From Newsgroup: comp.os.linux.advocacy

I remember coming across this little quirk in the early days of OS X.
That was decades ago. And then when I asked about it a few years
later, someone confirmed it was still there. I haven’t touched a
(newer) Mac lately, so I wonder if it’s fixed yet: feel free to try
and let me know.

In normal multiwindow GUIs, windows are independent of each other in
the stacking order, even if they might be owned by the same
application. Not so on macOS: there seems to be some weird connection
between windows belonging to the same application, which doesn’t
really make sense from a UI standpoint.

To illustrate, let’s say you have two windows, A1 and A2, belonging to application A, and a single window B1 belonging to application B.
Let’s say their stacking order, from top to bottom, looks like this:

A2
B1
A1

So in this stacking, window A2 is the active (topmost) one. Suppose
the user closes window A2: which window becomes active next?

You would expect B1 to be the active window now. And in all normal
multiwindow GUIs (including regular Unix and Linux ones), this would
be true. But on macOS, it’s not: instead, the next window to become
active is the next window from the same app, namely A1!

Where is there any sense in this from a UI standpoint? I can’t think
of any. If the user chooses to stack the windows in that order, surely
they did so for a reason, so why should the system behave otherwise?
If they wanted A1 to be the next window to come forward after A2, they
can choose to stack things that way. And on normal GUIs, they have the
choice.
--- Synchronet 3.22a-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

On May 10, 2026 at 1:05:10 PM MST, "Lawrence D´Oliveiro" wrote <10tqodl$kc60$3@dont-email.me>:

I remember coming across this little quirk in the early days of OS X.
That was decades ago. And then when I asked about it a few years
later, someone confirmed it was still there. I haven’t touched a
(newer) Mac lately, so I wonder if it’s fixed yet: feel free to try
and let me know.

In normal multiwindow GUIs, windows are independent of each other in
the stacking order, even if they might be owned by the same
application. Not so on macOS: there seems to be some weird connection
between windows belonging to the same application, which doesn’t
really make sense from a UI standpoint.

To illustrate, let’s say you have two windows, A1 and A2, belonging to application A, and a single window B1 belonging to application B.
Let’s say their stacking order, from top to bottom, looks like this:

A2
B1
A1

So in this stacking, window A2 is the active (topmost) one. Suppose
the user closes window A2: which window becomes active next?

You would expect B1 to be the active window now. And in all normal multiwindow GUIs (including regular Unix and Linux ones), this would
be true. But on macOS, it’s not: instead, the next window to become
active is the next window from the same app, namely A1!

Good observation... I tend to use tabs and while I have seen this before I do not think of it much. With tabs it makes sense (you are closing a tab in the same window).

Where is there any sense in this from a UI standpoint? I can’t think
of any. If the user chooses to stack the windows in that order, surely
they did so for a reason, so why should the system behave otherwise?
If they wanted A1 to be the next window to come forward after A2, they
can choose to stack things that way. And on normal GUIs, they have the choice.

I think the idea behind macOS is having the Application be more centric than the document, at least for this -- but I agree it is not intuitive or beneficial in most cases.
--
It's impossible for someone who is at war with themselves to be at peace with you.
--- Synchronet 3.22a-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

On 2026-05-10 4:05 p.m., Lawrence D’Oliveiro wrote:

I remember coming across this little quirk in the early days of OS X.
That was decades ago. And then when I asked about it a few years
later, someone confirmed it was still there. I haven’t touched a
(newer) Mac lately, so I wonder if it’s fixed yet: feel free to try
and let me know.

In normal multiwindow GUIs, windows are independent of each other in
the stacking order, even if they might be owned by the same
application. Not so on macOS: there seems to be some weird connection
between windows belonging to the same application, which doesn’t
really make sense from a UI standpoint.

To illustrate, let’s say you have two windows, A1 and A2, belonging to application A, and a single window B1 belonging to application B.
Let’s say their stacking order, from top to bottom, looks like this:

A2
B1
A1

So in this stacking, window A2 is the active (topmost) one. Suppose
the user closes window A2: which window becomes active next?

You would expect B1 to be the active window now. And in all normal multiwindow GUIs (including regular Unix and Linux ones), this would
be true. But on macOS, it’s not: instead, the next window to become
active is the next window from the same app, namely A1!

Where is there any sense in this from a UI standpoint? I can’t think
of any. If the user chooses to stack the windows in that order, surely
they did so for a reason, so why should the system behave otherwise?
If they wanted A1 to be the next window to come forward after A2, they
can choose to stack things that way. And on normal GUIs, they have the choice.

I can confirm that it behaves in exactly the way described on my own
computer. Nevertheless, I doubt anyone would have an issue with this behaviour.
--
CrudeSausage
M4 MacBook Air
Islam is the worship of Satan
--- Synchronet 3.22a-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

On May 10, 2026 at 2:45:37 PM MST, "CrudeSausage" wrote <6a00fc81$0$23$882e4bbb@reader.netnews.com>:

On 2026-05-10 4:05 p.m., Lawrence D’Oliveiro wrote:

I remember coming across this little quirk in the early days of OS X.
That was decades ago. And then when I asked about it a few years
later, someone confirmed it was still there. I haven’t touched a
(newer) Mac lately, so I wonder if it’s fixed yet: feel free to try
and let me know.

In normal multiwindow GUIs, windows are independent of each other in
the stacking order, even if they might be owned by the same
application. Not so on macOS: there seems to be some weird connection
between windows belonging to the same application, which doesn’t
really make sense from a UI standpoint.

To illustrate, let’s say you have two windows, A1 and A2, belonging to
application A, and a single window B1 belonging to application B.
Let’s say their stacking order, from top to bottom, looks like this:

A2
B1
A1

So in this stacking, window A2 is the active (topmost) one. Suppose
the user closes window A2: which window becomes active next?

You would expect B1 to be the active window now. And in all normal
multiwindow GUIs (including regular Unix and Linux ones), this would
be true. But on macOS, it’s not: instead, the next window to become
active is the next window from the same app, namely A1!

Where is there any sense in this from a UI standpoint? I can’t think
of any. If the user chooses to stack the windows in that order, surely
they did so for a reason, so why should the system behave otherwise?
If they wanted A1 to be the next window to come forward after A2, they
can choose to stack things that way. And on normal GUIs, they have the
choice.

I can confirm that it behaves in exactly the way described on my own computer. Nevertheless, I doubt anyone would have an issue with this behaviour.

I can see having a preference against it... but not a huge deal. Still, I
think having it be document (window) centric and not application centric makes sense. Certainly not the end of the world, but a reasoned complaint and well described.

