Hacker Newsnew | past | comments | ask | show | jobs | submit | gschizas's commentslogin

There's no 0x80-0xFF in the UTF-8 encoding. Only up to 0x7F (127) it's the same.

UTF-8 is not technically a character set (because it has way more than 256 characters). Characters 32-127 in UTF8 are the same as ASCII, which is the same as the OEM/CP437 and the ANSI/ISO-8859/CP1252.

The characters in CP437 (and other OEM codepages) actually come from the ROM of the VGA (and EGA/CGA/MCGA/Hercules before them).

What you are referring to is those (visually), right? I'm missing some characters in the first line, because HN drops them.

    0123456789abcdef
   0...♥♦♣♠•◘○◙..♪♫.
   1►◄↕‼¶§▬↨↑↓→←∟↔▲▼

As far as I know, the equivalent control characters (characters 0-31) don't have any representation in CP1252, but that's also dependent on the font (since rendering of CP1252 is always done by Windows)

As to their origin, originally the full CP437 character set was taken from Wang word processors. I don't know where Wang took it from, but they probably invented it themselves.

EDIT: There's a more complete history here: https://www.os2museum.com/wp/weird-tales/

EDIT 2: The CP437 character set didn't seem to come directly from Wang; it's just that they took some (a lot) of characters from Wang word processors character sets. The positions of those "graphic" characters was decided by Microsoft when they made MS-DOS (at least according to Bill Gates).


In my screen there is indeed about thirty icons. When I executed the program on xterm, they were different and when I pasted them on LibreOffice they were again different. And now it seems this shit is also different in every country.

The world is broken.


ASCII 60-63 is just <=>?

You probably mean 28-31 (∟↔▲▼, or ␜␝␞␟)

Unless this is octal notation? But 0o60-0o63 in octal is 0123


The character "μ" itself is called μι or μυ (both pronounced "me"), and is the exact equivalent of Latin "m". For some reason, English speaking countries tend to pronounce it "mew" (they also pronounce "π" as "pie"). But English speaking countries mispronounce so many words it's par for the course anyway.

Pet peeve: I really hate when they replace "μ" with "u". Completely different letters. Of course replacing capital Latin "E" with capital Greek sigma ("Σ") is even worse.

As to the μlauncher: It's anyone's guess as to how the author meant for it to be pronounced. I'd call it "me"launcher or "micro"launcher (micro is also pronounced wrong in English BTW, it's not MY-crow, it's more like mee-CRAW )


"u" has the advantage of correctly rendering basically everywhere, which"μ" does not. My initial attempt to share this was automatically corrected to "Mlauncher" for example. I'm pretty sure this is the reason the one symbol is used so often where it should be the other.

The solution is just to stick to the Latin alphabet, but you can't deny that mixing in a little Greek every now and then is fun.


Android 15/16 does allow you to control notifications even lower, at the level of notification category, but indeed the app must have chosen to use them.

Most apps that are in need of notification control either:

a) bundle everything in one category, from critical notifications without which the app can't fulfill its purpose to "HEY YOU HAVEN'T USED ME IN A DAY, USE ME NOW" spam

b) create a new category for spam every time they feel enough users have turned off the previous one, which is often


Companies do it with email unsubscribe categories to, which is skirting laws for sure.

I wish app review checked for this. It's hostile anti user behavior.

OEMs like Samsung turn off that ability by default unless you toggle it in "advanced settings". Thank you Samsung!

Yeah the problem is a lot of apps dump their ads in the wrong category if they even use categories :(

3.5" floppies are still floppy. The case may be hard, but the floppy flops.


Given that CScript is the name of the "Console Based Script Host" for Windows, that can run JScript (Microsoft's old variant of JavaScript) and VBScript, and supposedly other pluggable scripting languages (I've never seen one in the wild), calling this "Cscript" is not a good idea.


It's a joke language, I doubt it will cause any significant ambiguity ever.


There's no Greek in there, so that's unfortunately another skip for me. I know most people won't care, but at least try for a WGL4 (https://en.wikipedia.org/wiki/Windows_Glyph_List_4) charset! All Greek-derived alphabets (Greek, Latin, Cyrillic) have very similar characteristics (unlike other writing systems such as Arabic, Hebrew, Chinese, Japanese or Korean), and it should be much easier to expand your system to cover those.

EDIT: As many people have said, the "custom Python engine" would be far more interesting than the font itself.


if you file a ticket or email me, i can try to get to the glyphs that you need... when i have time... thanks!


Could be worse, it could be syncing to Google Photos.

Oh, you want to download all photos so that you can free the cloud space? Too bad, you can only download them one by one. Or use Google Takeout and leave half of them undownloaded.


Security!

(read the article)


I read the article, but is there any evidence suggesting the Apple measuring app is insecure? It only collects data not linked to you (usage data and diagnostics) and is using the same privacy policy as the Apple camera app, which pretty much every iPhone user is trusting already.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: