6 ms·

Nobody seems to mention a pretty good reason also: standard western text (and content in general) is oriented right-to-left; therefore covering only one side se

Nobody seems to mention a pretty good reason also: standard western text (and content in general) is oriented right-to-left; therefore covering only one side seems to me intuitively less obstructing (we can read perfectly up to the click spot, instead of being confused by what's underneath it)

lotharbot 13y ago
In addition, it mimics what a right-handed person already does with their hand when pointing at something on a page or on a screen right in front of them, especially left-to-right text (using the right hand so it doesn't block the text). Pointing at something off to the side (outside the width of your shoulders) makes your hand tilt outwards; pointing at something directly in front makes your hand tilt inwards.
- dllthomas 13y ago
  "(outside the width of your shoulders)" Strictly speaking (after observing myself pointing) it's "outside your elbow", right? Which coincides roughly with "width of your shoulders" when your elbow is at your side or your elbow is straight, but can differ by up to the length of your upper arm...
ronaldx 13y ago
Yep: a better reason (than those given on SO) is that it's designed for minimal UX annoyance - it is least likely to cover up what you need to see while still serving its purpose; the asymmetry allows you to more quickly understand which point is important. It's instructive to observe the circumstances when the cursor changes: there is some indication of context change, but also the design of the cursor in each context is important.
- tlunter 13y ago
  A better reason in hindsight isn't the right answer. Likely they had to do it per limitations of the computers of the time. Trying to tie it to UX like a modern day designer is just wrong.
  - ronaldx 13y ago
    Absolutely disagree. As discussed in the accepted answer, vertical line cursors predate the slanted shape. Changing to a different style only makes sense as a design decision.
  - rch 13y ago
    It's probably not a good idea to underestimate the depth of the deliberations that actually took place at the time, based on a necessarily pragmatic justification for the accepted decision.
ds9 13y ago
Some of the Linux GUIs used to have a feature that, if you set the mouse to left-handed, the cursor would tilt the other way. It was very ergonomic and seemed natural (and I would like it back :)
- pbhjpbhj 13y ago
  FWIW on KDE there are several cursor sets available via the "download new content" system with left-handed primary cursors; but yes it seems you have to select them yourself rather than them being automatically applied.
- Moru 13y ago
  Most older computers had at least some utility to do this too. I used it a lot on my Atari ST's since I was having troubles with my right hand and had to switch left/right quite a lot.
PakG1 13y ago
Being pedantic, but don't you mean left-to-right?
nollidge 13y ago
Nobody mentions it because there's no evidence for that fact being an influence on the design. There's historical evidence for the other reasons. Yours is a just-so story.
- gojomo 13y ago
  Take the oldest, crudest mouse you can find. Place your right hand on it and position it to the right of your keyboard. Put your finger on its primary (or single) button. Look at that finger from your head's natural perspective. Your pointing/clicking finger will appear to be pointing in an upper-left direction, somewhere between 0' and 45' – exactly like the classic arrow-cursor. Even if a designer with contemporaneous experience explicitly denied that was n influence in choosing the orientation, I wouldn't believe them. The congruence is too strong.
  - nollidge 13y ago
    OK? I'm not sure what that's supposed to do with the post I replied to.
    - gojomo 13y ago
      Oops! Indeed, I mis-perceived your comment as being a response to lotharbot's comment, about the tilt of right-hand pointing being an influence. Sorry. However, I do think there's something more than "just so" to the darkmighty post you replied to, as well. Anything that's hand-manipulated, and especially English writing on paper, must face the risk a dominant right-hand will partially obscure it. Thus we've got longstanding design tendencies, echoed on pixel screens, that more primary labels and information tends to be above and/or to the left, and more-ignorable (or even just defer-able) information goes to the bottom/right. If damaging the visibility of some information for another purpose – as with an overlaid pointing indicator – a designer will thus have good reasons for allocating that damage in the down-right direction. (That is, tend towards obscuring the later/lesser information.) People immersed in our culture's visual design will even make that tradeoff subconsciously. So a "this makes intuitive sense" story, when reasoning about long-ago design decisions, is perfectly worthy of mention alongside other "historical evidence".
Stratoscope 13y ago
But the arrow cursor is not used for selecting text. The cursor switches to an I-beam over selectable text. This was the case as far back as the original Mac; I'm not sure which cursors were used on systems before that.
- thwarted 13y ago
  Not all on-screen text is selectable!
  - oneeyedpigeon 13y ago
    And that's a problem at least an order of magnitude more serious than which few pixels the cursor is covering.

