Post A8ORGuS5ouDBB3ogvA by TerrorBite@meow.social
(DIR) More posts by TerrorBite@meow.social
(DIR) Post #A8OPZbjjQbu4j4r5Si by Elizafox@mst3k.interlinked.me
2021-06-17T23:54:42Z
1 likes, 2 repeats
I'm going to fart in programming church and keep it real: Unicode is a God damn mess.
(DIR) Post #A8OPZcHPPOqGPWbyxE by Elizafox@mst3k.interlinked.me
2021-06-17T23:58:14Z
0 likes, 2 repeats
Like I know there is nothing better, I'm not mad at things like CJK ideograph unification (duplicating thousands of similar characters for what essentially amounts to political reasons is no way to run an encoding standard) but UTF-16 is a horrible bodge, 21-bits is a horrible bodge to support UTF-16, and the "round-trip compatibility" requirement has bloated the character set and caused countless homoglyph issues all for what? Sops for ancient encodings?
(DIR) Post #A8OQ45MhpmURIycNyC by Elizafox@mst3k.interlinked.me
2021-06-18T00:03:05Z
0 likes, 1 repeats
I know UTF-8 is a hack that was basically required to be compatible with Unix (Unix was never going to gain Unicode support without it, because all Unix knows is bytes... for worse, really, but that's another story). But my Gods, it is annoying.The sheer amount of security issues created by the fucking encoding system itself is mind-boggling. Non-fixed width encodings are a mess, and decoding them has created countless headaches (a lot of them due to UTF-8 and UTF-16). The homoglyph problem is easily managed enough but it took far too long for anyone to recognise the problem. Normalisation is also shitty (choose one of four amazing normalisations, none of which probably do quite what you want).And there is still no standard way to handle Unicode in C or C++, the two most used languages in the world. ICU is the closest thing to a standard we got, but it's huge.
(DIR) Post #A8OQR94kCnnrTNDEVU by Elizafox@mst3k.interlinked.me
2021-06-18T00:08:17Z
1 likes, 1 repeats
Also, the Unicode standard includes a LOT of useless things. I'll be honest, it's too big and bloated with unnecessary cruft.When was the last time YOU had to write anything with Hieroglyphs? (history majors need not answer). It makes a headache for implementers, the people who have to draw the fonts (what do you think the fonts just fall from the sky?), all for what... something that could have been handled in LaTeX... like it was before?
(DIR) Post #A8OQax6XZIVFF0jFia by Elizafox@mst3k.interlinked.me
2021-06-18T00:09:36Z
0 likes, 3 repeats
Oh I'm sorry, no, they included hieroglyphs so you could do this I presume: 𓂺 /sNext time you see a proposal rejected from Unicode for what is essentially corporate reasons, just remember there is a literal dick (well, several) encoded in Unicode.
(DIR) Post #A8OR0MeLjENthxeza4 by Elizafox@mst3k.interlinked.me
2021-06-18T00:11:43Z
1 likes, 2 repeats
Yeah. That's another problem I have. A bunch of suits are now the arbiters of the way we communicate, a committee of largely Western self-appointed experts. They're trying to make a universal script, and they get to decide what goes in it and what doesn't. A bunch of people who you've never met, who you'll never know, all with their own little interests, sensibilities, etc. And that's why you will never get things like a frowning pile of poo, yet you get a fucking dick in Unicode.
(DIR) Post #A8ORGuS5ouDBB3ogvA by TerrorBite@meow.social
2021-06-18T00:17:27Z
1 likes, 2 repeats
@Elizafox They added hieroglyphs but refused to add Tengwar and Klingon
(DIR) Post #A8OWp1R2rZG0S4PbpQ by portpupper@social.sakamoto.gq
2021-06-18T01:20:24.902228Z
0 likes, 0 repeats
@TerrorBite @Elizafox I’ll give a shit about proper Unicode support once they add Tengwar. Less facetiously, I once forced a project to migrate to python3 by spamming the bug tracker with unexpected behavior when I’d slide in emojis and Zalgo text into the input fields.