This is a really interesting way of tackling the “mostly static” case of the virtual DOM. I wonder how feasible a hybrid approach here can be, given that you do end up with a certain encapsulation here.
To be honest this is pretty close to the Angular approach, where DOM elements end up getting recycled/reused. The disadvantage in the Angular approach is working through the HTML, so you lose a lot of typing advantages. But here you still get to write your DOM in a typed space.
I’ve enjoyed working on it because many of the decisions were made automatically, by observing any issues and then simply following the types, as I’ve tried to illustrate here.
And it was a joy reading this for exactly that reason. It’s always a pleasure when people manage to turn application develoment into a logical, deductive process.
This is a really interesting way of tackling the “mostly static” case of the virtual DOM. I wonder how feasible a hybrid approach here can be, given that you do end up with a certain encapsulation here.
To be honest this is pretty close to the Angular approach, where DOM elements end up getting recycled/reused. The disadvantage in the Angular approach is working through the HTML, so you lose a lot of typing advantages. But here you still get to write your DOM in a typed space.
And it was a joy reading this for exactly that reason. It’s always a pleasure when people manage to turn application develoment into a logical, deductive process.