Sibai
banner
sibai.bsky.social
Sibai
@sibai.bsky.social
Enthusiastic reader & movie buff
I am going to open source it once the project is done. Tho I have absolutely no idea when that will be haha

It needs some work to be fully ready. It doesn't FULLY implement important standards such as UAX #14 or UAX #29 (grapheme clustering) for example
January 27, 2026 at 8:34 PM
Gradien to generate the gradients :-
gradien.sibai.me
Grd Edtr - Gradient Editor by Sibai Isaac
Create beautiful gradients using shaders. Free to use, export your canvas at any size, and explore endless color possibilities.
gradien.sibai.me
January 24, 2026 at 9:52 PM
Well personally I’m not interested in another cash grab game where the game is basically abandoned, and servers are basically running for bots to farm items and cheaters to play
January 24, 2026 at 12:57 AM
Yep 🙌
January 24, 2026 at 12:54 AM
Oh if they turn it into some money cow like they did with Cs then it’s over before it even started
January 23, 2026 at 11:43 PM
would be shocking if it doesnt tbh. They put a lot of work in it and I cant see them abruptly killing the game and its growth
January 23, 2026 at 10:41 PM
For sure. The basics are the basics, whatever language you end up using. I would even argue that using a less forgiving language teaches you to handle memory better.
January 23, 2026 at 10:40 PM
Really unfortunate. Hopefully it makes a strong comeback. But I feel like that's the case with most new online games. They get a LOT of hype when they initially release, but maintaining interest and a player-base seems borderline impossible nowadays.
January 23, 2026 at 10:36 PM
Here is a preview:
bsky.app/profile/siba...

I will be posting more of the journey on here.
The whole rendering engine is written in Rust and compiles to WASM. Performance is extraordinary and collab works as smooth as butter.
A showcase of my Rust rendering engine for a web based collaborative board!

My latest addition: PDF files. Drag'n'Drop a PDF file directly into the board and start working on it!

Can't believe I am saying this, but choosing Rust was one of the best decisions I made for this project.
January 23, 2026 at 10:34 PM
Other than that, the dev-experience is great, especially when using modern IDEs/Code-Editors.
January 23, 2026 at 9:52 PM
My current work is 95% if not more, in Rust WASM. Web based, WebGL2 renderer in Rust.
It is really great. But try to limit the interop between JS and Rust to a MINIMUM. Very expensive, especially when trying to hit 60+fps in my case.

Also you can't simply use dependencies that rely on a C++ lib.
January 23, 2026 at 9:52 PM
Even tho the idea is tempting, browser support is still very weak for anything serious at the moment. I think I will have to stick to WebGL2 for my needs for a while at least. Hopeful for the future of WebGPU tho!
January 23, 2026 at 9:40 PM
yea wish this game got more attention
January 23, 2026 at 9:39 PM
yea i get it
January 23, 2026 at 9:38 PM
Contrary to most, but years ago, I liked his books "Sapiens" and "Homo Deus" even. But I cannot stand his horrendous takes on almost every other genre he tries to comment on, especially AI. He has some of the worst takes on AI.
January 23, 2026 at 9:37 PM
I am currently knees deep into Rust's WASM. It is superb but, I found the best architecture is the one where there is barely, if any at all-interop between JS and WASM. And I assume it is probably the same using Tauri's ICP? Keep in mind I am talking about performance demanding software.
January 23, 2026 at 9:33 PM
I thought the AI devices experiments all died a couple of years ago alongside Rabbit R1 or Humane AI pin which was ridiculously bad?
January 23, 2026 at 9:29 PM
Yes BUUUUT, the rust compiler is very very slow. So slow that it significantly affects the development times if the project is big enough. Other than that, it is great.
January 23, 2026 at 9:10 PM
Oh I also wrote one very recently in Rust! Mine is for a layout engine tho and definitely took longer 2 hours. 😂
January 23, 2026 at 9:05 PM
Hard agree with this take! Often times with these takes I find that the programs in thought are surface level/non-critical.
AI will not in at least 10 years produce enterprise level high demand distributed systems for example.
January 21, 2026 at 10:34 PM