A
Aireyanna Havaska
I am using Windows 10 1909 x64, display language English (United States), and of course, default font Segoe UI. There has been a problem that has troubled me for a very long time, and this one problem's cause is rooted in the design of Windows NT.
The problem is very very simple, that is, some texts with different character set encoding from the system default's, they show up as garbled codes, and completely unintelligible! I don't know if I conveyed my meanings right, so here are some screenshots that help explain what I mean.
As you see the first is a screenshot of file explorer, what should be Chinese characters are displayed as god-knows-what, the second is a screenshot of the program-which is in Chinese, again almost all Chinese characters are not displayed correctly, and the third and last is a screenshot of the console of programming language Julia, in the help section, what should be 17 mathematical symbols are displayed as square tofu-like substitute characters. The aforementioned problems do not limit to Chinese.
Way off the topic I would like to say I really dislike this Segoe UI font, which is sans-serif, so it has so many issues in itself, for one thing the upper-case I and lower-case l look exactly the same: Il, they are homoglyphs, and there is not any distinction between I and l in sans-serif fonts, which causes too much confusion. Of course in normal conditions, in a context, one can never confuse these two characters, such as in Intelligence, however there is a famous song Ievan Polkka, which is Finnish, and many people mistakenly believe the first character is an el instead of an i, resulting it often misnamed as Levan Polkka, and in programming there isn't much context to guess what the characters really are. And, nowadays almost everything is in Unicode, UTF-8, the character set which essentially contains everything, however Segoe UI is an ASCII font?, it has only so many characters.
Obviously the first two problems are caused by character mapping correspondence failure, the characters exclusive to UTF-8, beyond the code point limit of ASCII (or ANSI, I am not sure what Windows NT uses as default character set, nothing I can find online can elaborate me on that), only UTF-8 has the right characters at these code point entries, and ASCII has way fewer entries, the UTF-8 exclusive characters got mistakenly decoded in ASCII, which doesn't have them, so the computer divides the entry point by the limit of ASCII entries and displays the ASCII characters mapped to the remainders, which aren't the right ones. It is so easy to figure out why for me, but the difficult one is, how? How to solve this problem?
I have already installed Chinese (Simplified) language pack and fonts, but Windows is not using the right font and charset while displaying these Characters, instead uses the horrible Segoe UI as default font, uses it to display what it doesn't support. Obviously, the simplest way to solve the first two problems would be to switch the display language to Chinese, but I really don't like my system display Chinese by default, almost all my installed programs(more than 500 with exceptions less than ten) are in English. Besides, even if I make my system display Chinese by default, many English programs would display scrambled text thus making them unusable, for example Raw Therapee, this is one of the reasons why I switched my system to display English in the first place!
The last problem of cource is because there is the entry point, but the code point doesn't have any drawing mapped to it, the font doesn't include the character, so it displayes tofu character instead.
I have installed texlive in the hopes of solving this problem, it took good long hours, longer than installing Windows 10 itself, only to find it's unrelated to the problem.
How do I change default font of console programs? How do I change Windows 10 default character encoding/character set to utf-8?
Is there any reistry editing can do this? Please, I am rather good at regedit.
Continue reading...
The problem is very very simple, that is, some texts with different character set encoding from the system default's, they show up as garbled codes, and completely unintelligible! I don't know if I conveyed my meanings right, so here are some screenshots that help explain what I mean.
As you see the first is a screenshot of file explorer, what should be Chinese characters are displayed as god-knows-what, the second is a screenshot of the program-which is in Chinese, again almost all Chinese characters are not displayed correctly, and the third and last is a screenshot of the console of programming language Julia, in the help section, what should be 17 mathematical symbols are displayed as square tofu-like substitute characters. The aforementioned problems do not limit to Chinese.
Way off the topic I would like to say I really dislike this Segoe UI font, which is sans-serif, so it has so many issues in itself, for one thing the upper-case I and lower-case l look exactly the same: Il, they are homoglyphs, and there is not any distinction between I and l in sans-serif fonts, which causes too much confusion. Of course in normal conditions, in a context, one can never confuse these two characters, such as in Intelligence, however there is a famous song Ievan Polkka, which is Finnish, and many people mistakenly believe the first character is an el instead of an i, resulting it often misnamed as Levan Polkka, and in programming there isn't much context to guess what the characters really are. And, nowadays almost everything is in Unicode, UTF-8, the character set which essentially contains everything, however Segoe UI is an ASCII font?, it has only so many characters.
Obviously the first two problems are caused by character mapping correspondence failure, the characters exclusive to UTF-8, beyond the code point limit of ASCII (or ANSI, I am not sure what Windows NT uses as default character set, nothing I can find online can elaborate me on that), only UTF-8 has the right characters at these code point entries, and ASCII has way fewer entries, the UTF-8 exclusive characters got mistakenly decoded in ASCII, which doesn't have them, so the computer divides the entry point by the limit of ASCII entries and displays the ASCII characters mapped to the remainders, which aren't the right ones. It is so easy to figure out why for me, but the difficult one is, how? How to solve this problem?
I have already installed Chinese (Simplified) language pack and fonts, but Windows is not using the right font and charset while displaying these Characters, instead uses the horrible Segoe UI as default font, uses it to display what it doesn't support. Obviously, the simplest way to solve the first two problems would be to switch the display language to Chinese, but I really don't like my system display Chinese by default, almost all my installed programs(more than 500 with exceptions less than ten) are in English. Besides, even if I make my system display Chinese by default, many English programs would display scrambled text thus making them unusable, for example Raw Therapee, this is one of the reasons why I switched my system to display English in the first place!
The last problem of cource is because there is the entry point, but the code point doesn't have any drawing mapped to it, the font doesn't include the character, so it displayes tofu character instead.
I have installed texlive in the hopes of solving this problem, it took good long hours, longer than installing Windows 10 itself, only to find it's unrelated to the problem.
How do I change default font of console programs? How do I change Windows 10 default character encoding/character set to utf-8?
Is there any reistry editing can do this? Please, I am rather good at regedit.
Continue reading...