|
Sorry, that should say public and cleanliness. As far as orthography goes, english is pretty abhorrent. There are some advantages to having an alphabet over an ideographic script, but if we're going to go down this route, featural scripts are fantastic. English is the horror. quote:English is tough stuff
|
# ? Nov 18, 2012 13:32 |
|
|
# ? May 14, 2024 11:04 |
|
tef posted:I was just wondering if you had any pubic sources on the deanliness on the forgivingness of english. tef posted:As far as orthography goes, english is pretty abhorrent. There are some advantages to having an alphabet over an ideographic script, but if we're going to go down this route, featural scripts are fantastic. The Gripper fucked around with this message at 13:40 on Nov 18, 2012 |
# ? Nov 18, 2012 13:33 |
|
That's a cool poem, but pronunciation doesn't really matter when programming and neither does grammatical correctness. I must have heard at least 10 different pronunciations for the word 'virtual' but never had a problem figuring out what the person meant.tef posted:
Could be a good way to separate language keywords from variables and such actually, but I wouldn't want to have to get a separate keyboard just for coding.
|
# ? Nov 18, 2012 17:39 |
|
BliShem posted:Could be a good way to separate language keywords from variables and such actually, but I wouldn't want to have to get a separate keyboard just for coding. You know that the letters on the top of the keys aren't anything to do with the characters on the screen right?
|
# ? Nov 18, 2012 22:02 |
|
BliShem posted:I wouldn't want to have to get a separate keyboard just for coding. No one does, which is why I still only see upsides to letting people write tokens in whatever characters are convenient.
|
# ? Nov 18, 2012 22:08 |
|
If you're not using an Optimus keyboard, you can't call yourself an international developer!
|
# ? Nov 18, 2012 22:13 |
|
Zombywuf posted:You know that the letters on the top of the keys aren't anything to do with the characters on the screen right?
|
# ? Nov 18, 2012 22:22 |
|
tef posted:The spelling and pronunciation have nothing in common though? I thought you ought to know about this being tough and thoroughly challenging for newcomers to plough through, a linguistic hiccough or cough perhaps. I really hope this sentence was composed on purpose
|
# ? Nov 18, 2012 22:28 |
|
Volte posted:They have something to do with them, in the brains of the average person. Does anyone know all the symbols that OS X lets you type with the option key? Because I sure don't, but I might if they were written on the keys. Open the keyboard viewer and hold the option key. There you can see all of them. I know all the ones I use daily by heart (stuff like é, è, ê, ü etc), and those alone are already quite a lot to be honest.
|
# ? Nov 18, 2012 22:29 |
|
Volte posted:They have something to do with them, in the brains of the average person. Does anyone know all the symbols that OS X lets you type with the option key? Because I sure don't, but I might if they were written on the keys. I don't know all the ones, but I know the ones I find myself using often; the accented characters are pretty easy to remember because they're compositional (alt-[`einu] followed by an applicable character). Doc Hawkins posted:No one does, which is why I still only see upsides to letting people write tokens in whatever characters are convenient. Anything I can type in under four strokes on my Mac's US English QWERTY keyboard is convenient for me. However, some people use Dvorak, AZERTY, QWERTZ, or QWERTY with a different localized layout. If I use parentheses, braces, brackets, dollar signs, or digits, it makes typing difficult for these other keyboard layouts. On the other hand, they're easier for me to read and understand than if I limited myself to just the upper- and lower-case alphabet and spaces. If I'm telling the majority of the world that uses non-US keyboards to deal with it, I might as well tell the minority in my programming circles that don't use Macs to comply with Rules 12 and 36 or quit complaining.
|
# ? Nov 18, 2012 22:41 |
|
Volte posted:They have something to do with them, in the brains of the average person. Does anyone know all the symbols that OS X lets you type with the option key? Because I sure don't, but I might if they were written on the keys. You think "the average person" uses the Latin alphabet! That's cute.
|
# ? Nov 19, 2012 10:41 |
|
Only the elite use the Latin alphabet
|
# ? Nov 19, 2012 10:49 |
|
The Gripper posted:If you're not using an Optimus keyboard, you can't call yourself an international developer! Only if your definition of international means "US physical key layout" Anyways, all layouts should have an "Alt Gr" key.
|
# ? Nov 19, 2012 11:49 |
|
Volte posted:They have something to do with them, in the brains of the average person. Does anyone know all the symbols that OS X lets you type with the option key? Because I sure don't, but I might if they were written on the keys. I'm confused, how does the 'A' on my 'a' key tell me that when I hold option it produces 'å'?
|
# ? Nov 19, 2012 11:51 |
|
Zombywuf posted:You know that the letters on the top of the keys aren't anything to do with the characters on the screen right? Yes, but learning touch typing on a keyboard with a different layout than the characters you're typing should not be a prerequisite to learning programming.