For most apps I use tabs -- to test this and make sure it still acts this way (I knew it did in the past) I had to make sure I was not using tabs in the two apps I tested (Pages and Numbers). When using tabs, of course it makes sense
to stay in the same window when you close a tab.
--
It's impossible for someone who is at war with themselves to be at peace with you.
--- Synchronet 3.22a-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

On 10 May 2026 22:36:55 GMT, Brock McNuggets wrote:

For most apps I use tabs -- to test this and make sure it still acts
this way (I knew it did in the past) I had to make sure I was not
using tabs in the two apps I tested (Pages and Numbers). When using
tabs, of course it makes sense to stay in the same window when you
close a tab.

Don’t you have a function to close the entire window, regardless of
however many tabs it may have in it?
--- Synchronet 3.22a-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

On 10 May 2026 20:34:44 GMT, Brock McNuggets wrote:

I think the idea behind macOS is having the Application be more
centric than the document, at least for this ...

That sounds to me like a retrograde step.

Back in the 1990s, there was a push for more “document-centric” rather
than “application-centric” approaches to computing. And Apple was at
the forefront of this push, as part of one of the two warring camps.

On one side, you had Microsoft with OLE 2.0, and on the other side,
you had Apple (and I think Novell as well) with OpenDoc.
--- Synchronet 3.22a-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

On May 10, 2026 at 3:41:06 PM MST, "Lawrence D´Oliveiro" wrote <10tr1i1$n8tt$3@dont-email.me>:

On 10 May 2026 20:34:44 GMT, Brock McNuggets wrote:

I think the idea behind macOS is having the Application be more
centric than the document, at least for this ...

That sounds to me like a retrograde step.

Back in the 1990s, there was a push for more “document-centric” rather than “application-centric” approaches to computing. And Apple was at
the forefront of this push, as part of one of the two warring camps.

Right. I am guessing they kept the windowing to have fewer menu "jumps" -- but even the top menu I am guessing is no longer an advantage on larger screens as it was in the '80s and '90s.

So for those who say I defend Apple right or wrong, while I do not think your issue is a big one, I do think it is a real one, and there are some cases
where I can understand being frustrated by it. I personally prefer document centric, but have to say this has rarely if ever been a big issue for me.

On one side, you had Microsoft with OLE 2.0, and on the other side,
you had Apple (and I think Novell as well) with OpenDoc.

--
It's impossible for someone who is at war with themselves to be at peace with you.
--- Synchronet 3.22a-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

On May 10, 2026 at 3:38:38 PM MST, "Lawrence D´Oliveiro" wrote <10tr1de$n8tt$2@dont-email.me>:

On 10 May 2026 22:36:55 GMT, Brock McNuggets wrote:

For most apps I use tabs -- to test this and make sure it still acts
this way (I knew it did in the past) I had to make sure I was not
using tabs in the two apps I tested (Pages and Numbers). When using
tabs, of course it makes sense to stay in the same window when you
close a tab.

Don’t you have a function to close the entire window, regardless of
however many tabs it may have in it?

Yes. And all other tabs, leaving just the one you have selected open. Works very much like tabs in most browsers. THAT is a windowing feature I use all
the time... currently have:

* Pages: 5 tabs
* Numbers: 3 tabs
* Pixelmator Pro: 2 tabs
* Safari: 7 tabs
* BBEdit: 15 tabs (though those show on the side)

Have some other apps open, but not with multiple documents. There are times I want two windows on a single app, but it is rare for me.

Having the tabs in 5 windows is a lot easier for me than having 32 separate windows (in addition to other apps open).

And that is just on this desktop... I have three desktops (though the others are not as busy).
--
It's impossible for someone who is at war with themselves to be at peace with you.
--- Synchronet 3.22a-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

On 10 May 2026 22:58:10 GMT, Brock McNuggets wrote:

... but even the top menu I am guessing is no longer an advantage on
larger screens as it was in the '80s and '90s.

That’s a controversial one. On the one hand, you have Fitts’ Law,
which says that the time to move a pointer to a target is directly
proportional to the square root of how far away from the current
pointer position it is, and inversely proportional to how big it is.

Note that square-root factor: if the screen is twice the dimensions,
that’s still only about a 40% increase in the time taken to get from
one edge to the opposite one. Think of a menu bar at the very top of
the screen as being infinitely tall, since you can’t move past it; so
you can just slam the mouse any sufficient distance in roughly the
right direction, and you’re there in a jiffy.

On the other hand, there is a definite UI fashion factor involved.
People seem more used to a menu bar attached to each window, closer to
where the action is, regardless of the fact that its thinness makes it
harder to hit. Maybe it’s because we (or at least I) don’t use menu selections that much? The most common functions end up getting invoked
via their keyboard shortcuts.
--- Synchronet 3.22a-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

On May 10, 2026 at 4:07:29 PM MST, "Lawrence D´Oliveiro" wrote <10tr33h$n8tt$7@dont-email.me>:

On 10 May 2026 22:58:10 GMT, Brock McNuggets wrote:

... but even the top menu I am guessing is no longer an advantage on
larger screens as it was in the '80s and '90s.

That’s a controversial one. On the one hand, you have Fitts’ Law,
which says that the time to move a pointer to a target is directly proportional to the square root of how far away from the current
pointer position it is, and inversely proportional to how big it is.

Note that square-root factor: if the screen is twice the dimensions,
that’s still only about a 40% increase in the time taken to get from
one edge to the opposite one. Think of a menu bar at the very top of
the screen as being infinitely tall, since you can’t move past it; so
you can just slam the mouse any sufficient distance in roughly the
right direction, and you’re there in a jiffy.

If I recall it is not exponential but logarithmic, but that does not hurt your point.

On the other hand, there is a definite UI fashion factor involved.
People seem more used to a menu bar attached to each window, closer to
where the action is, regardless of the fact that its thinness makes it
harder to hit. Maybe it’s because we (or at least I) don’t use menu selections that much? The most common functions end up getting invoked
via their keyboard shortcuts.

I have seen macOS deemphasize it over time. And add more to Context menus
where there is easier targeting and minimal travel.

A factor for me is where the window is. If I have, say, four windows in the quadrants of the screen (left/right, top/bottom) I find the menu works well
for the two on the left, but seems odd for the two on the right. Seems I am going into the "territory" of the other app. This is where I get my idea that larger screens change the cost / benefit. But that is my intuition -- have not seen research on this (nor looked).

Asking AI (which of course is not authoritative) it says:

-----
So the research picture is roughly:

* For single-window classic desktop use on moderate screens, global edge menus are often faster.
* For large displays, multi-monitor setups, and dense multi-window workflows, local/contextual controls often become more efficient overall because they preserve locality of attention and reduce cursor travel cycles.
-----

This fits my intuition well... and is about as much time as I want to put into it now, but I might look into it more, later.

