Wikipedia talk:Wikipedia Signpost/2014-09-03/Recent research

Discuss this story

Thanks for posting the article about the AfC. I have felt for years now that the AfC project is "a really bad thing", but have been unable to express why exactly in terms that will make any difference. No offense to the people who have devoted so much energy to the cause, but after talking about this with User:Kudpung at Wikimania in London, I only feel more certain that this project should be shut down completely. Jane (talk) 13:38, 6 September 2014 (UTC)[reply]

AfC has extreme problems, like the Signpost author I find the results on timeliness sufficiently surprising to be unbelievable without independent scrutiny. Articles that would be (or would have been) perfectly acceptable in main-space, may languish at AfC until they get g13'd. However the purpose of AfC is to protect newbies from new page patrollers (broadly construed) rather than the other way around. Many of the posts we get at Teahouse relate to AfC issues, and I certainly feel that much of the effort that goes to assist these editors will eventaully be g13'd along with the article. All the best: Rich Farmbrough, 18:22, 6 September 2014 (UTC).

Although a growing number of established editors are suggesting AfC be done away with, I would just like to avoid any misuerstanding that while I am deeply concerned about AfC and have discussed it with many people my solution would not to be to close AfC down. The system is in a mess for all the issues mentioned in the report although as per Rich the report got a couple of things slightly wrong. Closing AfC down would lump all new registered users creations onto WP:NPP which in spite of having an excellent, professionally designed suite of software to work with, is in an even bigger mess. The only advantage of closing AfC down completely as far as I can see, would be that it would put a stop once and for all of creations by IPs. Something was begun by the Foundation that would have been at the same time an excellent replacement for the Article Wizard and a proper landing place for would-be new creators of articles. Why we don't have either this yet, and/or a professinally engineered replacement for AfC are discussed in recent threads at WT:AFC. KudpungMobile (talk) 06:41, 7 September 2014 (UTC)[reply]

Not really. IIRC, the research shows that on non-English Wikipedias the unregistered users create better articles on average than registered (newbie) users, and that on en.wiki articles created directly in ns0 have better survival rates. So the solution looks rather simple: enable unregistered users article creation to improve the average contribution; stop encouraging users to use AfC, to improve average editor experience. --Nemo 09:24, 7 September 2014 (UTC)[reply]

This image shows some of the difficulties, the trend was downwards before AfC, and although there was a step change downwards, it is not clear that there is long term damage or benefit. Conversely the figure is per creator and we might consider that if AfC, with all its careful warnings and guidance, and the investment of significant volunteer time doesn't produce something substantially better than "Create page" then something is wrong. All the best: Rich Farmbrough, 13:40, 7 September 2014 (UTC).

There are two problems with the essay, which were well-discussed earlier at AfC: success and editor experience. What is success? That some unreferenced POV nonsense created by a newbie survives past 30 days? There's a false metric in the essay that any content that survives 30 days is a positive contribution. With a decreasing user base it takes longer to identify problematic content that isn't obvious vandalism and some hoaxes have persisted for years. What is editor experience? New users or IPs are able to contribute without being bitten? If we need to address their editing then their comfort has to take a back-seat to the project.

In my role as a campus ambassador I tell my students to avoid AfC because it isn't meant for registered users and it will delay their move into the main namespace. However, I'm watching them like a hawk to ensure their edits don't sour the community on WEF's outreach. Other newbies often don't have a Wikipedian sponsor and AfC (and maybe the Teahouse) is the only outlets formally charged with providing help with article development in contrast to what NPP and AfD do.

In my mind, AfC is a bug zapper that catches the editors unable to otherwise navigate the system. The bug-zapper works as well as it can stopping problematic editors from harming Wikipedia. I fully support Kudpung's suggestion to disallow IPs from creating articles but disestablishing AfC will simply move the logjam to a different set of editors. Chris Troutman (talk) 17:15, 7 September 2014 (UTC)[reply]

I'm one of the editors who would like to remove the afc system entirely. In addition to all of the problems mentioned: it encourages editors to resubmit unacceptable drafts indefinitely--G13 will not remove anything unless it's totally inactive (this is much less a problem with NPP); it does not catch duplication of existing articles right at the beginning (this is not a problem with NPP); it mentions the possibility of merging, but neither facilitates it nor exposes it to editors who might do it, such as editors of the article to be merged into (this is not a problem with NPP merge tags); it hides incompetent reviews and reviewers (this is much less a problem with NPP); successive reviews tend to be inconsistent and create confusion (this is much less of a problem in NPP)it provides no hint about what an article is about to permit screening (NPP is particularly good here); it does not permit referral to Wikiprojects (NPP does). In one key area both it and NPP have equal problems: they both encourage the use of canned reviews which do not address specific problems).

