2919: Arabic Typewriters Dec 18, 2024

Some alphabets lent themselves to type writers relatively easily, for example Greek, Cyrillic, Latin, and Hebrew because all the letters are separate and can be rendered with minimal complication for monospaced fonts. Others, like Arabic, were enigmatic for many years.

Arabic’s writing system presents several notable challenges. Each of the 28 letters usually takes 4 forms, whether it's at the beginning, middle, or end of a word, or standing alone, which would lead to far too many characters to reasonably fit. Compounding that is that since it is a connected, cursive font, the letters need to be able to line up with each other, and need to at times both go above or below the line of type. Further complicating the matter is that of width, though an automatic double-space was comparatively not hard to solve.

In the end, there were many attempts in the years around the turn of the 20th century, but the most successful patent was filed by Selim Haddad. Earlier attempts had tried cylinders—like how later electric typewriters would work—but this proved ineffective. Rather the arabic typewrite from Haddad used a simplified form of the 4 letters: connected that also served as word-initial, or disconnected which also served as word-final (for instance: سـ connected, and س disconnected). Lacking capital letters, this was performed similarly to what was done for English with the shift key. The point of connection was made level for all letters. A few exceptions had to be made where the final forms and isolated forms were too distinct to reuse and required two separate keys, but these were paired on keys with other exceptional letters that don’t have a connected form. Diacritics, while traditionally always included in Arabic writing, were by and large ignored. This reduced Arabic from requiring, by various estimates, at least 638 to merely 53 keys, compared to 44 on an English typewriter.