Jump to content

User talk:Sean.hoyland

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia
Khirbet 'Ein Karzaliyah (Arabic: خربة عين كرزلية), Jordan Valley: December 2013 - January 2014
Id'eis (Arabic: ادعيس), Jordan Valley: May 2014

EC protecting ARBPIA articles

Now that ArbCom agrees that PIA articles should be EC-protected by default, I made a quarry script to find all articles with an ARBPIA template but no EC-protection. See https://quarry.wmcloud.org/query/90245 . There are over 2000 hits so I'm not sure what it will be good for. One thing I didn't figure out is how to distinguish the cases where the template has the "relatedcontent=yes" attribute. Can you see how to do that? Zerotalk 13:42, 27 January 2025 (UTC)[reply]

I'll have a look, but I think we would need to pull talk page text and search that. There may be another issue though. The template:Contentious_topics/Arab-Israeli_talk_notice can be used in both cases, where ARBPIA coverage is full or partial e.g. Talk:Eurovision Song Contest 2024 Sean.hoyland (talk) 14:05, 27 January 2025 (UTC)[reply]
Looks like it's possible. I'll have a proper look tomorrow. I can pull the top section of the talk pages using the API and look for relatedcontent=yes. A couple of examples from the 49 rows where pg.page_title like 'Be%'
article page_id,page_title,template,talk page text
48163910,[[Bella Hadid]],ArbCom Arab-Israeli enforcement,{{ArbCom Arab-Israeli enforcement|relatedcontent=yes}}
171275,[[Ben & Jerry's]],ArbCom Arab-Israeli enforcement,{{ARBPIA|relatedcontent=yes}}

Sean.hoyland (talk) 17:38, 27 January 2025 (UTC)[reply]

Now I understand what you wrote above: Template:Contentious_topics/Arab-Israeli_talk_notice always says "parts of this article". It doesn't even have a parameter for partial vs full coverage. By experiment it ignores both relatedcontent and section (the latter is in Template:Contentious_topics). Was that ever discussed? If this is intentional, it is a waste of time trying to mechanically separate partial and full coverage. Zerotalk 05:38, 28 January 2025 (UTC)[reply]

I don't know the history of that template but that's why I avoid it. I think it's probably still worth identifying the relatedcontent cases because the vast majority of pages use the 'ArbCom Arab-Israeli enforcement' template or things that redirect to it, 1999 vs 88 for the unprotected pages. It's easy enough to produce the results. I just need to make sure I do it in a way that doesn't hit an API rate limit. Sean.hoyland (talk) 06:27, 28 January 2025 (UTC)[reply]

See User:Zero0000/sandbox2. I can remove the ones indicating "related content" but first I'm waiting for a response to Wikipedia:Arbitration/Requests/Clarification_and_Amendment#Clarification_request:_Palestine-Israel_articles_5. Zerotalk 07:10, 28 January 2025 (UTC)[reply]

I have a little bash script that finds those without "relatedcontent=yes" by loading the source using urls like https://wiki.riteme.site/w/index.php?title=Talk:Hamas&action=raw . It's slow but this only has to be done once. Zerotalk 07:28, 28 January 2025 (UTC)[reply]

Going via the API is pretty slow too. It's running now, with pauses every 100 calls. Let's see how long that takes. Sean.hoyland (talk) 07:51, 28 January 2025 (UTC)[reply]

I found 1977 out 2087 without relatedcontent=yes. Zerotalk 08:32, 28 January 2025 (UTC)[reply]

This is what I get. 23 minutes. I've included the article page_ids just in case that helps. I get one more without relatedcontent=yes I think. Sean.hoyland (talk) 08:45, 28 January 2025 (UTC)[reply]
Our results agree except I counted the template on Talk:Martha Pollack that is down the page and you didn't. I updated User:Zero0000/sandbox2. Zerotalk 11:17, 28 January 2025 (UTC)[reply]
Right. Still, it's encouraging that there was only one case where the template was not in section 0 of the talk pages. Sean.hoyland (talk) 12:44, 28 January 2025 (UTC)[reply]
There are some trickier issues. Many articles don't have a PIA template and it's not clear how to address that. Have a look at the 'United Nations Security Council resolutions concerning the Arab–Israeli conflict' category and its 2 subcats, a simple example where traversing the category tree doesn't cross the PIA - not-PIA boundary. Squares are categories. Circles are pages. Blue is EC protected, red is unprotected. The yellow and gray circles are talk pages. Gray is templated, yellow is untemplated. I think this kind of issue is quite widespread in the topic area. Sean.hoyland (talk) 09:58, 28 January 2025 (UTC)[reply]
That's a lot of extra articles, suggesting that the real total is substantially higher. Is there a way to bring up all pages in category "Arab–Israeli conflict" and its subcategories however deep? Would it make sense even? Zerotalk 11:17, 28 January 2025 (UTC)[reply]
Yes, it's possible, but I think you pretty rapidly hit a "six degrees of separation"-like issue and exit the topic area in a lot of branches. For example, for levels 0 to 5 with root Category:Arab–Israeli_conflict, the page+subcat counts go 72, 1011, 4801, 12988, 30113, 59382. Sean.hoyland (talk) 11:50, 28 January 2025 (UTC)[reply]

An idea that occurred to me, since there are so many pages, is to start by requesting EC-protection of those which have had a non-EC edit in the past X months. Zerotalk 11:17, 28 January 2025 (UTC)[reply]

I was hoping someone already something to mass-protect pages, as long as they have a list. Sean.hoyland (talk) 11:50, 28 January 2025 (UTC)[reply]
Right, someone should have a tool. However I'm involved™ so I can't do it. Zerotalk 12:02, 28 January 2025 (UTC)[reply]
I sent Dr. vulpes a list of 2180 titles. Zerotalk 01:46, 30 January 2025 (UTC)[reply]
Great. Looks like they've already started. Sean.hoyland (talk) 05:01, 30 January 2025 (UTC)[reply]

Question

Hello, do you happen to know of any tools or methods to find when or in which edit a specific bit of content was added to an article? IOHANNVSVERVS (talk) 13:01, 30 January 2025 (UTC)[reply]

There's WikiBlame. That can usually find it. Sean.hoyland (talk) 13:12, 30 January 2025 (UTC)[reply]