Talk:Diacritic

The short explanation of why has been necessary to use span style="font-family: serif" in each case where the diacritic is being demonstrated against a dotted circle (as "neutral grey background") is to mitigate a rendering limitation in Android (as of v13), that its default sans font fails to render "dotted circle + diacritic", so visitors just get a meaningless (to most) [X] mark. For the longer explanation, see template talk:Unichar#Combining diacritics are displaying as tofu on Android - fault may be in cwith= handling?. JMF (talk) 15:19, 24 September 2023 (UTC) Reply

Can anyone explain why the section Diacritic#Diacritics that do not produce new letters exists? Is there any reason not to merge it into Diacritic#Languages with letters containing diacritics? As far as I know (which is not very far at all) the only case of where a letter+diacritic is considered unique and distinct for sort order, is ñ in Spanish (as is or was ll). If we really need to go into that detail (and I question that too), surely the section should be about those few exceptions. JMF (talk) 17:50, 17 March 2024 (UTC) Reply

Per WP:BRD, I reverted a bold edit by user:2601:c6:d200:e9b0:6136:61c3:9ee3:bb8c . Per WP:BRD, the change is suspended pending debate.

The long-standing pre-existing arrangement was to use the letter ⟨a⟩ as base letter as much as possible. If it is to be changed, a discussion and consensus is needed. The dispute is whether to use the same base letter so that the emphasis is on the diacritic, or to choose a letter which "best" (?) displays the diacritic in use.

(The section deals with precomposed characters, so the option to use the generic place-holder symbol U+25CC◌DOTTED CIRCLE is not really relevant – and would require a lot more work to use combining diacritics.)

The debate is now open. JMF (talk) 15:02, 18 June 2024 (UTC) Reply

I am not really fond of all examples in a particular example being of the same idea. Can we try to keep them all different, sticking with my version for now? 2601:C6:D200:E9B0:6136:61C3:9EE3:BB8C (talk) 15:18, 18 June 2024 (UTC) Reply

Thank you for engaging with the BRD process.

Sorry but "I am not really fond" is not a very rational argument. In your edit summary, you wrote Using different letters for most examples would flow better as some letters flow better with some diacritics than they do with others. Can you elaborate, using examples (bearing in mind that all the examples based on the letter ⟨a⟩ are real code-points, not artificial constructions?) [Meanwhile, WP:STATUSQUO applies.] --JMF (talk) 15:55, 18 June 2024 (UTC) Reply

Well, I'd like to point out that the base letter ⟨u⟩ has combining forms with all of the diacritics with examples listed expect for the overdot and undercomma. Meanwhile, ⟨a⟩ doesn't combine with the double acute and undercomma, ⟨e⟩ and ⟨i⟩ both don't combine with the double acute, overring, and undercomma, and ⟨o⟩ doesn't combine with the overring and undercomma. So maybe we should be using ⟨o⟩ as our base letter example, which is good because it also has ⟨ø⟩, ⟨ơ⟩, and ⟨ǫ⟩ for the slash overlay, horn, and ogonek, respectively. 2601:C6:D200:E9B0:A4ED:D8FB:CD17:1710 (talk) 13:49, 21 June 2024 (UTC) Reply

Yes, that would equally satisfy the need for a common base letter in all (or at least almost all) cases to that the focus of attention is the diacritic not the base. And yes, ⟨o⟩ is a better choice than ⟨a⟩ for the reasons you set out. Would you like to go ahead and do that change? --JMF (talk) 19:41, 21 June 2024 (UTC) Reply

This article often use different formats for letters or accented characters:

• ⟨ö⟩ shows ö in angle brackets, often for visual clarity in lists.
• ö links directly to the article ö.
• ö (or even ö ) uses double quotes, typically for code or technical text.

Suggestion for consistency: When the character has an article (like ⟨ö⟩), it's clearer to use ö to allow readers to click through.

⟨⟩ can still be used where linking is not needed (e.g., in abstract lists or phonetic examples). Milan Berta 05:35, 12 May 2025 (UTC) Reply

Oppose

• In a section about a letter as a letter (compare Words as words and WP:WAW), it is conventional to use angle brackets to 'isolate' it for analysis. This usage is widespread and must be retained.
  • For diacritics in isolation, as they are usually tiny (and they are glyphs, not graphemes), our local {{char }} is the best option, ` (grave).
    • (The 'correct' answer is to use U+25CC◌DOTTED CIRCLE with the combining diacritic form but that is maybe too difficult for most editors unless they are using {{unichar }}. That will enable U+0300◌̀COMBINING GRAVE ACCENT .)
• When a letter is wlinked, it is underlined on most systems (maybe not on iMacs?). This is intolerable when the topic is diacritics: is the underlining a diacritic or just noise? will it obliterate a real 'below baseline' diacritic? Take even a widely known example, U+00E7çLATIN SMALL LETTER C WITH CEDILLA and look at the effect of ç. On unfamilial graphemes, which in some contexts may have a diacritic above and below, how is the reader expected to know which one is being used? This method should never be used when diacritics are significant. When linking is needed, apply the link to the description of the letter, not the letter itself.
• If the context requires italics, then there is nothing wrong with using the "two apostrophes on either side" Wikipedia markup. If it is ok to write o then it is equally ok to write ö. (As already explained, it is generally sloppy and bad practice to write ö, with or without italics.
  • [I assume that this is what you meant when you wrote "double quotes". If you really did mean "ö", please explain, because again if it is ok to write "o" then it is equally ok to write "ö" – except that it is usually wrong in any case!]

--JMF (talk) 12:51, 12 May 2025 (UTC) extended --12:57, 12 May 2025 (UTC)Reply