A preview look of my hobby free software font editor - gerb

Manos Pitsidianakis

NOTE: This is in prototype state, lacks basic features such as saving modifications or exporting to binary formats (.ttf, .otf, etc). It's more of a "what if?" than a tool at this early point.

• *gerb ʰ-: reconstructed Proto-Indo-European root, meaning to carve

So, for fun I started making a UFO font editor that's libre/gratis and open source. The technology stack is the gtk GUI library and the Rust programming language. I'm a programmer by trade so I'm interested to know what type professionals think of it and whether it's a worthy endeavor to pursuit for actual type development.

The development is on github: https://github.com/epilys/gerb

The UI is unpolished because it's going through changes at this early stage. As stated it in the github site, the following list describes the goals of the project:

• Fun
• Good UX not necessarily tied to prior art
• Reasonable performance
• Configurability
• Use visual feedback for all kinds of operations to inform the user of the current state (for example, a Bézier path tool that shows you the current curve's degree and progress).

Some other goals:

• Compatibility with python scripts for Glyphs and Fontforge
• Complete support of UFO format
• Everything (Project, masters, glyph groups, glyphs, contours, standalone curves/splines, guidelines, etc) is an object whose properties can be introspected and edited like in Blender.

Some videos/images I've made the past few weeks showcasing what I was currently working on:

Le Murmure, by Velvetyne Font Foundry https://www.velvetyne.fr/fonts/le-murmure/

Guidelines as introspect-able objects (Video)

https://gist.githubusercontent.com/epilys/51f07f5350fdf36877993bdf7132c038/raw/e56d362951d9a9b22e6555fe872926fabe478b8a/gerb-guidelines.mp4?raw=true  

 

Raw unfinished bezier pen tool (Video)

https://gist.github.com/epilys/51f07f5350fdf36877993bdf7132c038/raw/e56d362951d9a9b22e6555fe872926fabe478b8a/gerb-first_bezier_pen.mp4?raw=true 

Editing existing bezier curves (Video) not final behavior/look

https://gist.github.com/epilys/51f07f5350fdf36877993bdf7132c038/raw/e56d362951d9a9b22e6555fe872926fabe478b8a/gerb-handles.mp4?raw=true

Simon Cozens

This is very interesting! Can I recommend you have a look at the norad crate for handling UFOs? The kurbo crate may be helpful for handling Bézier curves too.

Manos Pitsidianakis

Simon Cozens said:

This is very interesting! Can I recommend you have a look at the norad crate for handling UFOs? The kurbo crate may be helpful for handling Bézier curves too.

Will certainly do, thank you! kurbo will be useful in the future, when spline modifications are implemented.

Alex Visi

How come there’s not feedback here?

epilys said:

so I'm interested to know what type professionals think of it and whether it's a worthy endeavor to pursuit for actual type development.

It is great to see someone having some interest in making a type design app!
However, looking at the demos and description, I don’t really understand what makes it different from the existing apps. If you’re willing to spend so much time on it, I think it would make sense to collaborate with a practicing type designer/foundry and start by finding the pain points of the current apps. The problem is that those pain points are easy to fix with plugins. In my view, making a whole new useful editor is possible, but it requires a lot of work and understanding of the font design process.

Nonetheless, I think there’s a market for plugins. Also, a lot of people could benefit from something better than Adobe apps.

Andreas Stötzner

> I'm interested to know what type professionals think of it and whether it's a worthy endeavor to pursuit for actual type development.

I think that there are some 4 or 5 matured font creation programs available already and, to make a case for your development, you might highlight the peculiar strengths or stand-alone features of your product a little more. “It’s free” is of no interest for me because I have invested in professional programs years ago and I obtained experience with them needed in my work process.

Sérgio Martins

I agree with Simon, this is quite interesting!

What would your be your OS support goal? Personally, I would love to see more font editor options available for OSs other than mac OS.

Justin Penner

This looks like a fun project, but what differentiates it from other Rust-based open-source font editors like MFEK and Runebender? It seems like a very big project to take on when there are already other projects that you could perhaps contribute to.

