[kristopolous]

You can see similar things in the Apple Lisa source code as well: https://info.computerhistory.org/apple-lisa-code

The linked SO page is a page of complete speculation.

History isn't just a bunch of logical thought exercises, it's an assembling of documentation and evidence.

As far as I can see, there is no contemporaneous documentation claiming intentionality so the question remains unanswered.

A smoking gun would be a file with a name like cursor.bitmap or some code like "declare cursor_default = [ [ 1, 0 ... ] ];" from a major source (ms/xerox/apple) say, pre-1988 or so, with some comment above it explaining the rationale of why that cursor style in particular. I'd even accept a more minor source like Acorn, Digital Research, Quarterdeck, NeWS, VisiOn or MIT Athena (X).

Finding something that talks about say, lightpens and then defends the mouse cursor style in that way is working backwards from a hypothesis. It's weak and doesn't preclude other possibilities. Let's be rigorous and get it right.

[Stratoscope]

> A smoking gun would be a file with a name like cursor.bitmap or some code like "declare cursor_default = [ [ 1, 0 ... ] ];" from a major source (ms/xerox/apple) say, pre-1988 or so, with some comment above it explaining the rationale of why that cursor style in particular.

The Inside Macintosh pages from 1985 I cited above may be what you're looking for.

Especially page 158 (I-146).

It doesn't give a longwinded rationale of why you need an X/Y hotspot offset, it does much better than that. It shows you several cursors with their hotspots, so you can see why a hotspot is needed. And it lists the data structure to support it.

[kragen]

but that is 4 years later than the xerox optical mouse tech report, and from a different company which copied their default mouse pointer style from xerox. it doesn't bear on the question of whether xerox was implementing cursors without hotspot coordinates at the time that they adopted the left-leaning shape

(i suspect xerox mouse cursors always had variable hotspot coordinates because it's, what, six microseconds extra in the screen update to subtract them? and i think smalltalk-76 mouse cursors have hotspots. but 01988 or even 01985 is way too late)

[Stratoscope]

In one of my comments on the Stack Exchange answer, I linked to a couple of Xerox Alto cursors with different hotspots:

https://ux.stackexchange.com/questions/52336/why-is-the-mous...

[kragen]

this is great information, thanks! i think it confirms that even back in the mid-70s they used different hotspots

[lupire]

Within 8 thousand years, people will figure out variable length storage and processing for integers. I promise.

[kragen]

do not underestimate human stupidity