Good to see another "UI junkie" who cares about these things. :)
--
It's impossible for someone who is at war with themselves to be at peace with you.
--- Synchronet 3.22a-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

Brock McNuggets <brock.mcnuggets@gmail.com> wrote in news:6a010ecf$0$24$882e4bbb@reader.netnews.com:

On May 10, 2026 at 3:38:38 PM MST, "Lawrence D´Oliveiro" wrote <10tr1de$n8tt$2@dont-email.me>:

On 10 May 2026 22:36:55 GMT, Brock McNuggets wrote:

For most apps I use tabs -- to test this and make sure it still acts
this way (I knew it did in the past) I had to make sure I was not
using tabs in the two apps I tested (Pages and Numbers). When using
tabs, of course it makes sense to stay in the same window when you
close a tab.

Don’t you have a function to close the entire window, regardless of
however many tabs it may have in it?

Yes. And all other tabs, leaving just the one you have selected open.
Works very much like tabs in most browsers. THAT is a windowing
feature I use all the time... currently have:

* Pages: 5 tabs
* Numbers: 3 tabs
* Pixelmator Pro: 2 tabs
* Safari: 7 tabs
* BBEdit: 15 tabs (though those show on the side)

Have some other apps open, but not with multiple documents. There are
times I want two windows on a single app, but it is rare for me.

Having the tabs in 5 windows is a lot easier for me than having 32
separate windows (in addition to other apps open).

And that is just on this desktop... I have three desktops (though the
others are not as busy).

That's a lot of tabs.
You have 15 BBEdit tabs open?
Are you a programmer or just doing something unusual?

--- Synchronet 3.22a-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

On May 10, 2026 at 4:47:19 PM MST, "Jim" wrote <XnsB448C94CD88DE10293847@62.164.182.22>:

Brock McNuggets <brock.mcnuggets@gmail.com> wrote in news:6a010ecf$0$24$882e4bbb@reader.netnews.com:

On May 10, 2026 at 3:38:38 PM MST, "Lawrence D´Oliveiro" wrote
<10tr1de$n8tt$2@dont-email.me>:

On 10 May 2026 22:36:55 GMT, Brock McNuggets wrote:

For most apps I use tabs -- to test this and make sure it still acts
this way (I knew it did in the past) I had to make sure I was not
using tabs in the two apps I tested (Pages and Numbers). When using
tabs, of course it makes sense to stay in the same window when you
close a tab.

Don’t you have a function to close the entire window, regardless of
however many tabs it may have in it?

Yes. And all other tabs, leaving just the one you have selected open.
Works very much like tabs in most browsers. THAT is a windowing
feature I use all the time... currently have:

* Pages: 5 tabs
* Numbers: 3 tabs
* Pixelmator Pro: 2 tabs
* Safari: 7 tabs
* BBEdit: 15 tabs (though those show on the side)

Have some other apps open, but not with multiple documents. There are
times I want two windows on a single app, but it is rare for me.

Having the tabs in 5 windows is a lot easier for me than having 32
separate windows (in addition to other apps open).

And that is just on this desktop... I have three desktops (though the
others are not as busy).

That's a lot of tabs.

Yes. I am trying to cut down -- I do occasionally run into issues with app memory. Pages / Numbers can be hogs when you have lot of images and the like.

You have 15 BBEdit tabs open?

I do... but most are just for temp items and I should just close them.

Are you a programmer or just doing something unusual?

Not a programmer -- but do a lot of editing and have text files I use for copy and paste. Several are tied to discussions / responses I have made on Usenet (really too many). Others are tied to some research I am doing for an anxiety group I am in, another for looking for into on local plumbers (I had some work done), and then even though I am not a programmer I have some projects open where I am having AI code for me. Here is one I shared before -- tied to a Daggerheart game I run:

https://rpg-cb1.pages.dev/GreenGlassDoor/

Also some notes for a person in my community I am helping with computer
issues. Used to do this as a side job, not mostly just to help people
generally for free.

I really can (and should) close some of these -- and generally do every day or two.
--
It's impossible for someone who is at war with themselves to be at peace with you.
--- Synchronet 3.22a-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

On 2026/5/10 4:05 PM, Lawrence D’Oliveiro wrote:

I remember coming across this little quirk in the early days of OS X.
That was decades ago. And then when I asked about it a few years
later, someone confirmed it was still there. I haven’t touched a
(newer) Mac lately, so I wonder if it’s fixed yet: feel free to try
and let me know.

Its not a bug, its a design choice.
--- Synchronet 3.22a-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

On May 10, 2026 at 7:56:32 PM MST, "Nick Charles" wrote <FpmcnS1ZBsAd2Jz3nZ2dnZfqnPWdnZ2d@supernews.com>:

On 2026/5/10 4:05 PM, Lawrence D’Oliveiro wrote:

I remember coming across this little quirk in the early days of OS X.
That was decades ago. And then when I asked about it a few years
later, someone confirmed it was still there. I haven’t touched a
(newer) Mac lately, so I wonder if it’s fixed yet: feel free to try
and let me know.

Its not a bug, its a design choice.

True... but one that seems hard to justify. Why be app focused and not
document / window focused.

Also not a huge deal... and with the top menu it does stop it from changing as much which might be distracting.
--
It's impossible for someone who is at war with themselves to be at peace with you.
--- Synchronet 3.22a-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

On Sun, 10 May 2026 22:56:32 -0400, Nick Charles wrote:

On 2026/5/10 4:05 PM, Lawrence D’Oliveiro wrote:

You would expect B1 to be the active window now. And in all normal
multiwindow GUIs (including regular Unix and Linux ones), this
would be true. But on macOS, it’s not: instead, the next window to
become active is the next window from the same app, namely A1!

Its not a bug, its a design choice.

What is the rationale behind that choice, can you enlighten us?
--- Synchronet 3.22a-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

On 2026-05-10 13:05, Lawrence D’Oliveiro wrote:

I remember coming across this little quirk in the early days of OS X.
That was decades ago. And then when I asked about it a few years
later, someone confirmed it was still there. I haven’t touched a
(newer) Mac lately, so I wonder if it’s fixed yet: feel free to try
and let me know.

In normal multiwindow GUIs, windows are independent of each other in
the stacking order, even if they might be owned by the same
application. Not so on macOS: there seems to be some weird connection
between windows belonging to the same application, which doesn’t
really make sense from a UI standpoint.

To illustrate, let’s say you have two windows, A1 and A2, belonging to application A, and a single window B1 belonging to application B.
Let’s say their stacking order, from top to bottom, looks like this:

A2
B1
A1

So in this stacking, window A2 is the active (topmost) one. Suppose
the user closes window A2: which window becomes active next?

