Paul Kinlan

Declarative Partial Updates and the Dynamic Markup (streaming HTML insertion) APIs have the potential to change web development.

https://declarative-partial-updates.paulkinlan-ea.deno.net/

Declarative Partial Updates experiments

May 27 at 10:44pmWeb
Show threadStephenMay 27
@paul this looks interesting, I'd love to be able to go back to it being possible to disable JavaScript again.
Show threadniu tech5d ago
@subtl @paul Declarative partial updates have long been possible without JS using my Pure HTML Out-Of-Order Streaming (#PHOOOS) technique based on Declarative Shadow DOM: https://kraksped.pl/phooos/
Pure HTML Out-Of-Order Streaming

Pure HTML Out-Of-Order Streaming without JavaScript

Show threadniu tech4d ago
@paul As for #HTMX emulation, you can do it in just a single 166-byte #HTML tag: `<iframe hidden name=htmz onload="setTimeout(()=>document.querySelector(contentWindow.location.hash||null)?.replaceWith(...contentDocument.body.childNodes))"></iframe>`, no special syntax using processing instructions `<?...>` - see #HTMZ: https://leanrada.com/htmz/
htmz - a low power tool for html

Show threadPaul Kinlan3d ago

@niutech right. But you're loading an iframe which are very heavy weight for simple interactions. I explored this in a sandboxing demo and got a huge amount of feedback from people who've been trying this for years that it's too complex to deploy at scale to users. That being said, all platform apis are like this - you can hack it, few people want to use them and want to wait until there's a platform feature. E.g, xmlhttprequest.

Nice work on what you've got.

Show threadniu tech2d ago
@paul Thanks for the reply! But I don't get it - a single hidden empty `<iframe>` with a single onload event handler in #HTMZ is very heavy weight, whereas 51KB of minified JS `<script>` without `defer` put in `<head>` thus blocking rendering as in #HTMX is considered lightweight?
Show threadPaul Kinlan2d ago
@niutech I didn't say to add in all the JavaScript of htmx