Wikipedia talk:Content assessment

A class vs GA vs B

First off can an article be a and ga or just one also which one is higher a or ga and why are there so few a articles? TropicalGalaxy (talk) 13:53, 6 March 2025 (UTC)

TropicalGalaxy, A-Class is paradoxically higher than a Good Article (GA). There are so few such A-class articles or better, because folks (1) like to create articles because it's a quick way to get that "dopamine rush", (2) only have the time, energy, or temperament to make small changes and fixes, but (3) often don't have the time, energy, or temperament necessary for making big changes. There's politically incorrect reasons for that phenomenon. For example, it's taking me months to improve dress code. By the way, can you assess it? Please and thank you. Bearian (talk) 12:50, 19 March 2025 (UTC)

Sorry, but I didn't see that you changed your user name to OrangeLolipopSnail. Bearian (talk) 12:55, 19 March 2025 (UTC)

But wouldn't A be easier to get than GA because for A just 2 people have to agree but for GA an admin or something like that (I forgot the name) has to review it. OrangeLolipopSnail (talk) 21:29, 20 March 2025 (UTC)

WP:Good Article says Any impartial editor may review the article, so as I understand it the process is more or less the same. And at that point, if you're going to go through all the work, I suppose you may as well just keep shooting for FA rather than settle for A Viv Desjardin (talk) 02:40, 30 March 2025 (UTC)

Ok thanks OrangeLolipopSnail (talk) 02:49, 30 March 2025 (UTC)

Some editors do have the expertise, energy, and temperament to take articles all the way to FAC. An article that passes an A-class review is likely to pass FAC. But FAC only allows one nomination at a time (two if one is a co-nomination). It is therefore of benefit to run an article through A-class to resolve as many issues as possible before the FAC nomination, because this will minimize the time an article spends in the queue at FAC, which can run into months, and the chance of a failure at FAC, which attracts a two-week penalty. Hawkeye7 (discuss) 19:59, 30 March 2025 (UTC)

The highways projects required that any article nominated at WP:HWY/ACR be a GA. If passed through ACR, that article would have both statuses, as Pierce Stocking Scenic Drive does. Imzadi 1979 → 03:09, 30 March 2025 (UTC)

Thank you OrangeLolipopSnail (talk) 08:50, 30 March 2025 (UTC)

@OrangeLolipopSnail: The main reason why there are so few A-Class articles is that A-Class reviews only happen in very few WikiProjects (especially Military History; I don't actually know of any other project that currently has a reasonably active review process). For most non-Military History purposes, you can safely ignore A-Class. —Kusma (talk) 20:23, 30 March 2025 (UTC)

Thank you OrangeLolipopSnail (talk) 02:38, 31 March 2025 (UTC)

thx for clarifing 49.131.86.202 (talk) 13:46, 24 April 2025 (UTC)

Start-class "graphic"

I'm going through as many Law stubs as I possibly can this year, with a view at more accurately assessing Start- and C-class articles mislabeled as law-stubs, or never assessed at all. I've come across many articles about legislation and court cases with a seal or flag (especially UK or US) in the Infobox. Does that count as a "graphic" as defined by the Start-class criteria or factor? Yes or no, as as the attorney cross-examining a hostile witness says. I know that you are not hostile; I just want to efficiently get through assessment of 16,700-plus stubs. Bearian (talk) 12:39, 19 March 2025 (UTC)

In my opinion, as a fellow stub reviewer, having a seal as a graphic might contribute to it being Start-class, but only in conjunction with other Start-class features. It is a graphic in the literal sense but it doesn't support an article to the same extent as a relevant photo or diagram would.

Especially when you consider WP's description of a reader's experience of a Start-class article: Provides some meaningful content, but most readers will need more. If the article is otherwise stub-like, having a seal probably wouldn't make a reader feel like the article provides some meaningful content.

Not a very good yes-or-no answer but unfortunately there really isn't a good, consistent definition of what makes an article a stub vs start-class :) Viv Desjardin (talk) 20:44, 29 March 2025 (UTC)

I find the Start-class description unhelpful. A more practical definition is: A Start-class article has too many words/sentences to be a Stub, but doesn't meet the criteria for C-class.