You would expect B1 to be the active window now. And in all normal multiwindow GUIs (including regular Unix and Linux ones), this would
be true. But on macOS, it’s not: instead, the next window to become
active is the next window from the same app, namely A1!

Where is there any sense in this from a UI standpoint? I can’t think
of any. If the user chooses to stack the windows in that order, surely
they did so for a reason, so why should the system behave otherwise?
If they wanted A1 to be the next window to come forward after A2, they
can choose to stack things that way. And on normal GUIs, they have the choice.

By "bug" you mean:

"mode of operation I don't agree with".

Just to be completely clear.
--- Synchronet 3.22a-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

On May 10, 2026 at 8:27:58 PM MST, "Lawrence D´Oliveiro" wrote <10tribu$r55l$1@dont-email.me>:

On Sun, 10 May 2026 22:56:32 -0400, Nick Charles wrote:

On 2026/5/10 4:05 PM, Lawrence D’Oliveiro wrote:

You would expect B1 to be the active window now. And in all normal
multiwindow GUIs (including regular Unix and Linux ones), this
would be true. But on macOS, it’s not: instead, the next window to
become active is the next window from the same app, namely A1!

Its not a bug, its a design choice.

What is the rationale behind that choice, can you enlighten us?

Unlike Windows and Linux, macOS is -- in terms of windows -- application centric. It is not going to jump you to another application unless you specifically say you want to go to that application (or quit the app you are in). This is possible because macOS has the ability to have document focused apps be open with no application at all. In short, macOS separates things that Windows and Linux do not:

* closing a window
* switching apps
* quitting an app

None of those push the others. If you close the last window in an app you can still hit Command+N to get a new one. You cannot do that in Windows and Linux.

As I noted earlier, I absolutely get why you would not want this... I know I tend to think at the document level and not the app level. I think it makes sense to do so for most people ... though many do think of documents being
"in" an app -- as in not even considering the file system.
--
It's impossible for someone who is at war with themselves to be at peace with you.
--- Synchronet 3.22a-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

On 11 May 2026 05:18:19 GMT, Brock McNuggets wrote:

On May 10, 2026 at 8:27:58 PM MST, "Lawrence D´Oliveiro" wrote <10tribu$r55l$1@dont-email.me>:

On Sun, 10 May 2026 22:56:32 -0400, Nick Charles wrote:

On 2026/5/10 4:05 PM, Lawrence D’Oliveiro wrote:

You would expect B1 to be the active window now. And in all
normal multiwindow GUIs (including regular Unix and Linux ones),
this would be true. But on macOS, it’s not: instead, the next
window to become active is the next window from the same app,
namely A1!

Its not a bug, its a design choice.

What is the rationale behind that choice, can you enlighten us?

Unlike Windows and Linux, macOS is -- in terms of windows --
application centric. It is not going to jump you to another
application unless you specifically say you want to go to that
application (or quit the app you are in). This is possible because
macOS has the ability to have document focused apps be open with no application at all.

For a company which was such a stickler for intuitive UIs in its early
days, with clear cues to the user as to what is happening at any
moment, that seems very unintuitive -- obscure, even. Also, remember
they were champions of a more document-centric approach, consciously
moving away from an application-centric focus, at one point. Which
kind of compounds the mystery of why they would embrace such a
seemingly perverse idea.
--- Synchronet 3.22a-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

On May 10, 2026 at 10:42:26 PM MST, "Lawrence D´Oliveiro" wrote <10trq82$spn3$2@dont-email.me>:

On 11 May 2026 05:18:19 GMT, Brock McNuggets wrote:

On May 10, 2026 at 8:27:58 PM MST, "Lawrence D´Oliveiro" wrote
<10tribu$r55l$1@dont-email.me>:

On Sun, 10 May 2026 22:56:32 -0400, Nick Charles wrote:

On 2026/5/10 4:05 PM, Lawrence D’Oliveiro wrote:

You would expect B1 to be the active window now. And in all
normal multiwindow GUIs (including regular Unix and Linux ones),
this would be true. But on macOS, it’s not: instead, the next
window to become active is the next window from the same app,
namely A1!

Its not a bug, its a design choice.

What is the rationale behind that choice, can you enlighten us?

Unlike Windows and Linux, macOS is -- in terms of windows --
application centric. It is not going to jump you to another
application unless you specifically say you want to go to that
application (or quit the app you are in). This is possible because
macOS has the ability to have document focused apps be open with no
application at all.

For a company which was such a stickler for intuitive UIs in its early
days, with clear cues to the user as to what is happening at any
moment, that seems very unintuitive -- obscure, even.

If you think in terms of applications it is intuitive. And I can see why: If I have a hammer and I finish with one nail I still have a hammer in my hand. I
do get the reasoning, especially with the ability to have no windows open.

But I also get your point of it being a "bug" (even if not technically correct), and I agree I tend to think in terms of documents and not applications. With that said, I rarely if ever run into this as an issue... based on tab use.

Also, remember
they were champions of a more document-centric approach, consciously
moving away from an application-centric focus, at one point. Which
kind of compounds the mystery of why they would embrace such a
seemingly perverse idea.

Where do you think they are more document centric?
--
It's impossible for someone who is at war with themselves to be at peace with you.
--- Synchronet 3.22a-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

On 11 May 2026 05:50:32 GMT, Brock McNuggets wrote:

If you think in terms of applications it is intuitive.

This is all just speculation on your part. Nobody from Apple’s UI
group has ever explained the rationale behind this.

But I think I have an explanation, in the form of the usual
“historical reasons” -- it’s a hangover from the original 1980s MacOS.
If you remember, that was single-tasking. So it was quite possible to
have an app running with no windows open, and this was less likely to
confuse the user, since the desktop would be completely blank, while
the menubar still gave a hint that an application was running. The
reason for allowing this was it often took significant time for an app
to start up, so it made sense to keep it running until the user
explicitly quit it.

When multitasking was added to MacOS, this legacy behaviour simply
continued on. It was now potentially more confusing, but the saving of
time restarting an app that the user didn’t want to quit still
applied. And when MacOS was replaced by OS X/macOS, something of the
original GUI underpinnings must have been carried over. Why? For the
usual reason: it was easier to leave things that way than try to
figure out how to fix the code.

Unix systems, on the other hand, were multitasking right from the
beginning. So they never felt the need to implement such a concept.

Where do you think they are more document centric?

*Were* -- back when they were championing OpenDoc.
--- Synchronet 3.22a-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

On May 10, 2026 at 11:19:10 PM MST, "Lawrence D´Oliveiro" wrote <10trscu$tabj$2@dont-email.me>:

On 11 May 2026 05:50:32 GMT, Brock McNuggets wrote:

If you think in terms of applications it is intuitive.

This is all just speculation on your part. Nobody from Apple’s UI
group has ever explained the rationale behind this.

