<html><body><div class="gmail_quote">
<div dir="ltr" class="gmail_attr">On Nov 2, 2023 at 12:11:51, Ken Moffat via tex-live <<a href="mailto:tex-live@tug.org">tex-live@tug.org</a>> wrote:<br></div>
<blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex" type="cite">
In Small Caps, there are some fonts where<br>latin i and dotless i both lack a dot, making the small caps useless<br>in e.g. turkish.<br>
</blockquote>
</div>
<br>
<div dir="ltr">
Not necessarily: I would encode all Turkish text using ı and ı̇ (the latter being 0x0131+0x0307 rather than 0x69). (I would note that the recommendation for case-folding is to case fold on text in NFD form which, unfortunately, won’t automatically transform 0x69 to 0x0131+0x0307 since normalization is not language dependent, but Unicode stability means this will never change and software may need to enage in some manual string manipulation for Turkic languages).</div><div dir="ltr"><br></div><div dir="ltr">Small caps are handled through glyph substitution (yes, there are code points for 25 of the 26 latin letters scattered through IPA extensions, phonetic extensions and Latin extended-D, but those should not be used as small caps directly as they greatly impair accessibility) and knowing whether a small caps i should be rendered as I or İ would require knowledge of the language that most software would not have). The lack of a semantically distinct i-with-dot for those languages that have the pairs I/ı and İ/i is a weakness in Unicode that often shows up in surprising places (e.g., older versions of PHP which break when the language is Turkish thanks to language-specific case-folding being applied to identifiers, just one of many bad design choices in that language).</div><div dir="ltr"><br></div><div dir="ltr">-dh</div></body></html>