Talk:Community Wishlist/Wishes/Improve review speed

This page is for discussions related to the Community Wishlist/Wishes/Improve review speed page.   Please remember to: Sign posts using the four tildes (~~~~) Remain civil and polite during discussions. Place new text under old text (start a new post).

Invalid wish?

This sounds like a specific project issue. It might be resolved via getting more users to perform draft reviews and this is something not really feasible to do by coders/maintainers. We don't even need to add new features as the toolbox for draft reviewing was significantly revamped a few years ago. I suggest that an admin/community wishlist maintainer archives this wish as invalid/out of scope. That does not negate the existence of this problem, it just isn't the right venue to alert about this problem. Best regards, A09|(pogovor) 18:27, 28 July 2024 (UTC)

I agree with the above. However, I think generally it may be valid if the user had provided a specific concrete way to speed up the reviews. Prototyperspective (talk) 19:17, 28 July 2024 (UTC)

While true, I do think that there are lots of 'tasks' and 'queues' that would benefit from better software support. In how far the lack of those features is responsible for this specific backlot backlog (which I assume to be for English Wikipedia), that is another questions indeed. —TheDJ (talk • contribs) 14:45, 29 July 2024 (UTC)

I'd like to leave just a friendly note and also appreciation in sign of support of your comment. When implementing things on other wikiprojects that already exist on the English Wikipedia we must acknowledge that English Wikipedia systems are far over complicated to a small- or even middle-sized project thus we shouldn't straightforward copy the existing enwiki code but rather try to localise it and fit to the local needs (ie. a good example, RFCH tool). A09|(pogovor) 14:58, 29 July 2024 (UTC)

@A09 this wish was extremely well articulated; not every wish should propose a solution. In fact, this wish is really helpful as it articulates the backlog of reviews as a problem for newcomer retention.

If, for example, the wish proposed a solution to "build an AI tool for reviewing new content," then we would be mincing words on the veracity of this solution. Rather, this gives us space to think about the conditions upstream and downstream that could impact the review queue. JWheeler-WMF (talk) 20:32, 29 July 2024 (UTC)

Saying it was extremely well articulated makes this sound quite implausible. It's also not elaborating much on the problem either or contributing to any solution such as proposing several example ideas for what could be done. It's like saying "Wikipedia needs more articles" which is overly broad. Maybe wishes that are basically just topics make sense but I think that has been unclear and kind of misses ideas underneath them and a topic to related ideas connection and so on. One major condition is that there's only few people doing these reviews and e.g. things that would facilitate more editors to review articles and adequately so would not just apply to that but to any task-type where more people are needed and as such is a kind of broader discussion. In short, I think broad problems as subject may be helpful but I think would be much better if they proposed some example ideas about what to do and if it was made clearer that the Wishlist is also about such broad things or topics. Prototyperspective (talk) 20:54, 29 July 2024 (UTC)

Thats fair - it wasn't "extremely well articulated," and it could offer more specifics to the problem. But, it does point out the user challenge and frustration of waiting days or weeks to find out their edit is reverted.

This specific topic is a big problem, with many sub-problems nested underneath it. There isn't a "magic pill" to solve everything. Still, I think it's a well-written wish that gets to the heart of a major issue, and I would love to see more users connecting their wishes to problems like this. JWheeler-WMF (talk) 21:08, 29 July 2024 (UTC)

On the English Wikipedia a lot of the people that are in a real hurry to get their drafts moved to mainspace are those who promote companies or products. To them, any delay is unacceptable. People who write an article about their favourite book for example are a lot more chill. Polygnotus (talk) 20:53, 4 September 2024 (UTC)

Well I think the title of the wish is "extremely well articulated", sure it seems a bit vague but we can figure out technical solutions on this talk page. The review delay on English Wikipedia is really frustrating for submitters. Back in the old days you could just cobble together three sentences, no references, no categories and the article would survive and be improved. Try that today. To help this wish we could:

1. Suggest drafts to review (to established users). I mean the system detects an established user's interests and matches them with drafts recently created. In my case I probably edit topics in the area of Australia and Tunnels. Really any draft in these areas I would be happy to take a look at, but I am not inclined to search them out. The suggestions could be opt-out.
2. Separate out dodgy drafts. I am not interested in reviewing promotional drafts, or drafts about topics that have an extremely low chance of meeting policy standards. There should be a way of marking a draft as "probably legitimate". This is subjective of course. Then perusing the list of drafts becomes more manageable.
3. I think TheDJ mentioned queues. I had a quick look and couldn't find a queue for drafts, but probably a bot maintains one somewhere. Combined with the above two features I think a queueing system, at least a list with most recent draft submissions at the top, would increase review speed.

Thanks for the wish @Feyan Li. Commander Keane (talk) 23:12, 25 October 2024 (UTC)

I think wishes should described technical solutions even if not a complete list of approaches or vague but nothing of that sort is done here so I think it would be best if (a) new wish(es) were created in its place that describe how to plausibly increase review speed with descriptive title(s).

1. Good idea! I think that would be an interesting task type to add into the Personalized suggested tasks system proposed in wish Suggested tasks based on contributions history (user interests) also for experienced editors and I'll probably add this to its example tasks.

3. Maybe he meant Category:AfC submissions by date (?)

Prototyperspective (talk) 16:25, 26 October 2024 (UTC)

Draft or new article ?

One question I have @Feyan Li, is if you were thinking about Drafts, Articles for creation or new page patrolling process specifically, or simply all of these kinds of processes ? Would love to hear more about your experiences. —TheDJ (talk • contribs) 08:53, 28 August 2024 (UTC)

In the current Wikipedia system, when we create a draft and submit it for review in the new page patrol, we often have to wait over a month for “volunteers” to review it. Unfortunately, our drafts typically do not get approved on the first try. When a draft is marked as “needs improvement,” we have to revise and resubmit it, which means waiting another month. This back-and-forth process can take several months just to get an entry approved. I believe this is unacceptable for those who are committed to contributing to Wikipedia.

For example, on Chinese Wikipedia, there are currently only two or three active users handling new page patrol (since most drafts are reviewed by them), which further exacerbates the review time. On the other hand, on Baidu Baike, which is often criticized for plagiarism, new page reviews typically take only a week, and sometimes even as little as three days.

I think dismissing this concern as irrelevant is irresponsible. As discussed by other users, this issue is very real. I am just a user, and the question of how to improve the review speed should be a matter for the community to discuss and solve, rather than simply ignoring the problem. Feyan Li (talk) 10:02, 28 August 2024 (UTC)

It's called Community Wishlist, not Community Problemlist or list of problems. I don't see why this wouldn't needs to be archived.
If there are many ideas how to solve the problem of slow review speed then these maybe could be part of one focus area about that problem. Prototyperspective (talk) 23:38, 31 August 2024 (UTC)

A good way to redcuce the review backlog

The most effective way to speed up the review process would be to deter new users from submitting drafts, until they're experienced enough to do so competently. This would also have an even bigger benefit: if new users didn't so often embark on trying to create new articles, they'd be less likely to become disheartened, and give up contributing altogether. Maproom (talk) 21:21, 29 October 2024 (UTC)