There are three possible reasons why we need some system of the sort: the one mentioned, of preventing inappropriate articles from getting in main space (but this has no effect on the articles not going through AfC; it might be better addressed by applying NOINDEX to all articles from new contributors for the first hour or so); we need to provide a space for people to work on unfinished articles (the draft space is indeed probably better than user subpages, but the AfC procedure doesn't help the articles get improved); it provides a way for COI and even paid editors to legitimately submit articles (we would need an alternative here).

It will at the very least be easier to build up NPP than to fix AfC, and I am not convinced those currently operating AfC are even willing to fix it to the radical extent that would be necessary. But the key problem affecting both is the lack of participation of experienced editors, if even one hundred experienced editors who understood the need for personal non-templated advice looked at only two or three articles a day, it would solve the problem. The effect will be to decrease work overall--the more problems we can solve at the start, the fewer will need to be dealt with later. DGG ( talk ) 00:00, 8 September 2014 (UTC)[reply]

DGG G13 is only the procedural method for cleaning out drafts that the "advocate" for the draft has lost interest in. If a advocate resubmits patently inappropriate content that multiple reviewers agree has no chance at being accepted (and surviving in mainspace) then a MFD should be started explaining why the draft has no hope of being accepted. A consensus of Wikipedia volunteers will adjudge the draft and if appropriate a admin will enact whatever consensus is established. From my experience the editors who repeatedly resubmit after being told no is 1% of 1% of the overall G13 pool. Shouldn't we be assuming good faith upon the activities of editors who remember that they had a draft that needed to be worked on? Hasteur (talk) 12:33, 9 September 2014 (UTC)[reply]

Hasteur, I'm wary of just how far our use of the expression Good Faith should be extended to the hundreds of drive-by SPA who drop off a new 'article' at AfC (or in mainspace} and be interpreted to mean that the patrollers/reviwers should be expected to make respectful articles out of them; frankly, we just don't have the calories. I'm all in favour of G13 and although I'm not a deletionist, I still think we pander too much to the possibility that many G13 have something salvageable. AfC is neither the RA response team nor the ARS task force (send them there if you like, remembering however, that we have deprecated the incubator), but sending them to MfD would defeat the object of reducing bureaucracy and backlogs at XfD. In searching for solutions for AfC we must avoid constantly proposing ones that merely shove the problems onto someone else's lap.KudpungMobile (talk) 22:43, 9 September 2014 (UTC)[reply]

So Chris troutman, you perform a one man "Review Board" by encouraging your charges to not submit to AfC where there are multiple volunteers? Did you consider that perhaps one of the AFC volunteers might have more experience with the content and therefore could provide a more accurate review so that the advocate for the article knows exactly what challenges they are going to face? Hasteur (talk) 12:39, 9 September 2014 (UTC)[reply]

Yes, I am the review board; that's my responsibility. I doubt having my students participate in AfC (permanently backlogged) would result in their work getting better oversight than what I already provide. Each semester I reach out to the applicable WikiProjects and typically get no interaction from any of the editors there. My courses are linked on my user page; please stalk them. Chris Troutman (talk) 16:05, 9 September 2014 (UTC)[reply]

What method is used, to alert participants in wikiprojects about requested articles in their area of interest? Jim.henderson (talk) 03:55, 8 September 2014 (UTC)[reply]

Method? Are you kidding, Jim? There are nearly two-and-a-half thousand Wiki projects... KudpungMobile (talk) 11:20, 9 September 2014 (UTC)[reply]

At the moment I have a script here that picks up any drafts who have a project template put on the talk page, and have made steps to try and categorised drafts, but other than that, nothing else at the moment.

I got involved with AfC because I did not like the way new editors were treated, which has been well documented in WP:NEWT (although I trust things have improved since then), and hoped it would be a nicer route into creating articles. On occasion, where a submission has piqued my interest, such as The Minories, Colchester or Rainthorpe Hall, I have tried to work with the editor and improve the draft. However, activities like this are very much in the minority.

What I would like to see is "drafting" as an alternative process for both deletion and AfC. A new article that did not meet any of the more blatant CSD criteria which have legal ramifications (eg: G10 - attack page, G12 - copyvio), or a borderline AfD (such as Little Sea (band)) could be moved to draft instead of told to get stuffed. The creator would work with the NPPer to improve it, preferably by bringing project experience in. This would replace the AfC process (and for all the posturing about a "project", I think "process" is the more correct term). Ritchie333 ^(talk) ^(cont) 13:29, 9 September 2014 (UTC)[reply]

Ritchie, I certainly agree that process is the more correct term, but you may just possibly be forgetting one thing, and which is the major problem facing both AfC and NPP: Where are we going to get all those reviewers/patrollers from? NPPers are generally unable/don't want to do most of the tasks that are required at WP:NPP for effectively protecting Wikipedia from copyright, libel, hoax, and vandalism issues. What we need are ideas how to attract more editors - and ones of the right calibre - to the task.