It is Application centric. If you think in terms of applications it makes sense.

Again: I think in terms of documents. I get the distaste for it (even if only
a minor issue, as far as I can see).

But I think I have an explanation, in the form of the usual
“historical reasons” -- it’s a hangover from the original 1980s MacOS. If you remember, that was single-tasking. So it was quite possible to
have an app running with no windows open, and this was less likely to
confuse the user, since the desktop would be completely blank, while
the menubar still gave a hint that an application was running. The
reason for allowing this was it often took significant time for an app
to start up, so it made sense to keep it running until the user
explicitly quit it.

Fair.

When multitasking was added to MacOS, this legacy behaviour simply
continued on. It was now potentially more confusing, but the saving of
time restarting an app that the user didn’t want to quit still
applied.

And there was a cost to switching apps that is not really an issue these days.

And when MacOS was replaced by OS X/macOS, something of the
original GUI underpinnings must have been carried over. Why? For the
usual reason: it was easier to leave things that way than try to
figure out how to fix the code.

I think they redid the code. But they kept the behavior to make the new OS
feel like the old. Except they moved to a browser-like Finder.

Unix systems, on the other hand, were multitasking right from the
beginning. So they never felt the need to implement such a concept.

Could be. That and the difference in menus... which does push toward thinking of apps (you can have an open app with no document)

Where do you think they are more document centric?

*Were* -- back when they were championing OpenDoc.

Ok. If I recall, the idea was to have a document which could be tied to multiple apps... you could have a spreadsheet in Pages which feels very much like Numbers, etc. Still have that to some extent with Word and Numbers and
the like, but not in the same way. But Apple, as you noted, moved away from that.
--
It's impossible for someone who is at war with themselves to be at peace with you.
--- Synchronet 3.22a-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

On 2026-05-10 22:42, Lawrence D’Oliveiro wrote:

On 11 May 2026 05:18:19 GMT, Brock McNuggets wrote:

On May 10, 2026 at 8:27:58 PM MST, "Lawrence D´Oliveiro" wrote
<10tribu$r55l$1@dont-email.me>:

On Sun, 10 May 2026 22:56:32 -0400, Nick Charles wrote:

On 2026/5/10 4:05 PM, Lawrence D’Oliveiro wrote:

You would expect B1 to be the active window now. And in all
normal multiwindow GUIs (including regular Unix and Linux ones),
this would be true. But on macOS, it’s not: instead, the next
window to become active is the next window from the same app,
namely A1!

Its not a bug, its a design choice.

What is the rationale behind that choice, can you enlighten us?

Unlike Windows and Linux, macOS is -- in terms of windows --
application centric. It is not going to jump you to another
application unless you specifically say you want to go to that
application (or quit the app you are in). This is possible because
macOS has the ability to have document focused apps be open with no
application at all.

For a company which was such a stickler for intuitive UIs in its early
days, with clear cues to the user as to what is happening at any
moment, that seems very unintuitive -- obscure, even. Also, remember
they were champions of a more document-centric approach, consciously
moving away from an application-centric focus, at one point. Which
kind of compounds the mystery of why they would embrace such a
seemingly perverse idea.

1. You mistake something you're not used to for "unintuitive".

2. I don't remember any "document-centric approach". Perhaps you could elucidate...

...or just admit it's something you made up to make your whole argument
look better?
--- Synchronet 3.22a-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

Alan <nuh-uh@nope.com> wrote:

On 2026-05-10 22:42, Lawrence D’Oliveiro wrote:

On 11 May 2026 05:18:19 GMT, Brock McNuggets wrote:

On May 10, 2026 at 8:27:58 PM MST, "Lawrence D´Oliveiro" wrote
<10tribu$r55l$1@dont-email.me>:

On Sun, 10 May 2026 22:56:32 -0400, Nick Charles wrote:

On 2026/5/10 4:05 PM, Lawrence D’Oliveiro wrote:

You would expect B1 to be the active window now. And in all
normal multiwindow GUIs (including regular Unix and Linux ones),
this would be true. But on macOS, it’s not: instead, the next
window to become active is the next window from the same app,
namely A1!

Its not a bug, its a design choice.

What is the rationale behind that choice, can you enlighten us?

Unlike Windows and Linux, macOS is -- in terms of windows --
application centric. It is not going to jump you to another
application unless you specifically say you want to go to that
application (or quit the app you are in). This is possible because
macOS has the ability to have document focused apps be open with no
application at all.

For a company which was such a stickler for intuitive UIs in its early
days, with clear cues to the user as to what is happening at any
moment, that seems very unintuitive -- obscure, even. Also, remember
they were champions of a more document-centric approach, consciously
moving away from an application-centric focus, at one point. Which
kind of compounds the mystery of why they would embrace such a
seemingly perverse idea.

1. You mistake something you're not used to for "unintuitive".

2. I don't remember any "document-centric approach". Perhaps you could elucidate...

...or just admit it's something you made up to make your whole argument
look better?

I think it makes sense if you go from document / window A to B and then
close B to expect to go back to A. At least for my way of thinking. Others might think more in terms of apps.

With tabs I don’t expect to jump to another app when I close a doc — so maybe that is tied to it.
--
Personal attacks from those who troll show their own insecurity. They
cannot use reason to show the message to be wrong so they try to feel
somehow superior by attacking the messenger.

They cling to their attacks and ignore the message time and time again.
--- Synchronet 3.22a-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

Brock McNuggets <Brock.McNuggets@gmail.com> wrote:

Alan <nuh-uh@nope.com> wrote:

On 2026-05-10 22:42, Lawrence D’Oliveiro wrote:

On 11 May 2026 05:18:19 GMT, Brock McNuggets wrote:

On May 10, 2026 at 8:27:58 PM MST, "Lawrence D´Oliveiro" wrote
<10tribu$r55l$1@dont-email.me>:

On Sun, 10 May 2026 22:56:32 -0400, Nick Charles wrote:

On 2026/5/10 4:05 PM, Lawrence D’Oliveiro wrote:

You would expect B1 to be the active window now. And in all
normal multiwindow GUIs (including regular Unix and Linux ones), >>>>>>> this would be true. But on macOS, it’s not: instead, the next
window to become active is the next window from the same app,
namely A1!

Its not a bug, its a design choice.

What is the rationale behind that choice, can you enlighten us?

Unlike Windows and Linux, macOS is -- in terms of windows --
application centric. It is not going to jump you to another
application unless you specifically say you want to go to that
application (or quit the app you are in). This is possible because
macOS has the ability to have document focused apps be open with no
application at all.

