am going to do a rewrite for v2. For some reason custom elements and astro don't play well with each other for me. I believe this issue does not happen on Sveltekit, Next etc. but this project is very young so bear with me
I'm already looking at a v2 implementation as some others have correctly pointed this out. Since I'm already relying on customElements and web components, I should make that first class and change the framework implementations into thin wrappers. V2 will be pure js custom elements which can fall back to native html. wrappers will be even more lightweight. Hopefully this design is a strict win
Custom elements/web components are just another layer of abstraction, and I'm also wondering "V2 will be pure js custom elements which can fall back to native html" how would you fall back to native html? custom elements depends on javascript, it's end game otherwise, you could setup some fallback markup but with no interaction
Yes fallback markup is what I meant, and yes, it would not be super useful if JS is entirely disabled. I figured custom elements would be good since they can encapsulate state and styling, allowing for slightly more sophisticated behaviour compared to raw dom elements
This happened after I turned the demo site into Astro to demo all the frameworks at once. For some reason it messes with the hydration of the site, will keep debugging...
Seems like some people are confused by the design decisions of using React/Svelte/Vue and using it at runtime. The use case I had in mind was LLM interaction, which you can try in the Chat Demo (running a real llm btw!). LLMs require secure, runtime, and minimal DSL to generate interactivity, hence markdown-ui.
It's very cool and the fact it has a spec is the icing on the cake. There could easy be implementations for other runtimes. Probably, given the spec, an LLM could generate such an implementation.
Sometimes I've made little things that I envision for a specific purpose and then think that use case is only MY intention and in theory could be used more broadly. These are tiny little personal projects I'm thinking about so they never end up getting real feedback. But without a firm alternative use-case I've wondered if it's better to just frame it specifically for the one I have.
yes, that was! and I totally understand what you mean. Whilst building this I've often wondered if other people have totally unrelated use cases for it
