> How do you define 'core set of glyphs'? There are more than 150 different written scripts in the world — at least those that are well-recognized. Only a small portion of them are Latin-based ... Then, there's the Greek alphabet which is similar to Latin as well as Cyrillic. Then we have Arabic and Hangul
> Within the first year, before making it open source, I had something that covered the 200 most common Latin characters.
Seems like most of the time spent on Inter was the result of the creator not properly limiting his scope. Later he even talks about how the Cyrillic characters in his font seem completely wrong to actual users of Cyrillic because he can't read Cyrillic. As someone who deals with multiple different scripts on my computer, I always hated the fonts that try to set a uniform style for every script when many of the designers clearly had no idea how many of these scripts worked. It really only makes sense to try to support every script for fallback fonts like Unifont. Rarely do typographic concepts translate well between scripts. If you want an example, install the CJK font "Hana Mincho" and type some English text in it: It looks atrocious.
> Within the first year, before making it open source, I had something that covered the 200 most common Latin characters.
Seems like most of the time spent on Inter was the result of the creator not properly limiting his scope. Later he even talks about how the Cyrillic characters in his font seem completely wrong to actual users of Cyrillic because he can't read Cyrillic. As someone who deals with multiple different scripts on my computer, I always hated the fonts that try to set a uniform style for every script when many of the designers clearly had no idea how many of these scripts worked. It really only makes sense to try to support every script for fallback fonts like Unifont. Rarely do typographic concepts translate well between scripts. If you want an example, install the CJK font "Hana Mincho" and type some English text in it: It looks atrocious.