Wikipedia:Wikipedia Signpost/2018-12-24/WikiProject report

WikiProject Articles for Creation is the WikiProject whose active members volunteer to review submissions to Articles for Creation. While any autoconfirmed user can review drafts manually, it is strongly recommended that only participants in the WikiProject do so. Participants must have extended-confirmed rights in order to  use the Helper Script,  and permission  is accorded by administrators; the full criteria for  access to  the tools can be found here. The introduction of this control was passed by consensus on a RfC proposal made by in 2013.

Unlike its counterpart New Page Reviewing, as a project AfC has no formal selection or vetting process anchored in policy for its operators and their tasks, although it is often claimed that the required level of knowledge of notability and other guidelines is very similar. A move towards more formal acceptance of the project was established through the Wikimedia Foundation's 2018 efforts to develop a more structured approach by embedding a live feed for new AfC submissions in the New Pages Feed. Many AfC reviewers are holders of the NPR flag. Armed with this knowledge, any reasonably experienced editor should not find the tasks too  complicated – while they  may  however present challenge to less experienced users, the actual instructions for reviewing are easy to follow, and are supported by a flow chart.

AfC reviewers have multiple tasks that fall within their purview related to the Articles for Creation process. First and foremost, they assess submissions to determine their suitability for the encyclopedia. This is done with the help of a user script. Secondly, they answer questions from editors about their reviews, and provide feedback on how they can improve their article. Finally, they assist editors at the help desk.

​

Q and A with reviewers Nosebagbear, Legacypac and K.e.coffman
The Signpost sat down with reviewers Nosebagbear, Legacypac, and K.e.coffman to discuss their contributions.

What motivated you to become a member of WikiProject Articles for Creation?

Nosebagbear: 2 main reasons encouraged me to sign-up to help out with AfC – Articles for Deletion & ACPERM. The former, slightly counter to the idea of AfC, was my first real area in the back-end of Wikipedia, and one of the things that I found so frustrating in my early time at AfD were the number of new articles that were in no way ready for article-hood and would be deleted within 2 weeks of uploading. As a freely admitted inclusionist, I felt that helping editors through a less destructive beginning seemed like a great way to avoid both biting new editors and to ultimately bring in more, better, content to Wikipedia. [Secondly, I] finally turned my long-term account active while the post-ACTRIAL RfC was taking place. The consensus was overwhelming in two ways – yes it's needed & yes there are going to be some big problems. I felt it was needed so joining up seemed a great thing to do.

I lied. There was a third reason. It's good fun – and you learn some weird things before anyone else does!

Legacypac: The opportunity to help good new content get into the project. I also enjoy maintenance, clearing out junk, spam and otherwise useless pages to improve the average quality of pages available here.

K.e.coffman: I've been nudged a couple of times on my Talk page, but joined only recently. Early on, I accepted a few drafts that had been submitted three months prior, which seemed like a really long time to wait to get one's draft reviewed. This motivated me to keep going. The current backlog is just shy of four weeks, which is much more reasonable.

''There has been discussion about integrating Articles for Creation into the New Pages Patrol. What are your thoughts on this?''

Nosebagbear: Hmm. I'd say I have a mixed view. It obviously appears outwardly reasonable, but it depends on a couple of things. A properly shared reviewing platform/set-up has to be able to handle both sides well, rather than with an NPP focus. There is already a good script made by &, so any system that wasn't at least as good at that would be useless. In other ways, I just don't know, more clarity on quite how close a merger is needed – NPP and AfC both make processes that work for them, and so a tight merger risks disrupting a system designed for drafts (often 1st time drafts). I would say my biggest concern is being pushed into a less than suitable set of changes without a sufficient chance to discuss a proper set of proposals.

Legacypac: The skill set is the same and the qualifications should be fully aligned. The major difference between AfC and NPP is the location of the pages. Drafts have been brought into the NPP tool recently so NPP reviewers can review Drafts the same way they review in mainspace.

K.e.coffman: The two projects have overlapping goals – shepherding new articles into mainspace while maintaining the integrity of the encyclopedia – but I don't think that the functions are exactly the same. The NPP browser now allows users to toggle between the two processes, which is useful.

''The Articles for Creation backlog has fallen dramatically recently. How are reviewers keeping the backlog low?''

