libui VS tauri

You can kind of see the desktop UI train wreck in real time here.

We started with simple stable APIs for a common look and feel. For a while these were evolved and made available in other languages. This was back when native apps were consistent and intuitive and you could… uhh… actually write and ship them without bundling giant runtimes or checking a huge compatibility matrix.

Then around 2012 the train rounds the bend and screeeeech it hits some bad track and starts to derail. UI starts trying to emulate the web, a terrible UI platform, and sane compositional UI libraries and APIs are abandoned in favor of XML soup.

Since this stuff is a trash fire, this is followed by multiple incompatible attempts to replace or fix this. Most of these are abandoned dead ends.

Meanwhile the dev community just said fuck it and went to Electron, creating today’s world where a “hello world” app with an OK button is hundreds of megabytes and has to load an entire private copy of a language runtime and rendering engine.

Versions of this comedy of errors have occurred on every other platform, and of course there has been little effort to create a cross platform UI API that’s sane beyond Qt (with its own problems) and dozens of half completed OSS projects.

So enjoy Electron I guess.

There was one sane human being who tried to do this a while ago:

https://github.com/andlabs/libui

It’s the only sane desktop UI project I’ve seen in almost 20 years, an attempt to create an actual cross platform common API. But it’s abandoned of course, likely too difficult for one dev and nobody is going to provide financial support for anything that sane.

Maybe AI will get good enough some day that we can use it to do a thing like that.

What about https://github.com/andlabs/libui

Fantastic! This is similar to the C library `libui` since it also acts as a wrapper of native libraries of each platform.

If only there was a way to interface to these using some declarative minimal and highly opinionated programming language and paradigm...

https://github.com/andlabs/libui

For the GUI you will need a library or framework that interacts with your specifiv operating system and allows you to create windows and a canvas to which you can draw. You could give libui a chance.

libui-ng is a cross-platform GUI library with native widgets written in C. It is based on an earlier, (currently) inactive project known as libui. While Rust bindings for libui have existed for years (see ui-sys and iui), there is no solution for the new libui-ng; libui-ng-sys aims to fill this role.

libui

https://github.com/andlabs/libui is very nice, but unfortunately dead, if it serves your purpose consider using it, this is a fork under development https://github.com/libui-ng/libui-ng

https://github.com/andlabs/libui

Better yet, it has excellent DSLs that make it possible to build desktop apps in a way similar to HTML, but much better due to keeping all code dynamic in one language (no static/dynamic multi-language separation/mixing dissonance):

Maybe a module that uses https://github.com/andlabs/libui or a light HTML renderer?

Hi, dear Tauri! Long time no see. I published my first post, Developing a Desktop Application via Rust and NextJS. The Tauri Way almost a year ago. Since then, Tauri has become stronger. I'm happy about that! And now, I am very pleased to make a useful contribution to the Tauri community. As a full-stack developer, I frequently face situations where I need to start a DB-based UI project as fast as possible. It's stressful if I need to start the project from 100% scratch. I prefer to keep some boilerplates on hand, which will save me time and nerves and will be the subject of this article.

Hey Thanks!

Just electron & vite. I might actually migrate off electron, Tauri (https://tauri.app/) seems to be getting more stable and it's gotten great reviews.

I think this is the boilerplate I used though https://github.com/cawa-93/vite-electron-builder.

Well the great thing about WebAssembly is that you can port QT or anything else to be at a layer below -- thanks to WebAssembly Interface Types[0] and the Component Model specification that works underneath that.

To over-simplify, the Component Model manages language interop, and WIT constrains the boundaries with interfaces.

IMO the problem here is defining a 90% solution for most window, tab, button, etc management, then building embeddings in QT, Flutter/Skia, and other lower level engines. Getting a good cross-platform way of doing data passing, triggering re-renders, serializing window state is probably the meat of the interesting work.

On top of that, you really need great UX. This is normally where projects fall short -- why should I use this solution instead of something like Tauri[2] which is excellent or Electron?

[0]: https://github.com/WebAssembly/component-model/blob/main/des...

[1]: https://github.com/WebAssembly/component-model/blob/main/des...

[2]: https://tauri.app/

Welcome to the 2nd episode of our series “Building with Tauri”, where we chat with developers who build amazing projects and products using Tauri.

Tauri seemed like the "thing" I should switch to because everybody loves Rust (heh), and because it ships significantly smaller apps.

First of all, I will be using npm as my package manager but feel free to use whatever you prefer. Find more info here.

Today I'm talking about Tauri! Do you know all the various tools that allow you to develop native applications starting from web languages? They often need an intermediate compilation, in the middle of which you end up encountering various problems not always transparent and directly solvable with a language mostly detached from native development. On the other hand, there's still the ease of developing attractive and easily usable interfaces, which are more difficult to develop with low level languages.