Andj C

So a printer's ornament, and only an emoji when followed by VS16.

The equivalent using italic math symbols would require a separate keyboard for the italic symbols, and switching keyboards when you need to switch between regular letters and math italics. It would mean spell checking wouldn't work for something in italic symbols. It becomes difficult for the user.

Think in terms of a word processor, you type letters on keyboard, select text and make text bold or italic, this is changing to the bold or italic font within the font family being used.

Some people use mathematical symbols available in Unicode to visually simulate italic letters. There are not true italics, they are displayed with maths fonts, not text fonts, they only support English, not most other languages. The glyphs are designed to look good in equations not body text.

Italics or obliques are one of the typefaces within a typeface family. The characters within an italic font are the same characters as within a bold, bold italic or regular typeface. Probably easier to think of it as a calligraphic style.

The issue is that devs haven't introduced rich text for messaging ... Not to mention on most devices what you really need is obliques not italics due to prevalence of sans-serif as default fonts.

For Unicode, italic would be a presentational feature of a glyph.

except such non-consent codepoint has no legal framework to make it binding.

Considering copyright and licensing is often ignored in scraping and pulling together datasets, just one more thing for them to ignore.

The holy grail is bidi and complex rendering in a terminal.

And after that the SIP and then the TIP

I prefer to control normalisation at each IO boundary. In theory, the APIs should be handling it, but rarely do. And it's unfortunate that the APIs can't handle combing Jamo sequences.

Probably safer to normalise to NF(K)C. But in theory there shouldn't be any compatibility Jamo in data, so NFC out bound should be fine. But if Jeju or archaic syllables ... Them combining Jamo are unavoidable.

In other words poorly implemented and designed APIs from an i18n perspective.

Sounds like fun, something odd in the downstream code I assume.

But focusing on graphic design and being able to make custom adjustments to open source fonts, extending them for specific projects seems to be a valuable tool

Actually there was a larger of sarcasm there, I like my humour very dry. And 2a is one of those weird American things. Font design as the primary gig is a difficult field to break into.Seems to be more openings as a font engineer these days.

Unicode Consortium? Or rather Apple, Google, X, Microsoft and a number of small, independent font designers?

Gecko (Firefox) uses a different font fallback mechanism than Blink (Chrome/Chromium/Edge) so when fall back is occurring it's probable that they will use different fonts.

Blink font fallback failing to find suitable font on that platform. Other than the text font you'd need 2-3 other fonts as well. Probably failing at the math font.

An anthromorphic lion?

Greek and mathematical symbols opens up all sorts of complexities.

You can use the math symbols to typeset Fraktur ..
Too many missing characters, and a math font would have the required ligatures.

What I meant by standard German letters is that in Unicode Fraktur is presentational, ie it's a typeface German Fraktur font uses the exact same code points as a serif or sans serif typeface.

But those fonts would require more than 159801 glyphs since some characters need more than one glyph.

There is German text set in a Fraktur font, ie standard German letters. And there are the mathematical symbols that look like Fraktur letters.l which are intended for mathematical equations. You can write in these symbols.

In terms of OCR it's a question of training on sets of German text in. Fraktur font. Digital Fraktur is standard German letters displayed in a Fraktur font.

Interesting set of "existing" tools to benchmark against. I'd would have preferred testing against skeleton strings which is likely to include things not caught by transliteration or normalisation.

I think you'll need to blame the ASCII standard for that one, when they replaced the up arrow with what they called a circumflex in 1963.

Full list for Delta: ⵠ ᐃ 𖼚 𝞓 ∆ Ⲇ 𝛥 𝚫 𝜟 △ Δ 𐊣 🜂 𐊅 𝝙