Time to post the evergreen thoughts of Alan Kay from his 1997 OOPSLA talk: https://youtu.be/oKg1hTOQXoY?t=1491 https://youtu.be/oKg1hTOQXoY?t=1491 We've spend

Time to post the evergreen thoughts of Alan Kay from his 1997 OOPSLA talk: https://youtu.be/oKg1hTOQXoY?t=1491 https://youtu.be/oKg1hTOQXoY?t=1491 We've spend the last 25 years hacking a document renderer and scripting language aimed at 5 line form validation logic into the be-all and end-all of modern application development. Browsers sit at 10s of millions of lines of code, with thousands of people working on them. We depend on a standardization committee to have things like the common controls you could find in a Smalltalk-80 UI, and these are still insufficient for many purposes such that developers end up cramming together a million diffs to create the controls they want, in the style they want. The writing was on the wall once we had a canvas element, aka a drawing context, and webassembly, aka a vm target you can compile to. We can keep building up giant garbage heap of complexity and saying that's what we meant to do ( https://youtu.be/oKg1hTOQXoY?t=814 https://youtu.be/oKg1hTOQXoY?t=814 ), or we can take step back and ask if we can provide simpler lower level tools on top of that. With that said, I feel somewhat ambivalent about all of this. Of course the web is an open standard and has provided an incredible explosion of knowledge and utility for everyone. Any sweeping changes to that should elicit feelings of caution I think. Things like accessibility are important, although I think that enough people recognize this that we don't need to worry about it being left behind with this approach and I would also say the ocean of divs approach isn't exactly great for accessibility either. Will this take off and be the first of a new kind of application development? Maybe, or maybe we'll all look back on it in 10 years and wonder what the hype was. For my part I do feel like some sort of change is going to happen, the current approach of piling more crap on top of the current html app dev paradigm feels unscalable, while the strengths of how it allows for app distribution feel like something that we can never get away from. Will Flutter retain those strengths and leave behind the weaknesses? I hope so, and it make me excited for the future of web development in a way I haven't previously.