Wikipedia talk:Manual of Style/Superscripts and subscripts: Difference between revisions
Content deleted Content added
Kys Tags: Reverted Mobile edit Mobile web edit |
|||
| (16 intermediate revisions by 5 users not shown) | |||
Line 15:
If this page to be used, it should itself be made MOS-compliant with regards to the accessibility guidelines, namely [[WP:COLOR]] - the "correctness" of some of the examples is indicated by colour alone, and note that coloured text should really only be used if there's a pressing reason to do so. [[User:Knepflerle|Knepflerle]] ([[User talk:Knepflerle|talk]]) 16:35, 17 August 2010 (UTC)
:If you were refering to the comparison in the lead, it's now fixed. [[User:Jonkerz|jonkerz]][[User talk:Jonkerz|♠]] 09:45, 19 August 2010 (UTC)
== Apply to article title? ==
Do these guidelines apply to article titles? Could it explain why (or why not)? Thank you. Regards, [[User:RJHall|RJH]] ([[User_talk:RJHall|''talk'']]) 21:30, 12 July 2012 (UTC)
:Done use the characters like "²" in the title, but don't use <nowiki><sup><sub></sub></sup></nowiki> tags either, instead use <nowiki>{{DISPALYTITLE:}}</nowiki> to make it look right. [[User:Graeme Bartlett|Graeme Bartlett]] ([[User talk:Graeme Bartlett|talk]]) 11:28, 9 September 2019 (UTC)
== Lost semantic integrity ==
Line 21 ⟶ 25:
:Unicode causes many complications, for example in my browser I cannot type those suffix Unicode characters and I only know to copy and paste them from elsewhere. Also searching for "2" does not find "²". The Wikipedia search treats "²" as if it didn't exist, though a regex search can spot it. [[User:Graeme Bartlett|Graeme Bartlett]] ([[User talk:Graeme Bartlett|talk]]) 11:26, 9 September 2019 (UTC)
:: Thanks for your reply. Of course we can't have all thousands of [[Unicode]] characters on our keyboards, but they are available by many other means, most ostensibly the insert section at the bottom of this edit window. And searching for "2" (two) should not find "²" (square), they do not have the same meaning. The Wikipedia search should be fixed, but it is a problem of its own.--[[User:Marc Lacoste|Marc Lacoste]] ([[User talk:Marc Lacoste|talk]]) 13:26, 9 September 2019 (UTC)
::If I search for <code>2</code> I don't ''want'' to match <code>²</code> or <code>₂</code>, or vice versa. The point of a human performing a textual search is usually an approximation to find a semantic match; we already know whether we want an ordinal, or a quantity, or an exponent, or a footnote, or an index. On the rare occasion that I don't know which (or want to find them all), I simply search for them all: <code>[2²₂]</code>.
::If other humans don't want that behaviour, then the search tool can be adjusted, to treat superscript, subscript, & plain numerals as equivalent when "case insensitive" is selected.
::[[User:Martin Kealey|Martin Kealey]] ([[User talk:Martin Kealey|talk]]) 02:55, 6 August 2025 (UTC)
More in [[Unicode subscripts and superscripts]]. So, the font designers are dumb, but we continue to use <nowiki><sup>/<sub></nowiki> hacks instead of switching to a more sane font?--[[User:Marc Lacoste|Marc Lacoste]] ([[User talk:Marc Lacoste|talk]]) 13:34, 9 September 2019 (UTC)
Line 32 ⟶ 40:
::::We're ''still'' encountering ''new'' page summaries and audio page readers that claim such nonsense as that the mass of the sun about 2 tonnes, specifically 1.9889×1030 kg, and it's our fault, not those devices. For further discussion see [[#Different guidance for non-HTML rendering]] below.
::::[[User:Martin Kealey|Martin Kealey]] ([[User talk:Martin Kealey|talk]]) 08:05, 5 August 2025 (UTC)
::::''“2 does not differ in meaning depending on its position”'' is ... absurd. Of ''course'' a digit's meaning depends on context; 20 is ten times bigger than 2; ¹⁄₂ is twenty-four times smaller than 12; and 5² is about half of 52; O₂ is a molecule, O²⁺ is an ion, ²He is an element, and ₂H is a nuclide. I am always frustrated when searching for, say, a "squared" exponent, that I have to wade through masses of false matches for other kinds of <code>2</code>. For me, using distinct superscript digits would ''improve'' the specificity of searches.
::::I assume you're a sane human who knows all this, so if "meaning the same" isn't the real point, what were you actually trying to say? Does the lack of specificity when searching simply not bother you? [[User:Martin Kealey|Martin Kealey]] ([[User talk:Martin Kealey|talk]]) 03:30, 6 August 2025 (UTC)
== Add exception to allow Unicode super/subscripts in COinS fields in {{tl|cite xxx}} templates? ==
Line 57 ⟶ 67:
== Different guidance for non-HTML rendering ==
Whilst Wikipedia only provides HTML rendering, in some cases that will be subject to further reduction to plain ASCII. Examples include: screen readers, web crawlers,
Where a client can be identified as preferring "plain text", we should ''avoid'' sending “10<sup>30</sup>” because that produces
''Please'' can we recommend, in this specific case,
Just to be pellucidly clear, I'm talking about what templates generate, not what editors write in wiki markup, ''except'' when they're editing the templates themselves.
[[User:Martin Kealey|Martin Kealey]] ([[User talk:Martin Kealey|talk]]) 07:46, 5 August 2025 (UTC)
:Perhaps a sufficient change would be to ''strongly'' recommend <code><nowiki>{{sup|</nowiki><i>exponent</i><nowiki>}}</nowiki></code> over <code><sup><i>exponent</i></sup></code> and <code><nowiki>{{sub|</nowiki><i>index</i><nowiki>}}</nowiki></code> over <code><sub><i>index</i></sub></code>, so that those templates can take care of rendering as non-HTML Unicode when indicated by the HTTP client. <!-- Template:Unsigned --><small class="autosigned">— Preceding [[Wikipedia:Signatures|unsigned]] comment added by [[User:Martin Kealey|Martin Kealey]] ([[User talk:Martin Kealey#top|talk]] • [[Special:Contributions/Martin Kealey|contribs]]) 04:15, 6 August 2025 (UTC)</small>
::Although {{t|sup}} doesn't do that. Are you proposing to change that template and similar ones to render non-HTML queries differently to HTML? [[User:Lithopsian|Lithopsian]] ([[User talk:Lithopsian|talk]]) 15:22, 6 August 2025 (UTC)
▲Where a client can be identified as preferring "plain text", sending “10<sup>30</sup>” produces a ''deeply'' unsatisfactory end result like “1030”.
:::In essence, yes, but that's not a MOS issue.
:::[[User:Martin Kealey|Martin Kealey]] ([[User talk:Martin Kealey|talk]]) 22:23, 6 August 2025 (UTC)
{{talk-reflist}}
▲''Please'' can we recommend, in this specific case, using either Unicode superscript and subscript digits with fraction-slash,<ref>(superscripts U+2070, U+00b9, U+00b2, U+00b3, U+2074 thru U+2079; subscripts U+2080 thru U+2089; fraction-slash U+2044)</ref> or insert circumflex/caret “^” symbols for exponents and solidus “/” for fractions. [[User:Martin Kealey|Martin Kealey]] ([[User talk:Martin Kealey|talk]]) 07:46, 5 August 2025 (UTC)
| |||