For a company which was such a stickler for intuitive UIs in its early
days, with clear cues to the user as to what is happening at any
moment, that seems very unintuitive -- obscure, even. Also, remember
they were champions of a more document-centric approach, consciously
moving away from an application-centric focus, at one point. Which
kind of compounds the mystery of why they would embrace such a
seemingly perverse idea.

1. You mistake something you're not used to for "unintuitive".

2. I don't remember any "document-centric approach". Perhaps you could
elucidate...

...or just admit it's something you made up to make your whole argument
look better?

I think it makes sense if you go from document / window A to B and then
close B to expect to go back to A. At least for my way of thinking. Others might think more in terms of apps.

With tabs I don’t expect to jump to another app when I close a doc — so maybe that is tied to it.

This is what AI says of the research. To me it makes sense and fits my experience.

——-

Cognitive research on task switching shows that adaptive behavior requires
both the ability to focus on a task and protect it from distraction
(cognitive stability) and to rapidly switch when circumstances change (cognitive flexibility) — and crucially, these are governed by a recency heuristic, where the system nudges control based on the recent history of
task demands. 

In plain terms: your most recent context is the most cognitively “loaded” one. Returning to the last-used document after a close respects that — it puts you back where your working memory was focused.

The macOS behavior (stay in same app) has a principled defense, but it’s philosophical not empirical
The macOS argument is: “If I’m working in an app and I finish using a document, I generally wish to stay in that app to create another document.” The OS treats the application as the primary actor, not the document or the prior context. 

That’s coherent within a document-centric model, but it assumes you closed that window within a workflow in that app — which isn’t the case in your scenario. You explicitly switched apps, then closed.
The key variable: user intent at the moment of close

The research on mental models is most directly applicable here. When a
user:
1. Is in App A
2. Switches to App B
3. Closes a window in App B
…the most plausible intent is “I’m done with this thing in App B.” Returning to App A honors the navigation they initiated. Staying in App B requires the OS to assume the app-switch was incidental, which is a much
bigger assumption.

Bottom line: The Windows behavior is better supported by the evidence on
how people actually track task context. The macOS behavior is internally consistent but only correct if you never switch apps before closing — a narrow and increasingly uncommon use case. The research on recency
heuristics and mental models both favor returning to the prior context
rather than anchoring to the current app.​​​​​​​​​​​​​​​​
--
Personal attacks from those who troll show their own insecurity. They
cannot use reason to show the message to be wrong so they try to feel
somehow superior by attacking the messenger.

They cling to their attacks and ignore the message time and time again.
--- Synchronet 3.22a-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

Brock McNuggets <Brock.McNuggets@gmail.com> wrote:

Brock McNuggets <Brock.McNuggets@gmail.com> wrote:

Alan <nuh-uh@nope.com> wrote:

On 2026-05-10 22:42, Lawrence D’Oliveiro wrote:

On 11 May 2026 05:18:19 GMT, Brock McNuggets wrote:

On May 10, 2026 at 8:27:58 PM MST, "Lawrence D´Oliveiro" wrote
<10tribu$r55l$1@dont-email.me>:

On Sun, 10 May 2026 22:56:32 -0400, Nick Charles wrote:

On 2026/5/10 4:05 PM, Lawrence D’Oliveiro wrote:

You would expect B1 to be the active window now. And in all
normal multiwindow GUIs (including regular Unix and Linux ones), >>>>>>>> this would be true. But on macOS, it’s not: instead, the next >>>>>>>> window to become active is the next window from the same app,
namely A1!

Its not a bug, its a design choice.

What is the rationale behind that choice, can you enlighten us?

Unlike Windows and Linux, macOS is -- in terms of windows --
application centric. It is not going to jump you to another
application unless you specifically say you want to go to that
application (or quit the app you are in). This is possible because
macOS has the ability to have document focused apps be open with no
application at all.

For a company which was such a stickler for intuitive UIs in its early >>>> days, with clear cues to the user as to what is happening at any
moment, that seems very unintuitive -- obscure, even. Also, remember
they were champions of a more document-centric approach, consciously
moving away from an application-centric focus, at one point. Which
kind of compounds the mystery of why they would embrace such a
seemingly perverse idea.

1. You mistake something you're not used to for "unintuitive".

2. I don't remember any "document-centric approach". Perhaps you could
elucidate...

...or just admit it's something you made up to make your whole argument >>> look better?

I think it makes sense if you go from document / window A to B and then
close B to expect to go back to A. At least for my way of thinking. Others >> might think more in terms of apps.

With tabs I don’t expect to jump to another app when I close a doc — so >> maybe that is tied to it.

This is what AI says of the research. To me it makes sense and fits my experience.

——-

Cognitive research on task switching shows that adaptive behavior requires both the ability to focus on a task and protect it from distraction (cognitive stability) and to rapidly switch when circumstances change (cognitive flexibility) — and crucially, these are governed by a recency heuristic, where the system nudges control based on the recent history of task demands. 

In plain terms: your most recent context is the most cognitively “loaded” one. Returning to the last-used document after a close respects that — it puts you back where your working memory was focused.

The macOS behavior (stay in same app) has a principled defense, but it’s philosophical not empirical
The macOS argument is: “If I’m working in an app and I finish using a document, I generally wish to stay in that app to create another document.” The OS treats the application as the primary actor, not the document or the prior context. 

That’s coherent within a document-centric model, but it assumes you closed that window within a workflow in that app — which isn’t the case in your scenario. You explicitly switched apps, then closed.
The key variable: user intent at the moment of close

The research on mental models is most directly applicable here. When a
user:
1. Is in App A
2. Switches to App B
3. Closes a window in App B
…the most plausible intent is “I’m done with this thing in App B.” Returning to App A honors the navigation they initiated. Staying in App B requires the OS to assume the app-switch was incidental, which is a much bigger assumption.

Bottom line: The Windows behavior is better supported by the evidence on
how people actually track task context. The macOS behavior is internally consistent but only correct if you never switch apps before closing — a narrow and increasingly uncommon use case. The research on recency
heuristics and mental models both favor returning to the prior context
rather than anchoring to the current app.​​​​​​​​​​​​​​​​

Asked another AI. To me this is interesting.

——-
The behavior you’re describing is a fundamental divergence in **Human-Computer Interaction (HCI)** philosophy. It touches on how
operating systems define the relationship between an "Application," a
"Window," and the "User’s Focus."
To determine which is "better," researchers typically look at **Mental Models**, **Fitts's Law**, and **Task Switching Costs**.

## 1. The Windows Philosophy: Document-Centric

In Windows, the window *is* the app (or a primary instance of it). When you close a window, the OS assumes you are done with that specific task and
returns you to the very last thing you were looking at, regardless of which "program" it belonged to.

