This is a summary of the full post over on my blog! Check it out for more details.

The problems with Dicier

1. big font files (the .otf files are almost all a few hundred kB)
2. maximum code length (Affinity apps can't handle codes/ligatures longer than 8 characters)
3. all-caps codes (not a huge problem, but might be kinda annoying)
4. locked codes (most icons have only one fixed code)
5. unique translations (all codes have to be unique, which causes problems when different languages (or even the same language) use the same word for different things; translations in general also involve a lotta work that could be done away with)

In short: the files are big and a little inconvenient (they have a lotta dead weight that you won't use on a given project), and the codes aren't customisable.

The solution

Add customisability: let you choose which icons go in your font (so your Dicier files are only as big as they need to be) and let you choose what codes they have (to suit your language, your technology, and your games).

The Dicery: a mix+match Dicier web app

My plan is to basically do what a bunch of icon webfonts already do, which is to have a web app (the "Dicery") that displays all the icons, lets you choose which to add to your font, and lets you write in the text for each icon (and it'd check for things like duplicate codes. The app would use opentype.js, a javascript library that lets you edit fonts in the browser.

The full post has more details, but basically it'd look like a more advanced version of the code tables in the Dicier User Guide:

...with checkboxes to pick icons, and text fields under each one to let you type in your own codes.

That'd also cut almost all the work involved in adding new translations (and I mean cut, not just offload on you; see full post for more info) and let me work on adding new icons and styles instead!

So that's the rough trajectory. The next announcement will be later this year or early next year when I hopefully have a working demo.