Wikipedia talk:Splitting
This project page does not require a rating on Wikipedia's content assessment scale. It is of interest to the following WikiProjects: | |||||||||||
|
Semi-protected edit request on 6 October 2023
Please change the month for template dates from September 2023 to October 2023. 99.209.40.250 (talk) 15:19, 6 October 2023 (UTC)
- Fixed: I'm not sure why, but the use of
{{#time:F Y}}
was still producing "September" in this page, and changing it to{{CURRENTMONTHNAME}} {{CURRENTYEAR}}
fixed it. — SMcCandlish ☏ ¢ 😼 15:34, 6 October 2023 (UTC)
Splitting categories
Is it possible to automatically split categories? I wish I could do this using {{category redirect}}, but I don't think it can redirect a category to more than one other category. Jarble (talk) 16:59, 4 November 2023 (UTC)
Numbers and changes
There's been a really long series of discussions at WT:SIZE about what critieria, on what rationales, should actually be used, to arrive at what modern size recommendations. This page doesn't seem to be keeping up with the (presently maybe nebulous) consensuses forming over there and changes to WP:SIZE.
In particular, here at WP:SPLIT, I see "> 15,000 words (> 100 kB) – Almost certainly should be divided or trimmed." But this seems like a much too low pair of numbers. We have many, many, many articles far larger than that, and they are almost always already split and full of sections that start with {{Main}}
or {{Further}}
because numerous side-articles have been spun off of them already. Almost any article on a country or major city is such an example. Los Angeles is 236K. Munich is 184K. Australia is 221K. Even Botswana is 145K, despite not being a country that attracts so much attention from English-speakers in the West and Global North.
While this page maybe need not change right this second, it is clearly badly out-of-step with actual practice, and whatever the numbers are changed to should be consistent with what is determined at WT:SIZE. — SMcCandlish ☏ ¢ 😼 09:01, 27 November 2023 (UTC)
- You're making the very common mistake of getting confused between total size of page (in history) and readable prose of page. The guideline is about readable prose, and therefore much less strict. For instance, Botswana is 9,364 words and 60k of readable prose, and falls squarely within the guideline (see xtools). Similarly Los Angelos has 9,596 words and 60k readable prose, and not much wrong with it according to the guideline. This is why there is a current discussion to omit the kb. Putting it in kb of readable prose is confusing given that article history displays something else. —Femke 🐦 (talk) 12:53, 27 November 2023 (UTC)
- Agree with Femke re common mistake here, and that it's unhelpful to split this discussion from a guideline page to an essay page. SandyGeorgia (Talk) 13:25, 27 November 2023 (UTC)
- That is asking the wrong question. Keep as guideline, but correct the numbers, and avoid use of 'prose size', as it leads to incorrect assessments about article size wrt splitting. See § below. Mathglot (talk) 01:00, 25 February 2024 (UTC)
The term 'prose size'
Related to this discussion, is the issue of the use of the term prose size in the column headers, or in the body. The term prose size has a specific meaning that is not the same as word count; it is calculated by various tools such as Xtools articleinfo, which is linked from the top of every history page.
The term prose size cannot be used in isolation as the sole condition of evaluating page size for possible splitting; it can only be used in combination with other terms, such as "prose size > X OR word count > Y" or perhaps: "prose size > X OR raw pagesize > 300kb". Otherwise it leads to absurdities such as the article [[List of Glagolitic manuscripts]] (not linked; I'll explain why in a second), whose prose size is 45 words. So for that article to be in "probably-split" territory under current conditions, it would have to grow from 45 to 9,000 words, i.e., a 200-fold increase. However, extrapolating from current size, that would result in an article of 138 mega-bytes, as its current raw size is already 690kb (that's why I didn't link it above). Clearly, it is absurd to evaluate that article for splitting on the basis of prose size alone.
In my opinion, we should not use the term prose size (or implied equivalents such as readable prose, etc.) on this page at all, but if we do, it should be accompanied by a note explaining that it can be radically different from the actual number of readable words on the page. (My software counts 32,790 words on the rendered page above the "See also" section, not 45 words.) Since the current version is untenable, I've restored SMcCandlish's rev. 1187194448 of 00:27, 28 November 2023, which mentions word count without alluding to prose size, and which leads to the proper conclusion when evaluating List of Glagolitic manuscripts and similar articles. Mathglot (talk) 00:50, 25 February 2024 (UTC)
- I wasn't actually aware of this technical problem, but it does seem to be, well, problematic. — SMcCandlish ☏ ¢ 😼 01:20, 25 February 2024 (UTC)
Splitting an article into two new titles
Seeking quick advice: I'm seeking to split an article into two articles with two new titles and leave behind a DAB at the original title (per this proposal). Is there a recommended way of doing this with regards to preserving the old edit history? (E.g. instead of leaving the history behind at the DAB, should I move the current article to one of the new titles first and create the other article from scratch? Or is there a way to copy the edit history to both new titles?) Thanks, R Prazeres (talk) 02:17, 18 July 2024 (UTC)
- PS: I've gone ahead and followed my own suggestion above (see Talk:Anatolian Seljuk architecture), but if anyone sees this belatedly and thinks there was something wrong with this method, let me know. R Prazeres (talk) 20:47, 20 July 2024 (UTC)
Splitting procedure needs a re-write
As a documentation writer of long experience, I know a poor procedure from a good one, and the Splitting Procedure in the WP:Splitting page qualifies as the former.
- Some steps - for example, Step 1 and Steps 8-9 - are not steps but either conditions or options that should be set apart to shorten the main procedure.
- Most steps need to be rewritten/restructured. For example, Step 3 should begin "In a new browser window or tab, open the section or full article", followed by bullets for each action or by new short-and-to-the point steps.
- Step 5 is decent in having a short intro followed by a series of bulleted actions. An alternative, though, would be to move Cleanup to a subsection below the main procedure.
The general principle here is simplify-simplify-simplify, which leads to rules on wording and structure. For example, keep sentences and steps as short as possible. Place "locators" at the beginning of steps (as in the Step 3 recommended above). If a procedure becomes too lengthy, break it into a series of sub-tasks.
I'll gladly re-write the procedure at an admin's request and then post it in a sandbox for reviewing and further editing. Allreet (talk) 14:43, 1 November 2024 (UTC)
- Thanks, but please don't. This procedure may not follow the rules you're familiar with but it's very effective and easy to follow. In particular, the fact that it's all in one ordered section is a strength, not a weakness. If it were split into multiple sections, people wouldn't read them all. If you want to make incremental changes like rewording Step 3, sure, that's the way Wikipedia progresses, but a rewrite is not indicated. Dan Bloch (talk) 18:48, 1 November 2024 (UTC)