* **The Research Strength:** This aligns with **Temporal Locality**.
Research shows that users often multitask between two specific documents
(e.g., a spreadsheet and a report). Returning to the "chronologically
previous" window minimizes the cognitive load of remembering where you just were.
* **The Weakness:** It can feel chaotic if you have many windows open, as
you might be "thrown" back to an app you forgot was running.

## 2. The macOS Philosophy: App-Centric
In macOS, the Application is a distinct entity that exists independently of
its windows. Closing a window doesn't "quit" the app; the focus stays
within the app's remaining windows or its menu bar.

* **The Research Strength:** This supports **Spatial and Categorical Grouping**. For power users or creative professionals, the "App" is a workspace. Staying within the same app prevents "context switching," which research identifies as a major drain on productivity. It maintains a stable environment for the next action within that same tool.

* **The Weakness:** It requires more manual "management" (e.g.,
Command+Tab) to get back to a previous app, adding an extra step to the workflow.

## Comparison Table
| Feature | Windows (Last Document) | macOS (Same App) |
|---|---|---|
| **User Intent** | Focuses on *Time* (What did I do last?) | Focuses on
*Tool* (What am I using?) |
| **Cognitive Load** | Lower for quick back-and-forth tasks. | Lower for deep-dive, single-app tasks. |
| **Predictability** | Can be "surprising" if the last app was buried. |
Highly predictable; you never leave the app. |
| **Efficiency** | Saves a click/keystroke when switching. | Requires an intentional switch to change apps. |

## Which is "Better"?
There is no definitive "winner" in academic research, as the "better"
behavior depends entirely on the **User's Expertise** and **Task
Complexity**:

* **For General/Casual Use:** The **Windows** approach is often cited as
more intuitive for beginners because it follows a "stack" logic (like
picking up the paper that was underneath the one you just threw away).

* **For Professional/Complex Workflows:** The **macOS** approach is
favored by those who treat apps as "environments." Research into **Flow
State** suggests that accidentally being kicked out of an application (like being moved from Photoshop back to a stray Chrome window) can break concentration.

### A Third Path: Linux Environments
Interestingly, some Linux desktop environments (like GNOME) allow users to toggle between these behaviors. User preference studies there suggest that people who do **"heavy" multitasking** (10+ windows) prefer the macOS style
for stability, while **"light" multitaskers** (3-4 windows) prefer the
Windows style for speed.

Ultimately, it’s a trade-off between **speed of return** (Windows) and **environmental stability** (macOS).
——-

Definite food for thought for me.
--
Personal attacks from those who troll show their own insecurity. They
cannot use reason to show the message to be wrong so they try to feel
somehow superior by attacking the messenger.

They cling to their attacks and ignore the message time and time again.
--- Synchronet 3.22a-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

On 11 May 2026 19:25:23 GMT, Brock McNuggets wrote:

Asked another AI. To me this is interesting.

——-
### A Third Path: Linux Environments
Interestingly, some Linux desktop environments (like GNOME) allow
users to toggle between these behaviors. User preference studies
there suggest that people who do **"heavy" multitasking** (10+
windows) prefer the macOS style for stability, while **"light"
multitaskers** (3-4 windows) prefer the Windows style for speed.

Did your AI say that neither Windows nor Mac offer the optimum
situation -- dynamically switching between the two paradigms as needed
-- but Linux does?
--- Synchronet 3.22a-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

On May 11, 2026 at 3:40:52 PM MST, "Lawrence D´Oliveiro" wrote <10ttltk$1g1he$4@dont-email.me>:

On 11 May 2026 19:25:23 GMT, Brock McNuggets wrote:

Asked another AI. To me this is interesting.

——-
### A Third Path: Linux Environments
Interestingly, some Linux desktop environments (like GNOME) allow
users to toggle between these behaviors. User preference studies
there suggest that people who do **"heavy" multitasking** (10+
windows) prefer the macOS style for stability, while **"light"
multitaskers** (3-4 windows) prefer the Windows style for speed.

Did your AI say that neither Windows nor Mac offer the optimum
situation -- dynamically switching between the two paradigms as needed
-- but Linux does?

What makes you think that is the optimum solution? If that is all that is changing it likely offers more cognitive overhead than giving back and forth between Mac and Windows.

As I think about this I do see the value more to how macOS does it. With the single menu and not having apps be tied to any given window (they can even
have zero documents open), and with the use of tabs I sorta get it. Right now if I am in Pages and I close a window I always stay in Pages -- but usually
the same window (just a different tab). It is not uncommon for me to close a Pages window and then want to open a new one. With Windows you would open the new one, go back to the old, and then close it.

It would be really bad in a tabbed app to close a document and jump to another app. That would be annoying and disorienting... like closing a tab in a
browser and having it jump you to MS Word!

Does not mean I rescind what I said before... I see the value of how Windows does it and think for many it is the better option (even if only slightly).
For example: say you are in Pages and you jump to the Finder to just use Quick Look on another file. Then you close the Finder window and... you get jumped
to ANOTHER Finder window. That has happened to me (more than once!) and it is not what I want. I want to go back to my Pages document. Thankfully there is a setting in System Settings to turn this off!
--
It's impossible for someone who is at war with themselves to be at peace with you.
--- Synchronet 3.22a-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

Brock McNuggets <Brock.McNuggets@gmail.com> wrote in news:6a022d22$0$22$882e4bbb@reader.netnews.com:

Brock McNuggets <Brock.McNuggets@gmail.com> wrote:

Brock McNuggets <Brock.McNuggets@gmail.com> wrote:

Alan <nuh-uh@nope.com> wrote:

On 2026-05-10 22:42, Lawrence D’Oliveiro wrote:

On 11 May 2026 05:18:19 GMT, Brock McNuggets wrote:

On May 10, 2026 at 8:27:58 PM MST, "Lawrence D´Oliveiro" wrote
<10tribu$r55l$1@dont-email.me>:

On Sun, 10 May 2026 22:56:32 -0400, Nick Charles wrote:

On 2026/5/10 4:05 PM, Lawrence D’Oliveiro wrote:

You would expect B1 to be the active window now. And in all
normal multiwindow GUIs (including regular Unix and Linux
ones), this would be true. But on macOS, it’s not: instead, >>>>>>>>> the next window to become active is the next window from the >>>>>>>>> same app, namely A1!

Its not a bug, its a design choice.

What is the rationale behind that choice, can you enlighten us?

Unlike Windows and Linux, macOS is -- in terms of windows --
application centric. It is not going to jump you to another
application unless you specifically say you want to go to that
application (or quit the app you are in). This is possible
because macOS has the ability to have document focused apps be
open with no application at all.

For a company which was such a stickler for intuitive UIs in its
early days, with clear cues to the user as to what is happening at
any moment, that seems very unintuitive -- obscure, even. Also,
remember they were champions of a more document-centric approach,
consciously moving away from an application-centric focus, at one
point. Which kind of compounds the mystery of why they would
embrace such a seemingly perverse idea.

