Beginner question about unicode

I'm working on my debut font. Just about to finish standard latin glyph. And I'm thinking to expand it to multilingual. The problem is, I don't understand what glyph/unicode to design in order for my font to support multilingual.

I tried to observe other fonts about multilingual support, and it turns out not every the unicode in latin extended A, B etc. is assigned with some design. In other words, there is some blank unicode in latin extended A, B, etc. So it doesn't help me understand which unicode range shall I design to  support multilingual.

Can anybody explain me the unicode range to design in order for my fonts to support multilinguals? Or maybe is there some concept/keywords shall I learn so I can understand about multilingual support? 

Thankss..
Tagged:

Comments

  • Wanting your font to have "multilingual" support is a bit vague. Do you mean the common Latin based European languages? If so covering the unicode blocks Basic Latin, Latin-1 Supplement and Latin Extended-A will add support for a lot of languages.

    I recommend using Alphabet Type's Charset Builder to create your character set, it's a fantastic tool.
    http://www.alphabet-type.com/tools/charset-builder/
  • The Letter Database is a useful resource for researching which characters are need to support which languages. 
  • George ThomasGeorge Thomas Posts: 631
    edited May 2018
    The blank spots in the Unicode charts are merely space for future additions, if any.
  • KP MawhoodKP Mawhood Posts: 294
    Adobe's Latin Character Sets can be useful.
    https://github.com/adobe-type-tools/adobe-latin-charsets
  • Theunis de JongTheunis de Jong Posts: 112
    edited May 2018
    Anyway, there is only very little overlap between Unicode ranges and actual language support. "New glyphs" may get added to the block in which one would expect only if there is enough empty space left.

    And then even not always; the Unicode cons. may be planning ahead and consider that if adding support for a glyph x, it could be expected to have to add x+1, x+2 etc. in short order. In such a case, they'd better off choosing a block with plenty of room for future expansion.

    Don't bother too much about actual Unicode values (blocks, ranges, or whatever). Look up what characters you need, and only then look up their Unicode values.
  • AzizMostafa AzizAliAzizMostafa AzizAli Posts: 103
    edited May 2018
    Worth exploring are:
    1. FiraGO 1.000 that supports multilinguals including Arabic.
    2. the attached QB-English.pdf
    Happy downloading and exploring.
  • Ben AdhaBen Adha Posts: 6
    woww.... thanks so much guys for helping me with this problem, especially for the references, resources and keywords. That's a lot of resources, a lot to learn, a long way to go. And thanks again to you guys for telling me where to start it from...
  • Simon CozensSimon Cozens Posts: 724
    This chapter might also be helpful for getting a handle on the basics of Unicode: https://simoncozens.github.io/fonts-and-layout/unicode.html
  • Mark SimonsonMark Simonson Posts: 1,652
    Unicode may seem a little daunting if you haven't dealt with it before, but it's way easier than the pre-Unicode days, when you had to ship different versions of your fonts with different encodings to support different languages—and for different platforms.
  • Ray LarabieRay Larabie Posts: 1,376
    edited May 2018
    Examine the free Bold style of Toxigenesis in your font editor. I'm not saying you should make your set exactly the same but I want to show you my bare minimum set. This is the repertoire I use for display typefaces which don't require historical characters and aren't intended for scholarly use like language studies, history books. I include Greek and Cyrillic in all my new fonts but you mentioned Latin so...

    Latin Extended A


    The missing glyphs with circumflexes are for Esperanto. So research Esperanto and make your own decision. 0138 and 0149 are historical. The Ldots are deprecated and shouldn't be included. Do some research on Ubreve and decide for yourself.

    Latin Extended B


    If you want to support more African languages for a display typeface, you'll have to do research to figure out what's in use, historical, what's mainly used for textbooks. I've never found a guide explaining this clearly. Oops just noticed I missed the flipped E at 018E. I think you should always include Vietnamese but if you don't want to, you can skip the hooked O and U.

    IPA Extensions


    Combining Diacritical marks


    Latin Extended Additional



    I don't recommend skipping the Vietnamese characters but if you decide to, keep the dotted vowels as other languages require them. Don't forget 1E9E...I usually don't miss that one. You need that for sure.

    Currency Symbols

  • AbrahamLeeAbrahamLee Posts: 262
    Unicode may seem a little daunting if you haven't dealt with it before, but it's way easier than the pre-Unicode days, when you had to ship different versions of your fonts with different encodings to support different languages—and for different platforms.
    Music notation fonts are going through a similar transformation. Each notation app has, since way before Unicode, just put music symbols where ever they wanted. For music font designers, that meant they needed to have a different file for each app, and potentially for each platform as well. This has been a real deterrent for many designers and creators of music fonts. A few years ago, however there had been some initiative to standardize music symbol encodings once and for all beyond the few symbols Unicode prescribes. The new standard is called Standard Music Font Layout (or SMuFL for short). Music notation apps are slowly becoming compliant with this, but it remains a slow process and I can’t quite see the light at the end of the tunnel yet. I keep dreaming of the day when the big industry players announce they’ve finally done it. Until then, it will remain a complex process that few embark on.
Sign In or Register to comment.