Thats pretty cool 😁

It's also possible to integrate DOSBox into the UI relatively seamlessly by painting a hole in the canvas with a transparent widget, composing it with the DOM element of jsdos underneath and doing some dynamic redirecting of mouse and keyboard events. The animation is from a year ago when I did a test implementation. The w95 UI now uses multiple canvases rather than one so this doesn't work directly, but probably can be adapted.

Wow, this is so cool!
Thanks for sharing!

Proper DOSBox integration works more or less. Sort of pointless though. I guess you could build a web app with this that takes in a ZIP of a DOS game, displays the files in an explorer, and runs the exes in DOSBox, for what it's worth. A bit roundabout.

I have to echo.... this is cool!

I will be playing with this very soon. 😀

Thanks for posting!

The framework is now in post-beta, with various changes. It's now closed-source with an open distributable. You can report bugs to the distrib repo.

You can feed the manual to Google AI Studio and have it build apps to spec:

GPT-4o is better at reconstructing UIs from mockup sketches, but the free version on ChatGPT doesn't have a large enough context window for the full manual.