Nosebagbear: So one of the big changes is the number of new reviewers (as well as lots of returning reviewers) has increased significantly in the last couple of months. In an ongoing sense, this helps reduce the demand on a handful of super-reviewers doing the bulk of the work. It's one of the big benefits that has shown up from the steps towards merging NPP & AfC – lots of the active page patrollers have spread some of their time and also contributed in AfC. That said, it is that 10% of most active reviewers that cover over half of each month's submissions, so I think credit must flow for that. Finally, and very controversially, we recently had an editor review about 1100 drafts – unfortunately, the editor in question was identified as a hostile Sock. This did trigger a silver lining though of review-reviewing, with a large number of both accepts and declines looked over. This burst has put us in a much better position to keep on top of the review timeline rather than let it build up.

Legacypac: Like most backlogs, it fluctuates mostly because one or a small group of editors changes their activity level. AfC always processes most drafts within the first two to three days of submission. The long backlog is usually the minority of pages that are harder to assess for some reason. We were able to cut that from 8+ weeks to 3 weeks through some dedicated effort. We also cut the total pending from around 2300 to under 1300. Ideally all submissions would be answered within a few days but that will require more volunteers.

K.e.coffman: The new rejection feature (to discourage resubmissions) is helping; I'm no longer seeing drafts that have been resubmitted two, three, four times. Fewer resubmissions mean fewer total submissions, and it's better to be upfront with the authors if the topic is not suitable, rather than implicitly asking them to invest more time into "hopeless" drafts.

Have you seen a difference in your workflow following the conclusion of WP:ACTRIAL?

Nosebagbear: Not having a 1st person experience before and after I can't really compare – however the month after I joined (in May) received the most submissions AfC has ever had, so it was a fun start time.

Legacypac: WP:ACREQ was a huge success. It cut down the submission of junk into mainspace dramatically. However there was a pretty big uptick in AfC submissions, but not nearly as much as the reduction in problematic mainspace submissions so overall it was a big win. At AfC we can decline page rather than always seeking deletion like in NPP so most of he junk just gets quietly deleted WP:G13 in six months or so. This is a much better use of volunteer time overall.

K.e.coffman: I only joined the project two months ago, so I don't have the first-hand experience with what has changed compared to before ACTRIAL. I would argue that the bar should be raised even further than ACTRIAL went – four days and 10 edits in no way prepare news users to write encyclopedia entries directly in mainspace. More drafts being channelled to AfC would hopefully relieve the pressure on NPP reviewers, while improving the overall quality of articles that reach mainspace. To keep the AfC backlog manageable, we would need more AfC reviewers, of course.

What are WikiProject Articles for Creation's most pressing needs?

Nosebagbear: New Reviewers! 40 new reviewers doing 20 reviews each month would let us keep the turnaround time to less than a week which would be brilliant.

Legacypac: We need some more reviewers. It is an interestimg mix of quality control and encouraging great new content. I learn a lot from reading random submissions, even the spam is interesting as I learn about new companies and ideas.

Some advice from Legacypac to anyone interested in joining:

There is a PERM like process to access the reviewing script at Wikipedia talk:WikiProject Articles for creation/Participants. If you are not ready for that, there are interesting ways to still help while you can learn more about the process and build up confidence. We are in the business of sorting out the junk and promoting the good. You can CSD spam/copyvio and MfD other problems in Category:Pending AfC submissions. If you prefer content creation or gnoming, please look for good notable topics in the same category and improve the pages. Many submitters are new, without the skills to add sources properly, format pages correctly, and other issues. Helping get good pages ready for mainspace and eliminating bad pages makes the AfC reviewer's job much easier.

K.e.coffman: The project can always use more reviewers. I found that AfD experience makes AfC a seamless transition. Exposure to AfD and various notability guidelines is helpful in knowing which subjects are and aren't likely to survive a deletion discussion. With that in mind, I would encourage AfD regulars to try their hand at AfC. There's also an interesting proposal on the table regarding "Drafts for assistance", which would hopefully draw other projects into lending a hand. I'd like to see this idea implemented.

The Signpost would like to thank the interviewees for their time. If you would like your WikiProject featured in a future report, please drop us a line at the Newsroom talk page.

Kudpung contributed to this report.

​