I’ve built a new font! Thoughts and feedback on my approach very welcome.

    • edentOP
      link
      fedilink
      111 year ago

      Thanks :-) It was a couple of days work. Mostly teaching myself stuff that I’d forgotten. I blogged about it so others can follow the process if they want.

  • sab
    link
    fedilink
    10
    edit-2
    1 year ago

    Fantastic project! And thanks for the heads up about the 17th century Dutch fonts - I might consider that for the printed version of my thesis that will probably rot away in the university library anyway. Might as well make it interesting looking.

    My favourite project of this kind is TT2020, which takes typewriter imitation to the next level by including irregularities within the font itself. Each (common) glyph has several versions, making the outcome look much more genuine and allowing for the occasional badly printed glyph without making the same error repeat itself to infinity.

    It’s a bit overkill maybe, but it would absolutely lend itself well to Shakespearian scripts. :)

    • edentOP
      link
      fedilink
      61 year ago

      Both! Unicode has some support for as-is ligatures. For example is U+FB01.

      Modern fonts can also use self-defined ligatures. That’s how fonts like https://www.sansbullshitsans.com/

      So my plan is (eventually) to add in ligatures where Unicode has defined them - and automatically replace typed text with self-defined ligatures where it doesn’t.

    • @winety@communick.news
      link
      fedilink
      4
      edit-2
      1 year ago

      Yes, they do. Part of the OpenType standard are the so called “OpenType features” which (amongst other things) allow for contextual alternates, i.e. different kinds of ligatures, and for stylistic alternates, e.g. a slashed zero, a single-storey ɑ, etc. All of these different glyphs are encoded in the font and can be enabled when typesetting using different selectors. This website shows them off.

      Some ligatures, like “ffl”, are a separate character in Unicode. Some were added because they can be considered a different character in languages other than English. Some (like “ffl”) were added because of legacy reasons; “no more will be encoded in any circumstances”.

  • HunterHog
    link
    fedilink
    21 year ago

    Oh, I love everything about this project so much! It looks lovely!

    It also looks mighty useful for making handouts for TTRPG campaigns ahaha.

    Was the letter-recognition Python codeblock of your own making?

    • edentOP
      link
      fedilink
      31 year ago

      Cheers. The code was cobbled together by me from various random tutorials and things I had laying around.

  • shepherd
    link
    fedilink
    11 year ago

    Oooh, I’m excited to try this out! But later, 'cause I’m just on mobile, and kbin doesn’t have a save function yet lol.

    • edentOP
      link
      fedilink
      21 year ago

      Cheers! I’m hoping to add some more letters and tidy up the rest when I have more time.

      • MisterMoo
        link
        fedilink
        11 year ago

        This is such a great project! I can see how there’s a little room for clean-up on that long S. Plus the comma-looking apostrophe 😵‍💫. Although maybe that’s how that punctuation mark acted back then?

        • edentOP
          link
          fedilink
          11 year ago

          Nah, that’s just me not adjusting the positioning of the apostrophe. Lots of clean-up to do :-)

  • Storksforlegs
    link
    fedilink
    11 year ago

    You’ve really retained the slightly blotted look of the print. Great work!

    • edentOP
      link
      fedilink
      21 year ago

      Thanks! I’m chuffed with how they turned out.