lotharbot 13y ago

In addition, it mimics what a right-handed person already does with their hand when pointing at something on a page or on a screen right in front of them, especially left-to-right text (using the right hand so it doesn't block the text). Pointing at something off to the side (outside the width of your shoulders) makes your hand tilt outwards; pointing at something directly in front makes your hand tilt inwards.

dllthomas 13y ago
"(outside the width of your shoulders)" Strictly speaking (after observing myself pointing) it's "outside your elbow", right? Which coincides roughly with "width of your shoulders" when your elbow is at your side or your elbow is straight, but can differ by up to the length of your upper arm...

ronaldx 13y ago

Yep: a better reason (than those given on SO) is that it's designed for minimal UX annoyance - it is least likely to cover up what you need to see while still serving its purpose; the asymmetry allows you to more quickly understand which point is important. It's instructive to observe the circumstances when the cursor changes: there is some indication of context change, but also the design of the cursor in each context is important.

tlunter 13y ago
A better reason in hindsight isn't the right answer. Likely they had to do it per limitations of the computers of the time. Trying to tie it to UX like a modern day designer is just wrong.
- ronaldx 13y ago
  Absolutely disagree. As discussed in the accepted answer, vertical line cursors predate the slanted shape. Changing to a different style only makes sense as a design decision.
- rch 13y ago
  It's probably not a good idea to underestimate the depth of the deliberations that actually took place at the time, based on a necessarily pragmatic justification for the accepted decision.

ds9 13y ago

Some of the Linux GUIs used to have a feature that, if you set the mouse to left-handed, the cursor would tilt the other way. It was very ergonomic and seemed natural (and I would like it back :)

pbhjpbhj 13y ago
FWIW on KDE there are several cursor sets available via the "download new content" system with left-handed primary cursors; but yes it seems you have to select them yourself rather than them being automatically applied.
Moru 13y ago
Most older computers had at least some utility to do this too. I used it a lot on my Atari ST's since I was having troubles with my right hand and had to switch left/right quite a lot.

PakG1 13y ago

Being pedantic, but don't you mean left-to-right?

nollidge 13y ago

Nobody mentions it because there's no evidence for that fact being an influence on the design. There's historical evidence for the other reasons. Yours is a just-so story.

gojomo 13y ago
Take the oldest, crudest mouse you can find. Place your right hand on it and position it to the right of your keyboard. Put your finger on its primary (or single) button. Look at that finger from your head's natural perspective. Your pointing/clicking finger will appear to be pointing in an upper-left direction, somewhere between 0' and 45' – exactly like the classic arrow-cursor. Even if a designer with contemporaneous experience explicitly denied that was n influence in choosing the orientation, I wouldn't believe them. The congruence is too strong.
- nollidge 13y ago
  OK? I'm not sure what that's supposed to do with the post I replied to.
  - gojomo 13y ago
    Oops! Indeed, I mis-perceived your comment as being a response to lotharbot's comment, about the tilt of right-hand pointing being an influence. Sorry. However, I do think there's something more than "just so" to the darkmighty post you replied to, as well. Anything that's hand-manipulated, and especially English writing on paper, must face the risk a dominant right-hand will partially obscure it. Thus we've got longstanding design tendencies, echoed on pixel screens, that more primary labels and information tends to be above and/or to the left, and more-ignorable (or even just defer-able) information goes to the bottom/right. If damaging the visibility of some information for another purpose – as with an overlaid pointing indicator – a designer will thus have good reasons for allocating that damage in the down-right direction. (That is, tend towards obscuring the later/lesser information.) People immersed in our culture's visual design will even make that tradeoff subconsciously. So a "this makes intuitive sense" story, when reasoning about long-ago design decisions, is perfectly worthy of mention alongside other "historical evidence".

Stratoscope 13y ago

But the arrow cursor is not used for selecting text. The cursor switches to an I-beam over selectable text. This was the case as far back as the original Mac; I'm not sure which cursors were used on systems before that.

thwarted 13y ago
Not all on-screen text is selectable!
- oneeyedpigeon 13y ago
  And that's a problem at least an order of magnitude more serious than which few pixels the cursor is covering.