Wikipedia:Village pump (miscellaneous)

The miscellaneous section of the village pump is used to post messages that do not fit into any other category. Please post on the policy, technical, or proposals sections when appropriate, or at the help desk for assistance. For general knowledge questions, please use the reference desk.

Discussions are automatically archived after remaining inactive for a week.

« Archives, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79

Talk:Evangelos Marinakis has an RfC

Talk:Evangelos Marinakis has an RfC for possible consensus. A discussion is taking place. If you would like to participate in the discussion, you are invited to add your comments on the discussion page. Thank you. D.S. Lioness (talk) 18:01, 22 July 2024 (UTC)[reply]

This appears to be a question about whether to include the word oligarch. WhatamIdoing (talk) 18:24, 23 July 2024 (UTC)[reply]

That's right! D.S. Lioness (talk) 18:25, 23 July 2024 (UTC)[reply]

Performing a random pages test on business articles

I've been focusing a bit on Wikipedia's articles on businesses recently. I think anyone who spends even a little time in the topic area knows we have problems with promotion, COI/paid editing, and dependence on non-independent sourcing. However, it's difficult to find hard numbers on just how big the problems are. To remedy this, I'm trying to obtain a random sample of a few hundred Wikipedia articles on companies to assess the extent to which they comply with our content policies. However, I'm having a bit of trouble working out how to get a good sample:

Idea 1: Hit Special:Random. If article is about a company, add to list. Otherwise discard. Repeat until satisfied with number of articles on the list. Problem: Time-consuming and impractical for single editor.
Idea 2: Use query service (Petscan or Quarry) to generate a list of company articles based on Wikipedia category system. Problem: Wikipedia category system not suited to this task. Structure of Category:Companies tree includes too many non-company articles (e.g. biographies in Category:People by company). Difficult to reliably filter.
Idea 3: Query Wikidata for items about companies with an enwiki article. Problem: Potential for systematic bias in results if Wikidata editors focus on creating items for highly notable companies.
Idea 4: Query DBpedia for company entities with enwiki articles. Problem: Classification as "company" apparently somewhat unreliable. Too many non-company articles in query results. Results seem to only appear if enwiki page was created before or up to ~2022.

I'm a bit stumped on what to do. Is there a way to adapt one of my ideas to produce good results? Or is there another idea I'm missing? – Teratix ₵ 16:05, 23 July 2024 (UTC)[reply]

For sure, trying to navigate the category tree from Category:Companies will just end up in a world of pain. You'll do much better using wikidata, say starting from company (Q783794). There's a dedicated query language for this sort of thing, see wikidata:Wikidata:SPARQL tutorial. RoySmith (talk) 16:15, 23 July 2024 (UTC)[reply]

Yep, I've written a SPARQL query that pulls a random sample of Wikidata items on companies with enwiki articles. My concern is that this sample may not necessarily be representative of a random sample of companies with enwiki articles, depending on how Wikidata editors select which enwiki businesses to create items for.

If, for example, Wikidata is more likely to have an item on a prominent rather than an obscure business (given both have enwiki articles), then a random sample of Wikidata items will feature more prominent businesses than a random sample of Wikipedia articles, which could lead to biased results.

But I don't really know much about how Wikidata editing works, so this could be wrong. – Teratix ₵ 16:33, 23 July 2024 (UTC)[reply]

(edit conflict)

Modify {{infobox company}} (used in ~85,000 articles) so that it emits a tracking category listing all articles that use that template? Allow the category to populate (could take a month or more). Add {{random in category}} to the category (or use Special:RandomInCategory) to fetch your random samples?

If there are other infoboxen that are commonly used in business articles, do the same with those templates; populate only the one common category.

When done, revert your template edits and delete the category.

