at Special:UserRights. Kudpung กุดผึ้ง (talk) 21:12, 10 October 2016 (UTC)
- Hmm wait a minute, there isn't a new user right, correct? We're just removing the "patrol" permission from autoconfirmed, confirmed, and reviewer groups, and moving it to a dedicated user group. It appears Page Curation is already looking for this permission, so no changes are actually necessary — MusikAnimal talk 23:15, 9 October 2016 (UTC)
- NPP has already dwindled to very ssmall pool of active New Page Reviewers - that's why we have the backlog. The person who caused the backlog to drop has serious problems with their patrolling (see multiple ANI cases) and the backlog will now rise again. As with other user rights which are a magnet to the hat collectors (many of them do a really good job though), we are trading on the likelihood that plenty of qualified users will flock to WP:PERM for the hat, and that some former reviewers will return and ask for the bit. The only user group that subsumes New Page Reviewer is Administrator/Sysop (or higher). Kudpung กุดผึ้ง (talk) 09:27, 12 October 2016 (UTC)
- @Kudpung: Can you give me a link to the RfC about restricting the patrol right so that I can file a Phabricator ticket for the change? Kaldari (talk) 17:47, 13 October 2016 (UTC)
- Nevermind, I found it. Should these changes be implemented immediately, or should it wait until the qualifications RFC is finished? Kaldari (talk) 17:50, 13 October 2016 (UTC)
- @Kaldari:, these changes must wait until the RfC has been formally closed. This is Wikipedia policy. en.Wiki RfCs usually run for at least 30 days but can be closed earlier if there is an overwhelming majority one way or another. Also, in anticipation of the result as per the proposal, we are still working on texts for some template messages, a newsletter, some minor interfaace changes which any admin can do, some Lab tools which MusikAnimal and other volunteers are developing, and some additions to the Mew Pages Feed and the Page Curation tool bar that can only be done for us by your team.Kudpung กุดผึ้ง (talk) 00:22, 14 October 2016 (UTC)
- @Kaldari:, I'll just add however, that consensus is already strong, so these fixes all need to be ready to deploy when the consensus is granted. Kudpung กุดผึ้ง (talk) 12:57, 14 October 2016 (UTC)