Thomas Phinney

Agree with @Justin Penner
I think we are well past the point in number of font editor projects (both in general and in Rust-based open source ones), where the world would benefit more from more people-per-project, rather than more projects.

Simon Cozens

> This looks like a fun project, but what differentiates it from other Rust-based open-source font editors like MFEK and Runebender?

Ooh, I can answer this one: simplicity and ergonomics. MFEK is a great project but Fred has (deliberately) never used a commercial font editor and his desired user interface and workflow is considerably different from that of most designers. Runebender is designed to be a pilot project for a new set of user interface widgets so its development tends to have far more time spent on reinventing the GUI world using new (and therefore unfamiliar) tools than actually making a font editor, which in turn makes it slow to emerge difficult to contribute to. But the libraries that both of these systems have produced (kurbo, norad, MFEKmath, etc.) are useful components to a more user-friendly and conventional editor.

If I had the time to hack on any of these things, I'd probably be working on gerb.

Thomas Phinney

And that shows why asking the question is more sensible than my initial rush to judgment.  

Manos Pitsidianakis

Alex Visi said:

However, looking at the demos and description, I don’t really understand what makes it different from the existing apps. If you’re willing to spend so much time on it, I think it would make sense to collaborate with a practicing type designer/foundry and start by finding the pain points of the current apps. The problem is that those pain points are easy to fix with plugins. In my view, making a whole new useful editor is possible, but it requires a lot of work and understanding of the font design process.

Nonetheless, I think there’s a market for plugins. Also, a lot of people could benefit from something better than Adobe apps.

Thanks so much for your comments! The differences are:

• It's free and libre (open source software). It won't be commercialised so it's inclusive to people who cannot afford professional type making software
• It's using a strong GUI foundation, the gtk framework, and will have a strong library foundation: all the existing open source libraries mentioned in this thread, kurbo etc.
• paypoints will be able to be fixed by people changing the source directly instead of "patching" a closed source program with plugins!

It's still an experiment for fun, so the useful editor you mentioned is far in the future.

By the way, thanks for recommending to collaborate with type designers. I have some libre typeface designers/foundries I can approach!

Justin Penner said:

This looks like a fun project, but what differentiates it from other Rust-based open-source font editors like MFEK and Runebender? It seems like a very big project to take on when there are already other projects that you could perhaps contribute to.

As Simon said these two projects have provided lots of great foundations for font editing but they are not targeting a conventional editor interface like Fontforge or Glyphs. Most of the code I have written and am writing is for making the interface instead of re-inventing the wheel on curve manipulation etc. Also, this is a hobby project made for personal enjoyment, if anything useful results from this it'd be secondary and a happy accident!

The interface experience I'm going for right now is something like Blender: every data object can be inspected and manipulated from anywhere in the Editor and plugins.

Thomas Phinney said:

where the world would benefit more from more people-per-project, rather than more projects.

I completely agree with this view. My plan is to focus on the UI, contribute to the existing Rust font ecosystem if needed.

Manos Pitsidianakis

Sérgio Martins said:

I agree with Simon, this is quite interesting!

What would your be your OS support goal? Personally, I would love to see more font editor options available for OSs other than mac OS.

I'm making this on Linux! The UI framework works on most other major OSes as well: Windows, Mac and BSDs.

In fact I've recently got a company macbook pro so I can polish it for both Linux and Macos.

Dave Crossland

epilys said:

The interface experience I'm going for right now is something like Blender: every data object can be inspected and manipulated from anywhere in the Editor and plugins.

I think self-extensibility is a critical feature of a good font editor; is it possible in your code design to add a Jupyter kernel REPL that can do that kind of introspection?

Manos Pitsidianakis

Dave Crossland said:

I think self-extensibility is a critical feature of a good font editor; is it possible in your code design to add a Jupyter kernel REPL that can do that kind of introspection?

It's possible to expose Rust data to python (I've written a small proof of concept here for running python scripts against a Rust API on a Rust application) but I don't have any experience with Jupyter notebooks let alone developing for them. Sounds very interesting and possible though! 🤔