@Bearian, if you could get something like m:Research:Screening WikiProject Medicine articles for quality/Stub prediction table created for law articles, and especially for the unrated or Stub-rated pages that are estimated to be C-class or higher, you might find that faster than individual manual review of everything. (I don't know how to make the table. I just ask Nettrom to do it as a favor every couple of years.) WhatamIdoing (talk) 01:27, 30 March 2025 (UTC)

Thank you for the feedback. It's a long-term project, and I think one that requires some human intervention and judgment, but the table looks intriguing. Nettrom, how do you create such a table? Bearian (talk) 01:32, 30 March 2025 (UTC)

@Bearian: I have a couple of Python scripts that I've written for this project. One of them I give the category of stubs, and it then uses wikitech:Machine Learning/LiftWing to get predictions of the current quality of those articles. Then I have another one that creates the table so I can easily copy & paste it into a page. If you want me to give it a try for law articles, is Category:Stub-Class law articles the one I should use? Cheers, Nettrom (talk) 16:18, 30 March 2025 (UTC)

Yes, please and thank you! Bearian (talk) 16:47, 30 March 2025 (UTC)

Pinging Nettrom. Bearian (talk) 11:54, 31 March 2025 (UTC)

Thanks for the ping, @Bearian! I had forgotten that I can subscribe to talk page threads now, and so I wouldn’t have seen it otherwise (but I’m now subscribed). I’ve got the script running, might be a day or two until I have the results. Cheers, Nettrom (talk) 14:53, 1 April 2025 (UTC)

Fantastic! Many thanks. Please ping me or drop a link on my talk page. Bearian (talk) 14:56, 1 April 2025 (UTC)

@Bearian: I've started listing the pages on User:Nettrom/sandbox/WikiProject Law stub predictions. Currently it only holds the articles with a >90% probability of not being a stub, fashioned after the WPMED page that @WhatamIdoing has. I might also add articles from the 80–90% and 70–80% categories, once I figure out how to get around the 2MB edit size limit (there’s a lot of articles in these tables). Cheers, Nettrom (talk) 13:54, 2 April 2025 (UTC)

Thanks, Nettrom, 90% is a good number. No need to be a perfectionist at the moment. Bearian (talk) 15:06, 2 April 2025 (UTC)

For a first pass, it might be worth being even more limited, say, 98%, or only those with a two-class difference (C-class or higher). Reviewing thousands of articles is a really daunting task. WhatamIdoing (talk) 16:21, 2 April 2025 (UTC)

How Bold Should We Be?

I was just reading an article and totally disagreed with its content assessment. Negative or positive? So what? Who cares? CA is completely subjective and unhelpful. I want it gone. G41rn8 (talk) 20:16, 11 May 2025 (UTC)

It's not completely subjective, there are established criteria. The processes that have developed around it, especially at the higher end, are quite helpful to article development. CMD (talk) 02:28, 12 May 2025 (UTC)

I agree with CMD, and if you think CA is subjective/unhelpful, you can always just ignore it. GoldRomean (talk) 16:15, 20 May 2025 (UTC)

Subjectivity isn't a bad thing (assuming you mean actual subjectivity, rather than arbitrary and capricious assessments), but the biggest problem with assessments is that they're often out of date. WhatamIdoing (talk) 21:02, 21 May 2025 (UTC)

List-class is problematic

I've always found List-class to be strange because it does not say anything about the quality of the article. And sometimes it is difficult to identify an article which should be List-class. What if a significant proportion of the article is a list, but not all. For example, should The Game Awards 2018 be List-class? — Martin (MSGJ · talk) 11:36, 12 June 2025 (UTC)

"List-class" says it is not a Featured List, which says something about the quality, but not very much. —Kusma (talk) 12:11, 12 June 2025 (UTC)

I'd prefer we did it more like military history and have Start-class lists, C-class lists, etc. all the way up to Featured lists — Martin (MSGJ · talk) 12:21, 12 June 2025 (UTC)

If we want to do this properly, "list" should perhaps be a yes/maybe/no switch in addition to the quality rating. —Kusma (talk) 19:35, 12 June 2025 (UTC)

I had supported this 2/3 years ago, and I will do so again if it is brought up. That "rating" to be separated from "type". List could be a separate switch, or it could be a defined as a pagetype, both are acceptable to me. —CX Zoom[he/him] ^{(let's talk • {C•X})} 22:47, 12 June 2025 (UTC)

The Game Awards 2018 is a non-list article that happens to have multiple Wikipedia:Embedded Lists in it.

I think that editors would find it difficult to decide what constitutes as Start-class list vs a C-class list. (Look, e.g., at 4 mm caliber, which MILHIST rated as C-class list, but I'd have guessed was no more than a Start.) In particular, if the list itself is the usual simple list of bullet points, often with less description than a Wikipedia:Disambiguation page, you'd likely have to know something about the subject matter to be able to determine whether it was passably complete. WhatamIdoing (talk) 04:35, 13 June 2025 (UTC)