|
# ? Nov 19, 2012 12:38 |
|
BliShem posted:Yes, but learning touch typing on a keyboard with a different layout than the characters you're typing should not be a prerequisite to learning programming. Learning a language is a different thing to learning programming. Next you'll be telling me that learning to type Ä shouldn't be a prerequisite to learning Finnish.
|
# ? Nov 19, 2012 13:06 |
|
Zombywuf posted:Learning a language is a different thing to learning programming. Next you'll be telling me that learning to type Ä shouldn't be a prerequisite to learning Finnish. Typing Ä on a keyboard without Finnish letter labels? Why should it be? Being able to type is only a prerequisite for computer programming languages, the rest of the world made do just fine without it for thousands of years.
|
# ? Nov 19, 2012 14:32 |
|
HORATIO HORNBLOWER posted:The question of using non-ASCII characters in source files is germane to this thread in the sense that the real argument against them is that there are so few systems that sanely handle Unicode end-to-end. After having been a complete purist and borderline nutjob about this for a while, I have come around to the school of thought that Unicode itself is the problem if there are zero programming languages or environments that actually implement it correctly. Even when you take out the problems that Unicode tries (and mostly fails) to solve that are essentially rather than accidentally hard, there are still a billion things that think they're at least Unicode aware that gently caress basic things up and leave you with boxes or wingdings. We need to declare Unicode bankruptcy or something, and start afresh with some other approach, maybe even reconsidering whether code points are the correct atom to solve the problems we want to solve.
|
# ? Nov 19, 2012 15:57 |
|
utf-8 is the new latin-1? Starting afresh will ensure that no-one will implement your new solution correctly either, though.
|
# ? Nov 19, 2012 16:03 |
|
BliShem posted:Typing Ä on a keyboard without Finnish letter labels? Why should it be? Being able to type is only a prerequisite for computer programming languages, the rest of the world made do just fine without it for thousands of years. In the modern world if you can't type a language can you really claim to have learned it? Otto Skorzeny posted:After having been a complete purist and borderline nutjob about this for a while, I have come around to the school of thought that Unicode itself is the problem if there are zero programming languages or environments that actually implement it correctly. Even when you take out the problems that Unicode tries (and mostly fails) to solve that are essentially rather than accidentally hard, there are still a billion things that think they're at least Unicode aware that gently caress basic things up and leave you with boxes or wingdings. We need to declare Unicode bankruptcy or something, and start afresh with some other approach, maybe even reconsidering whether code points are the correct atom to solve the problems we want to solve. So the unicode standard is 24 years old, and version 6 is only 2. If you think second system unicode is going to be better than first system unicode I think you are very much mistaken. The solution is more zealotry, not less. Either you get unicode right or you're a racist.
|
# ? Nov 19, 2012 16:22 |
|
There's a few fuckups in Unicode (UTF-16 should not exist; having both precomposed and composition characters makes things twice as exciting), but very little of the complexity of properly supporting Unicode is actually particularly related to Unicode. The actual hard parts are things like dealing with things like that sorting and capitalization are locale-dependent, and legacy non-Unicode things.
|
# ? Nov 19, 2012 16:42 |
|
tef posted:utf-8 is the new latin-1? Basically yes. I know I'll never be able to implement UTF-8 or any other real unicode at work on the chips I write code for, and yet I want to be able to write stuff that can be easily localized (since we sell poo poo to a bunch of countries) without resorting to some bastardized bespoke encoding per product version or some similar horseshit (which is pretty much what gets done now, to the extent that it gets done). If there were some less enormous spec that allowed me to display any character ever but did nothing else (and maybe was always 4 bytes per character, no exceptions, no you shut the gently caress up dad etc), no casefolding, no concept of cases even, just display the poo poo I tell you to, it would make my life much easier. Maybe my requirements are weird! or maybe the problem is just intractable.
|
# ? Nov 19, 2012 17:05 |
|
Kinda surprised no-one's made unicode-on-a-chip yet. I'd guess there's a big enough market for it.
|
# ? Nov 19, 2012 17:26 |
|
Zombywuf posted:Kinda surprised no-one's made unicode-on-a-chip yet. I'd guess there's a big enough market for it. What would it do and how would it grow?
|
# ? Nov 19, 2012 17:55 |
|
Unicode as a service
|
# ? Nov 19, 2012 18:08 |
|
ninjeff posted:You can't assume every English-based programming language's keywords have a direct mapping to words in every other natural language. It's generally the case for other languages from around Europe, sure. But there's no Japanese (for example) word that maps to "while" or "if" alone. Ok, so some people have to map a slightly more complicated idea. Do any of these work? http://translate.google.com/#en/ja/If http://translate.google.com/#en/ja/while I'm assuming Google is dumbing poo poo down for me there, but it looks like there are options. Editors could just display the file backwa, uh, I mean, whichever direction you want? Also, I've tried typing with a different keyboard language (Swiss) on a standard English keyboard and it was terrible even with a printout of the 'actual' layout taped to the monitor.
