@mos_8502 Personally, I'd use only Ubuntu Mono or similar but neurotypicals will go apeshit on #monospaced fonts so I think #OpenDyslexic and other #accessibility-optimized fonts would be great...

Also one would likely need a font with all the #unicode characters in cuz otherwise CJKTI* - languaged people will rage about this decision when there's no legitimate reason for it.
https://en.wikipedia.org/wiki/CJK_characters

And lastly if one adds in @libreoffice one has a pretty substantial collection of fonts.

I rebooted myself!

https://wiki.freebsd.org/GrahamPerrin#alum includes a shortlist of revised focus areas.

Can anyone help with the Unicode input regression?

#FreeBSD #Unicode

C'est quoi votre outil/site préféré pour récupérer facilement un caractère unicode un peu complexe ? (un point médian, une flèche, un exposant....)
J'avais un site cool mais il est passé hors ligne. Ensuite un workflow alfred mais il est cassé aussi :(

#unicode #tools

So I've invented a new number system 🙃
It's described on either of 2 websites that technically can host articles(its a bit too complex for 500 characters 🙃):
https://www.deviantart.com/torrapgib/journal/Innovations-in-needless-complexity-981277741
https://odysee.com/@Idontlikenames:7/Innovations_in_needless_complexity_and_the_encoding_of_quantity:d
#math #linguistics #unicode #conceptart #😳

By the way really interesting talk this https://invi.susurrando.com/watch?v=gd5uJ7Nlvvo from NDC Copenhagen 2022 by @dylanbeattie about plain text
#PlainText #ASCII #Unicode #SoftwareEngineering

Building Up Unicode Characters One Bit at a Time - The range of characters that can be represented by Unicode is truly bewildering. I... - https://hackaday.com/2023/09/07/building-up-unicode-characters-one-bit-at-a-time/ #peripheralshacks #truetypefont #codepoint #keyboard #unicode #binary #usbhid #glyph #utf-8

An Exhaustive Test Program for the Legendary Höhrmann UTF-8 Decoder

“The Höhrmann Decoder, implemented as a deterministic finite state machine, needs only a handful lines of C code and 364 bytes for a combined character class and state transition table.

[...]

hoehrmann-utf8-test.c exhaustively tests all possible inputs to the decoder, that makes 269492416 different byte sequences[^1], out of which 1112063[^2] are accepted.”

https://git.sr.ht/~slink/hoehrmann-utf8
http://bjoern.hoehrmann.de/utf-8/decoder/dfa/
#unicode #utf8

Musing about the multi-ocular O ꙮ the thought flitted across my mind that a #Unicode #burlesque show would be a lot of fun. And I write it here so that mine isn't the only imagination this idea now lives in.

@vermaden IBTD ... #Unicode has been around for long enough and the #UTF8 representation became the de-facto standard everywhere. The #FreeBSD #ports tree is not restricted to #ASCII, but anything outside ASCII needs to be UTF8 to ensure interoperability. What you see here is, well, a bug 🙈

Another quiet Friday makes for a good excuse to commune with the Unicode Archives. Today I'm the Unicode 15 update to the Cyrillic code block, which increased the number of eyes in the multi-ocular O character. The legitimacy of encoding this in Unicode at all is frankly a little dubious; it appears once, in one manuscript, in the phrase "many-eyed seraph" (as seen on my dress). But I love it anyways. #DHsewing #Unicode

Serious question to #unicode people: is there a current consensus about how to decode #utf8 correctly in the various error scenarios?

It is 2023 and I wonder: does any software decode UTF-8 correctly?

https://www.cl.cam.ac.uk/~mgk25/ucs/examples/UTF-8-test.txt

firefox, gnome-terminal and kitty all seem to get it wrong, and all differently. Or is the test wrong?

#unicode

Nice little helper for when you see unfamiliar / ambiguous "#characters" in a text and want to see how their #CodePoint is officially called in #Unicode - can you tell an Em dash from an En dash just by looking at it? No matter what the font is?

Also works the other way around, e.g. try searching for "not equal".

https://unicodeplus.com

Normal people: "Final Sigma" sounds cool, it would be a good name for an action sci-fi movie.

People who deal with #Unicode: It would be a horror movie.

BTW one thing I've noticed is that the cards have too much free space around them. I'm using the #Unicode playing cards block so each card is just the corresponding code point wrapped in a span, but by giving the span a visible background I see a lot of bleeding out of the glyph. I tried culling all the extra space I could think of in #CSS, but it still isn't as tight a wrap as I'd like. Do I suck at CSS or this is just impossible to achieve? Maybe the font glyphs are just that spacey?

Great talk about text, encoding and real world problems/solutions.

https://youtu.be/gd5uJ7Nlvvo?si=SnqDHsDMiC0BM4gu

#PikeMatchbox #softwareengineering #unicode

Ich habe nun auch eine Version mit #Unicode in #hexadezimal erstellt.

#Unicode 15.1.0 has been under scrutiny since February and left beta review in July, with a full release scheduled for 12 September 2023

I know I've seen draft revisions before, but I hadn't previously known that there was a way to keep track of what's coming next better than paying attention to UTC meeting minutes/mailing lists/etc.

TIL that there's a #Unicode technical site page dedicated to information about alpha/beta drafts of upcoming releases.

https://www.unicode.org/versions/beta.html

IDK so much about #Unicode rendering, and what bidirectional shenanigans or whatever is involved, but:

ߋ

U+07CB is my nemesis. Given the string "0\u07CB0", Emacs and Firefox render that as (using "o" to approximate "\u07CB") "00o" (maybe correct because bidi), while `git gui` renders it as "0o0" (matching the byte stream). When my code truncated 1 byte off the end of the string, and I saw "0o" instead of "00", I was convinced that there was a bug in Go's memmove!