Andrea Giammarchi

> Why not doing an extra step to handle also removed nodes and eventual new nodes As couple of examples/ideas around this: * `parentEl.reorderChildren(parentEl.childNodes)` would basically be a no-op *...

I am surprised react keys are not known today, but [this is the documentation about it](https://reactjs.org/docs/lists-and-keys.html#keys). Ported to this `reorderChildren` idea, let's take a basic list as example: ```html b...

> And there is also the [replaceChildren](https://developer.mozilla.org/en-US/docs/Web/API/Element/replaceChildren) API... As that name suggests, `replaceChildren` does, in fact, **replace** children, it doesn't re-order children. With this argument, we don't need any sorting...

@chrishtr it's worth mentioning that abusing the API already allows diffing (at least per elements, not per mixed elements and text nodes (or comments) ... example: ```js const diff =...

@pygy you just rewrote my initial comment around making this API more generic, and all your algorithms are already part of udomdiff, domdiff, majinbu, or speedy Mayern approach I’ve explored,...

@titoBouzout I feel like all those methods are library authors' choice, nothing needed to land in the specs, imho. I do implement `valueOf` and `toString` subscribing because I can then...

@littledan > I don't think it's a good idea to identify signals with their contents like this. I don't know who this answer is directed to ... but if it...

if adding anything, `peek()` like mechanism is really the only thing because once that exists, everything else can decide or do whatever it likes to subscribe or not, when implicit...

@NullVoxPopuli exactly ... I would use `.peek()` you would use `.get()` ... it's all good, accordingly with our own use cases, but primitives gotta primitive, everything else should be left...

I think we changed pretty much nothing there between 2024.2.1 and 2024.3.1 around the terminal so maybe this is an upstream XTerm issue ... by any chance you checked in...