Maybe a hot take, but I do think that this is correct for trackpads. Not for mice though.

I prefer the direction apple doesn't call "natural" and I don't appreciate the implication my preference is "unnatural" because of it.

The whole point is that people have different mental models of these things.

I was using a trackpad before I'd ever owned a device with touchscreen, and so my concept for trackpad is based on mouse:

• Moving the cursor with a finger is like moving a mouse on the desk - movement agrees with direction
• Scrolling is like treating the trackpad as a 'virtual mouse wheel' - towards you moves the content up.

And as for why I think a mouse wheel itself should behave like that? It's because my mental model of the mouse wheel from the beginning was that it's a literal physical cog like a roller pressed "against" the view content - and so when the top of the roller moves down the underside of the roller moves the content up.

All the people who grew up with phones will insist my preference is wrong and freaky and I don't blame them because I think theirs is.

There's no actual wrong or right here, just whatever collection of mental models built up since childhood our life paths caused us to form.

This sometimes reminds me of the "invert Y-axis" for camera controls in video games, where someone's preference can basically determined by the question "Are you old enough to have played the original MS Flight Simulator as a child?".

Someone's preference typically reflects their tactile experiences that they made with their hardware early on, and once something feels "right", there's usually no convincing them it's not (and indeed no point to it). It's an artificial thing you have to get used to, one way or another, and there's really no right or wrong way to go about it.

Isn't this the default Mac behavior?

Yes, scrolling moves the content. As if it's being gripped by the bottom of the wheel I'm rotating with my finger from the top.

Imagine a toy monster truck on a sheet of paper. Push the wheels forward, page moves backward.

Inverse scrolling is like a high-heeled rubber shoe attached to the tip of my finger advancing the page. There's some logic to it but that's not what a wheel is like unless it's an Apple mouse's touch wheel. But they can't be trusted to design a mouse: one had a circular cross-section, another charged from the bottom.

The big question is: which is natural? Fortunately, mice exist in nature so there's a way to find out. Hold one tightly like this, release the front hand... Which way will the cloth move?

Just set it to whatever you prefer. Who cares which direction other people scroll?

On a touch screen, this movement is natural because you're interacting with the content on the screen, not the view.

For a mouse, where you're likely navigating multiple views across different windows, you're manipulating the view instead of the content and in that context "natural" is an improper term to describe the motion.

It depends entirely on the convention of the environment.

In ancient times when planes had a simple stick for controls, you pull back to go up and push forward to go down.

what else is the scroll wheel for?

This is a heated debate in computer assisted modeling/drafting. I forgot between solidworks, unigraphix, or one of the autodesk softwares which had a reveresed scroll default but it was different and a lot of people were trained opposite of everyone else.

Thats before you even get into a debate about how a spacemouse is supposed to be set up.

The last time I was thinking that way was when I was 5 years or so old and first introduced to a mouse and playing around on the "die Sendungen mit der Maus" Website. Im not using Windows that intensive anymore, is that fucking real?

If you scroll UK, should the scrollbar go up? Or should the page go up?

I have long preferred "natural" ("Inverted" in KDE) for touchpads, and the opposite for mouse scroll wheels. Until reading this thread, I don't think I've thought about it that much.

I just experimented with switching, and I dunno... I might try using inverted for both for a while. They both feel fine for me after a few seconds of orientation.

Well, duh.