Category talk:Cyrillic upper- and lower-case pairs with font lines

Special cases in font line sets

Copied from User talk:Watchduck.
Hello, thanks for creating the set categories. I would create similar categories for the Latin script if you don‘t mind (as well as container categories where appropriate). However, one issue I found is that Category:Cyrillic letters on 7 font lines (serif with big margin) should, by its literal scope, also contain File:Cyrillic letter multiocular O (2).svg and File:Cyrillic letter unblended uk.svg, even if they do not depict an upper- and lower-case pair (similarly for File:Cyrillic letter Multiocular O.svg with Category:Cyrillic letters on 5 font lines (serif with big margin)). This might seem insignificant at first; however, for Latin letters, there are much more cases of this. The question is whether the upper- and lower case pair subsets should be split up from the set categories, in which case the Cyrillic set categories would contain that subset and two files (I don't think that would be problem), or not, in which case the non-set categories for upper- and lower case pairs would still be crowded. I think I'm leaning towards the first option. 𝟙𝟤𝟯𝟺𝐪𝑤𝒆𝓇𝟷𝟮𝟥𝟜𝓺𝔴𝕖𝖗𝟰 (𝗍𝗮𝘭𝙠) 21:07, 6 January 2021 (UTC)

@1234qwer1234qwer4: Hi, I just created a single letters subcategory for these two files, but then I realized, that the uk does not belong there, because it is a different (sans-serif) typeface. (The ꙮ with 5 lines does not belong to the other set for the same reason.) For now I just show the ꙮ with 7 lines in the set, instead of including it. Anyway, in general I think the creation of a single letters subcategory is the practical solution in such cases, even if the whole set is in an upper- and lower-case pairs category. It is in the nature of image sets, that they can never have perfect names. So unless there is a flood of single letters, I would avoid the "perfect" solution of moving the pairs down to pairs subscategories. Greetings, Watchduck (quack) 10:37, 8 January 2021 (UTC)

@Watchduck: I'd say the font is not really relevant for the image set in this case. Category:Cyrillic letters on 7 font lines (serif with big margin) should contain the images we create at the Russian Wikipedia, and we normally use Universalia or Times New Roman, but in some cases, there is no glyph in that font, or it is particularly bad, so we use a different font in the image, but it still belongs to the same image set. You also say that "the creation of a single letters subcategory is the practical solution" but don't explain why it is supposed to be more practical than obeying transitivity and having a "upper- and lower-case pairs" subcategory instead, which would further have the benefit of having files like File:K with hook - uppercase wrong and right.svg and File:Various Latin chi and stretched x.svg categorised under the appropriate image set (it shows neither a single letter nor an upper- and lower-case pair). See Category:Latin letters with font lines for more examples. 𝟙𝟤𝟯𝟺𝐪𝑤𝒆𝓇𝟷𝟮𝟥𝟜𝓺𝔴𝕖𝖗𝟰 (𝗍𝗮𝘭𝙠) 19:19, 8 January 2021 (UTC)

@1234qwer1234qwer4: My point about the practical solution is simply, that I think it's not a good idea to move many images, just because a few images differ from them. Take an image like the K with red and blue hooks as an example: I would view it as unpractical to now move all the other images to their own monochrome subcategory. On the other hand I am not a fan of just scattering the special cases between a large number of uniform images. So I have just created a special cases subcategory for the two 7-line images you brought up. I think that makes sense until there are more special cases. (When there are just one or two special cases, I usually just show them on the category page without including them.) --Watchduck (quack) 13:22, 10 January 2021 (UTC)

@Watchduck: I am currently only talking about letters with font lines vs. upper- and lower- case pairs with font lines, because that is the only property of the images that is also represented in the non-set categories. Having a sub-set category with the upper- and lower- case pairs would mirror this categorisation style and make letters with font lines that are not upper- and lower- case pairs with font lines as easily findable as they are in the non-set categories. Also, I'm not sure what is so "impractical" with moving a category; it's quite a standard process and is pretty easy to do with Cat-a-lot. @Tuvalkin: Do you have and opinion on this? 𝟙𝟤𝟯𝟺𝐪𝑤𝒆𝓇𝟷𝟮𝟥𝟜𝓺𝔴𝕖𝖗𝟰 (𝗍𝗮𝘭𝙠) 14:06, 10 January 2021 (UTC)

It's not like I am opposed to this. I just think it's not worth the effort. There are not many single letters in Cyrillic letters with font lines, so I don't see the point. (It would be much more useful to sort the 347 images in Cyrillic upper- and lower-case pairs into sets.) --Watchduck (quack) 15:56, 10 January 2021 (UTC)

But there are quite a few at Latin letters with font lines, so creating such categories would make sense there and then, for unification, also for Cyrillic letters. 𝟙𝟤𝟯𝟺𝐪𝑤𝒆𝓇𝟷𝟮𝟥𝟜𝓺𝔴𝕖𝖗𝟰 (𝗍𝗮𝘭𝙠) 15:58, 10 January 2021 (UTC)

I agree that it makes sense for the Latin letters. If you think the unification is needed, I am not opposed to that.

A general point I want to make is, that IMO it is ok to have exceptions in a category. E.g. Speicherstadt is in Gothic Revival brick buildings in Hamburg, because that is correct for the overwhelming majority of buildings there. But I think it would be perfectly fine to have subcategory for non-brick buildings in there, even if that is technically misplaced in the brick parent. This is not mathematics. We are just trying to sort images in a way that makes it easy for human users to find what they are looking for. --Watchduck (quack) 16:23, 10 January 2021 (UTC)

@Watchduck: Thanks. For now, I will create the categories for Latin letters. Most importantly, I wanted to know if such categorisation was okay for Latin letters. 𝟙𝟤𝟯𝟺𝐪𝑤𝒆𝓇𝟷𝟮𝟥𝟜𝓺𝔴𝕖𝖗𝟰 (𝗍𝗮𝘭𝙠) 16:53, 10 January 2021 (UTC)

Done: Category:Letters on 7 font lines (serif with big margin), Category:Latin letters on 5 font lines (serif with big margin). I'm not really sure about the margin; Category:Latin upper- and lower-case pairs with font lines contains quite a few images that probably belong in a set category. 𝟙𝟤𝟯𝟺𝐪𝑤𝒆𝓇𝟷𝟮𝟥𝟜𝓺𝔴𝕖𝖗𝟰 (𝗍𝗮𝘭𝙠) 19:02, 10 January 2021 (UTC)