—Trappist the monk (talk) 16:23, 23 July 2024 (UTC)[reply]

RandomInCatagory isn't very random (see T230700 and T200703). The strategy it uses is fine for an end user who wants to idly article hop, but I wouldn't use it for anything that requires statistical rigor. Keying off {{infobox company}} seems like a reasonable approach, but it suffers from a lot of the same problems the category ~~tree~~ ~~DAG~~ ~~graph~~ blob does. There's many similar templates, all related in a quasi-tree structure, but not easy to navigate. You might start from {{Organization infoboxes}}. It also suffers from the wikidata problem of people who write company articles being hit-or-miss about whether they add infoboxes of any kind.

You might want to go with multiple approaches to discover company articles and combine/deduplicate the results. Asking ChatGPT was amusingly useless. RoySmith (talk) 16:54, 23 July 2024 (UTC)[reply]

No need to mangle mainspace even temporarily with a tracking category; you can do something like this if you accept that transcluding {{Infobox company}} is good enough. —Cryptic 17:06, 23 July 2024 (UTC)[reply]

More than that, please don't "mangle mainspace". I know this idea was well-intended, but historically such things have been frowned upon. RoySmith (talk) 17:12, 23 July 2024 (UTC)[reply]

Please explain how tracking categories mangle mainspace. Link to the consensus discussion that states that such things have been frowned upon.

—Trappist the monk (talk) 18:04, 23 July 2024 (UTC)[reply]

I can't find it now, but there was a thread (perhaps on WP:VPT? a few months ago about one of the mobile apps adding tags for its own tracking purposes. The general consensus was that it was a bad idea. There's a related phab ticket at T360164. RoySmith (talk) 18:19, 23 July 2024 (UTC)[reply]

That was about edit summaries, not actual article content. WhatamIdoing (talk) 18:26, 23 July 2024 (UTC)[reply]

My wording was perhaps facetious, but - at least on my part - the intent wasn't so much as "this would be harmful" as "this would be impractical". It's going to take a while - perhaps a long while - for the category to get fully populated after you add it to the template (unless you null edit all its transcluders, which has its own problems). And you might have to go through several iterations of adding and removing templates to/from your dataset. —Cryptic 18:28, 23 July 2024 (UTC)[reply]

You can use toolforge:randomincategory for a more truly random selection from a category. It will be a bit slow the first time you run it on an 85,000 member category, but it should work (and it caches data for 10 minutes so subsequent runs should be fine). --Ahecht (TALK
PAGE) 18:02, 23 July 2024 (UTC)[reply]

Great idea. How about going the Petscan route but using one of the more systematised subcategories of Category:Companies, like Category:Companies by country? I imagine that almost all company articles are in that tree. – Joe (talk) 16:55, 23 July 2024 (UTC)[reply]

We've actually already tried that, specifically with cat:Companies by country. It gets way too many non-company articles way too quickly, even after pruning categories starting with "People by company". —Cryptic 17:13, 23 July 2024 (UTC)[reply]

Yeah, that's typical for category traversals. Honestly, having accumulated a few scars from trying things like this in the past, I think the wikidata route is your best bet. Or at least your least bad bet. RoySmith (talk) 17:19, 23 July 2024 (UTC)[reply]

You might be able to leverage Category:WikiProject Companies articles and the advice given to me at Wikipedia talk:PetScan#Help creating query. WhatamIdoing (talk) 18:28, 23 July 2024 (UTC)[reply]

Thanks for the input, everyone. I've reflected a bit and come up with a hybrid/kludge solution that might work: deliberately getting a larger than optimal sample from Category:Companies by country with Petscan/Quarry, filtering out non-company articles from the sample by checking with Petscan whether they match a Wikidata query for whatever unwanted types (e.g. biographies) tend to show up, then just manually discarding anything unwanted that sneaks through the filter. – Teratix ₵ 15:18, 25 July 2024 (UTC)[reply]

Porandla

-- GreenC 01:03, 25 July 2024 (UTC)[reply]

An impressive journey, originally located quite far inland, the village moved to the coast, then moved again back inland but more to the northeast. (The first and last both seem to be clear villages on google maps, and there is at the very least a street with that name in the location of the second one.) CMD (talk) 02:06, 25 July 2024 (UTC)[reply]