|
# ? Nov 19, 2012 18:23 |
|
Cocoa Crispies posted:how would it grow? EEPROM or some other nonvol storage, probably
|
# ? Nov 19, 2012 18:34 |
|
Sneaking Mission posted:Unicode as a service Scalable etherial unicode in the cloud.
|
# ? Nov 19, 2012 19:08 |
|
Plorkyeran posted:There's a few fuckups in Unicode (UTF-16 should not exist; having both precomposed and composition characters makes things twice as exciting), Han Unification *opens can of worms* See also, emojii: http://www.reigndesign.com/blog/love-hotels-and-unicode/ As for UTF-16, it was a mistake, but it came from the earliest notions of unicode http://www.unicode.org/history/unicode88.pdf Where they argue '16 bits should be enough for everyone', and rank the importance of a language based around GDP. Hilarious.
|
# ? Nov 19, 2012 19:27 |
|
Otto Skorzeny posted:Basically yes. I know I'll never be able to implement UTF-8 or any other real unicode at work on the chips I write code for, and yet I want to be able to write stuff that can be easily localized (since we sell poo poo to a bunch of countries) without resorting to some bastardized bespoke encoding per product version or some similar horseshit (which is pretty much what gets done now, to the extent that it gets done). I doubt any encoding that would solve this problem would be less burdensome than unicode as-is.
|
# ? Nov 19, 2012 19:35 |
|
tef posted:Han Unification *opens can of worms* See also, emojii: http://www.reigndesign.com/blog/love-hotels-and-unicode/ Yet again mobile phones ruin it for everyone.
|
# ? Nov 19, 2012 19:45 |
Cocoa Crispies posted:What would it do and how would it grow? Kickstarter. Doesn't matter if it makes sense, the answer is always kickstarter.
|
|
# ? Nov 19, 2012 20:27 |
|
Manslaughter posted:Kickstarter. Doesn't matter if it makes sense, the answer is always kickstarter. What if it's not just one processor but a whole matrix of them? You can swap processors in/out depending on what languages you want.
|
# ? Nov 19, 2012 20:31 |
|
tef posted:See also, emojii: http://www.reigndesign.com/blog/love-hotels-and-unicode/ Mother. Of. God. Edit: After a quick test, I'm extremely disappointed that the Scala compiler isn't accepting Emoji for operator identifiers. The bikini XOR has as much right to exist as any other!
|
# ? Nov 19, 2012 21:22 |
|
Haskell is ready for the future.
|
# ? Nov 19, 2012 21:32 |
|
Emoji symbols work quite well in Google email addresses on a Mac too e.g foo+emojisymbol@gmail.com As do they in many other Mac applications!
|
# ? Nov 19, 2012 23:56 |
|
ultramiraculous posted:Mother. Of. God. Ahahaha I love academics and what hapens when you stick them on committees to debate stupid poo poo. Long bewildering debates spanning years over the meanings of words. I was on the academic council (reasearch funding coordination committee) at Murdoch university in the mid 2000s , and there was an ongoing debate about what the meaning of "Multidisciplinary" was in the universitys constitution. The debate had been going for 15 years and was still unresolved, and would come up as a topic about every 3 months to the same shouting match between the postmodern leaning eng-lit ("its context dependent!") department, the ultra-analytical ("Words have meanings damnit!") philosophy department , and everyone else who would watch on in bemused bewilderment, and occasionally offering "Well , I think I know what multidisciplinary means. Yesterday I borrowed a stapler from the vetrinary science guys!". 15 loving years to define a word. Still unresolved to this day. But if you walked down into the courtyard and asked an undergrad, you'd get a 5-10 second silence followed by a consise and entirely useable answer good enough for about 90% of the time. This is for a concept that was supposed to be the defining trait of the university, yet none of the boffins could agree on what the defining trait was supposed to mean! And it was in this sort of committee environment unicode was born. duck monster fucked around with this message at 05:40 on Nov 20, 2012 |
# ? Nov 20, 2012 05:37 |
|
duck monster posted:15 loving years to define a word. It's exactly what scissors paper rock was invented for!
|
# ? Nov 20, 2012 09:10 |
|
Munkeymon posted:Ok, so some people have to map a slightly more complicated idea. Do any of these work? http://translate.google.com/#en/ja/If http://translate.google.com/#en/ja/while I'm assuming Google is dumbing poo poo down for me there, but it looks like there are options. The word you posted for 'if' literally means 'case' (the noun). The one for 'while' isn't really considered a word; it's just a verb suffix, like '-ing' in English.
|
# ? Nov 20, 2012 11:13 |
|
|
# ? May 14, 2024 11:04 |
|
ninjeff posted:The word you posted for 'if' literally means 'case' (the noun). I think the second one works fine though, if you change the keyword position a little bit... code:
|
# ? Nov 20, 2012 15:36 |