You have a script here that picks up any drafts who have a project template put on the talk page - but have you thoght for a moment how that project banner gets there in the first place? KudpungMobile (talk) 23:58, 9 September 2014 (UTC)[reply]

Kudpung To clarify, I'm talking specifically about modifying behaviours of the resources we do have to favour drafts where practical. The AfD and CSD queues are alive and well, with numerous people willing to wipe content off the face of the earth. It's them I want to target a behavioural change.

I'm aware that the script I knocked up requires people tagging and categorising stuff, but you have to start somewhere. Ritchie333 ^(talk) ^(cont) 13:01, 10 September 2014 (UTC)[reply]

User:Ritchie333: I am quite sympathetic to the idea of moving to drafts as an alternative to deletion, but we currently seem to have no sound process for doing so. A WP:BOLD move would leave a redirect behind, which would be probably undesirable. Sending an article to WP:AFD with the intention of moving to drafts would be probably met with editors saying "Speedy keep, this is articles for deletion, not articles for drafting.". WP:RQM seems rather unfitting. I think implementing this idea would require a major reform of the deletion process.

If I were to do that, I would probably create a WP:ADFV — Articles and Drafts For Verification, which would deal exclusively with notability and verifiability concerns. If you send an article to ADFV and it fails to be verified, the article is moved to drafts. If you send a draft to ADFV and it fails, it gets deleted. Regular AFD would be left for matters relating to project scope, when making a case for deleting an article regardless of verifiability. Also, it goes without saying that AFD should apply to drafts as well. — Keφr 18:29, 11 September 2014 (UTC)[reply]

What the Wikipedia community does to hamstring itself is throw up nearly impenetrable walls of bureaucracy which are intended (sincerely, I believe) to help and draw in newcomers. They do not. I would never have written my first article had AfC been around. I stopped contributing to DIY because the level of scrutiny (for example, "close paraphrasing," a standard which is incredibly subjective) became the norm. If we want more contributors, and more diverse contributors, we need less process and more knowledge. Every piece of bureaucracy is a form of ownership, cleverly disguised by people who have enough familiarity to work within this incredibly complex system. Nothing will change until everything changes, and it saddens me.--~T P W 15:24, 9 September 2014 (UTC)[reply]

I too avoid the processes that require using particular templates or graphic voting symbols. I am not convinced that any of them are necessarily done that way, and everything we do with a complicated template needs to be re-evaluated for first, whether it is worth doing oat all, and second, whether it can be done more directly. But standards will inevitably be subjective at the boundaries. DGG ( talk ) 03:22, 10 September 2014 (UTC)[reply]

I am also pretty aghast at the AfC process, having written hundreds of articles prior to now and never using it before, it was suggested to me to use it when I wanted to write an admin-locked page that is presently a re-direct. My article Draft:Godwin Grech has been rejected (after almost a week of nothing happening at all), and I don't agree with the reasons... so it's left me feeling dejected and not sure what to do next. Not a good feeling and probably particularly not-good for newbies! Clare. (talk) 08:39, 10 September 2014 (UTC)[reply]

That's a really interesting piece of feedback, Clare. Simply put, you and the reviewer have had a content dispute. He thinks the redirect should stand per a five year old AfD (where it was "speedy" deleted for WP:BLP1E, not a CSD criteria and hence an abuse of the tools, but that's old news), you think the article should stand on its own merits. Quite how AfC is supposed to resolve that, I've no idea. In your shoes, I would probably have gone straight to WP:AN without going near AfC, supplied links to a few sources and request desalting. Ritchie333 ^(talk) ^(cont) 12:49, 10 September 2014 (UTC)[reply]

I did offer to de-salt for User:Clare. as a seasoned article creator (User_talk:Ronhjones#Godwin_Grech), but she was happy to go with WP:AFC. Someone else has now asked for de-salting to move her article to the page (User_talk:Ronhjones#Godwin_Grech_2), which I have done. Ronhjones ^(Talk) 13:51, 12 September 2014 (UTC)[reply]

Thank you to everyone who helped me on this, I really appreciate it. Clare. (talk) 13:52, 12 September 2014 (UTC)[reply]