It also had a different name from 2011 before losing all its text in 2022, but seems never to have had any source. PamD 05:40, 25 July 2024 (UTC) expanded 08:49, 25 July 2024 (UTC)[reply]

This source supports the statement in the original version of the article, so perhaps we should revert to that and add the source - and choose whichever of the later-added coordinates seems appropriate. PamD 09:01, 25 July 2024 (UTC)[reply]

I would also say it should be reverted to maintain the original intent, but there will also be sources to support the current version of the article, as the new version is literally for another town. The telugu page (te:పోరండ్ల (జగిత్యాల)) has always been about the current (Jagtial) Porandla, as has the associated Wikidata item (wikidata:Q13003257). The original (Ranga Reddy) Porandla is at te:పోరండ్ల (మహేశ్వరం)/wikidata:Q16340753.

If the original wording is restored, the thing to do here would be to revert, split off Jagtial Porandla, disambiguate Ranga Reddy Porandla, and then switch the relevant Wikidata entries.

(As an aside, the one-up division, te:జగిత్యాల గ్రామీణ మండలం is one of the few Jagtial district#Mandals without an en.wiki article.) CMD (talk) 04:15, 26 July 2024 (UTC)[reply]

Vote now to fill vacancies of the first U4C

You can find this message translated into additional languages on Meta-wiki. Please help translate to other languages.

Dear all,

I am writing to you to let you know the voting period for the Universal Code of Conduct Coordinating Committee (U4C) is open now through August 10, 2024. Read the information on the voting page on Meta-wiki to learn more about voting and voter eligibility.

The Universal Code of Conduct Coordinating Committee (U4C) is a global group dedicated to providing an equitable and consistent implementation of the UCoC. Community members were invited to submit their applications for the U4C. For more information and the responsibilities of the U4C, please review the U4C Charter.

Please share this message with members of your community so they can participate as well.

In cooperation with the U4C,

RamzyM (WMF) 02:46, 27 July 2024 (UTC)[reply]

Arabic Wikipedia help

Hi, I don't speak Arabic well and have been blocked on Arabic Wikipedia. As such, I am having trouble getting assistance to appeal my block there. Can someone help me with that? Ogundareibrahim123 (talk) 19:58, 27 July 2024 (UTC)[reply]

This is the English Wikipedia, we have no control over the Arabic Wikipedia. And if you know you don't speak Arabic well, isn't it kind of rude to try to edit there knowing that it will make things more difficult for them? Thebiguglyalien (talk) 04:54, 28 July 2024 (UTC)[reply]

Maybe. I may not be able to make very useful edits, but I was still looking to see if I could fix small things, e.g. errors in links. I first made some test edits in the public sandbox there that may have been controversial, which is probably why I got blocked. But I actually wanted to meaningfully improve the Arabic Wikipedia later, but because they already blocked me I don't have a chance now. Is there a way to appeal their block if I don't speak Arabic? If so, how? Ogundareibrahim123 (talk) 06:33, 28 July 2024 (UTC)[reply]

@Ogundareibrahim123: You can communicate with the admin who blocked you as I did when I faced a similar situation on the Chinese Wikipedia (in that case I emailed them, but pinging would have worked too; the manual method of adding a user page link is universal and {{ping}} works everywhere, including in Arabic. The user who blocked you, Dr-Taher, is quite active here per their English user page. Graham87 (talk) 11:50, 28 July 2024 (UTC)[reply]

We here at the English WP can’t unblock you from a sister project. They have their own set of admins, and you need to talk to them to have the block lifted.

It probably isn’t a good idea to edit in languages you don’t understand, but if you must… my advice would be to leave a message on your Arabic userpage (either in English, or using machine translation), requesting the assistance of someone who understands English… perhaps they can help you sort things out.

Otherwise, just be patient and wait until the block expires… then stay away from anything even remotely controversial. Blueboar (talk) 12:19, 28 July 2024 (UTC)[reply]