Eagster

> technically a file is not strictly necessary, however it is motivating. Makes sense. Definitely worth keeping then, good call. > I am skeptical that assets as entities will really...

I made a discussion post #18949 that is an alternative to this. I don't have a crazy strong opinion, but I think my proposal is worth investigating before merging this....

Just a head's up that @andriyDev and I have come up with a followup to this PR. See [here](https://discord.com/channels/691052431525675048/749332104487108618/1367506312455258122). I think we need to start with merging this pr, but...

@urben1680 Yeah, I think a new method is the best. I would personally want `cmp_age_approx` (or similar) that gives an `Ordering`, and then some utility methods like `is_newer_than`, etc built...

My vote then is to remove the `Ord` impl in this pr and revisit it later. That sounds like the most controversial part of this. We should fix the bug...

> Thinking of it this probably should have the returned ordering reversed _or_ not be called "cmp_**age**_approx". 🤔 `First` is "older" than `First + 1`. That's fair. I don't have...

I have not looked very closely at the implementation yet, but here's a few thoughts. This is *probably* the best way to implement component scopes. In my own ecs engine,...

@caspark, you mentioned earlier that getting back some version of `alloc_at` would be nice for cloning worlds. I haven't been keeping up with this thread, but I think #19451 would...

> Hah, yeah, that's what I did first :). @Victoronz convinced me that `Box` and `Deref` are better here: [#18162 (comment)](https://github.com/bevyengine/bevy/pull/18162#issuecomment-2701890021). That makes a lot of since. I still hate...

 Here's what it looks like to me ^ I would just remove the lint, but I'm not familiar at all with the render code. Could be valid?