1. You mistake something you're not used to for "unintuitive".

2. I don't remember any "document-centric approach". Perhaps you
could elucidate...

...or just admit it's something you made up to make your whole
argument look better?

I think it makes sense if you go from document / window A to B and
then close B to expect to go back to A. At least for my way of
thinking. Others might think more in terms of apps.

With tabs I don’t expect to jump to another app when I close a doc
— so maybe that is tied to it.

This is what AI says of the research. To me it makes sense and fits
my experience.

——-

Cognitive research on task switching shows that adaptive behavior
requires both the ability to focus on a task and protect it from
distraction (cognitive stability) and to rapidly switch when
circumstances change (cognitive flexibility) — and crucially, these
are governed by a recency heuristic, where the system nudges control
based on the recent history of task demands. 

In plain terms: your most recent context is the most cognitively
“loaded” one. Returning to the last-used document after a close
respects that — it puts you back where your working memory was
focused.

The macOS behavior (stay in same app) has a principled defense, but
it’s philosophical not empirical
The macOS argument is: “If I’m working in an app and I finish
using a document, I generally wish to stay in that app to create
another document.” The OS treats the application as the primary
actor, not the document or the prior context. 

That’s coherent within a document-centric model, but it assumes you
closed that window within a workflow in that app — which isn’t
the case in your scenario. You explicitly switched apps, then closed.
The key variable: user intent at the moment of close

The research on mental models is most directly applicable here. When
a user:
1. Is in App A
2. Switches to App B
3. Closes a window in App B
…the most plausible intent is “I’m done with this thing in App
B.” Returning to App A honors the navigation they initiated.
Staying in App B requires the OS to assume the app-switch was
incidental, which is a much bigger assumption.

Bottom line: The Windows behavior is better supported by the evidence
on how people actually track task context. The macOS behavior is
internally consistent but only correct if you never switch apps
before closing — a narrow and increasingly uncommon use case. The
research on recency heuristics and mental models both favor returning
to the prior context rather than anchoring to the current
app.​​​​​​​​​​​​​​​​

Asked another AI. To me this is interesting.

Interesting.
here is what Ai thinks of you snit.

https://chatgpt.com/s/t_6a0283ad2c6081919c0160005c3d8484

https://chatgpt.com/s/t_6a02851b8ea08191ba3c25d928e3b2d2
--- Synchronet 3.22a-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

On May 11, 2026 at 6:41:17 PM MST, "Chris" wrote <XnsB449DC9F858F81982100000@62.164.182.26>:

Brock McNuggets <Brock.McNuggets@gmail.com> wrote in news:6a022d22$0$22$882e4bbb@reader.netnews.com:

Brock McNuggets <Brock.McNuggets@gmail.com> wrote:

Brock McNuggets <Brock.McNuggets@gmail.com> wrote:

Alan <nuh-uh@nope.com> wrote:

On 2026-05-10 22:42, Lawrence D’Oliveiro wrote:

On 11 May 2026 05:18:19 GMT, Brock McNuggets wrote:

On May 10, 2026 at 8:27:58 PM MST, "Lawrence D´Oliveiro" wrote >>>>>>> <10tribu$r55l$1@dont-email.me>:

On Sun, 10 May 2026 22:56:32 -0400, Nick Charles wrote:

On 2026/5/10 4:05 PM, Lawrence D’Oliveiro wrote:

You would expect B1 to be the active window now. And in all >>>>>>>>>> normal multiwindow GUIs (including regular Unix and Linux
ones), this would be true. But on macOS, it’s not: instead, >>>>>>>>>> the next window to become active is the next window from the >>>>>>>>>> same app, namely A1!

Its not a bug, its a design choice.

What is the rationale behind that choice, can you enlighten us? >>>>>>>

Unlike Windows and Linux, macOS is -- in terms of windows --
application centric. It is not going to jump you to another
application unless you specifically say you want to go to that
application (or quit the app you are in). This is possible
because macOS has the ability to have document focused apps be
open with no application at all.

For a company which was such a stickler for intuitive UIs in its
early days, with clear cues to the user as to what is happening at >>>>>> any moment, that seems very unintuitive -- obscure, even. Also,
remember they were champions of a more document-centric approach,
consciously moving away from an application-centric focus, at one
point. Which kind of compounds the mystery of why they would
embrace such a seemingly perverse idea.

1. You mistake something you're not used to for "unintuitive".

2. I don't remember any "document-centric approach". Perhaps you
could elucidate...

...or just admit it's something you made up to make your whole
argument look better?

I think it makes sense if you go from document / window A to B and
then close B to expect to go back to A. At least for my way of
thinking. Others might think more in terms of apps.

With tabs I don’t expect to jump to another app when I close a doc
— so maybe that is tied to it.

This is what AI says of the research. To me it makes sense and fits
my experience.

——-

Cognitive research on task switching shows that adaptive behavior
requires both the ability to focus on a task and protect it from
distraction (cognitive stability) and to rapidly switch when
circumstances change (cognitive flexibility) — and crucially, these
are governed by a recency heuristic, where the system nudges control
based on the recent history of task demands. 

In plain terms: your most recent context is the most cognitively
“loaded” one. Returning to the last-used document after a close
respects that — it puts you back where your working memory was
focused.

The macOS behavior (stay in same app) has a principled defense, but
it’s philosophical not empirical
The macOS argument is: “If I’m working in an app and I finish
using a document, I generally wish to stay in that app to create
another document.” The OS treats the application as the primary
actor, not the document or the prior context. 

That’s coherent within a document-centric model, but it assumes you
closed that window within a workflow in that app — which isn’t
the case in your scenario. You explicitly switched apps, then closed.
The key variable: user intent at the moment of close

The research on mental models is most directly applicable here. When
a user:
1. Is in App A
2. Switches to App B
3. Closes a window in App B
…the most plausible intent is “I’m done with this thing in App
B.” Returning to App A honors the navigation they initiated.
Staying in App B requires the OS to assume the app-switch was
incidental, which is a much bigger assumption.

Bottom line: The Windows behavior is better supported by the evidence
on how people actually track task context. The macOS behavior is
internally consistent but only correct if you never switch apps
before closing — a narrow and increasingly uncommon use case. The
research on recency heuristics and mental models both favor returning
to the prior context rather than anchoring to the current
app.​​​​​​​​​​​​​​​​

Asked another AI. To me this is interesting.

Interesting.
here is what Ai thinks of you snit.

https://chatgpt.com/s/t_6a0283ad2c6081919c0160005c3d8484

https://chatgpt.com/s/t_6a02851b8ea08191ba3c25d928e3b2d2

Notice it does not offer MIDs or quotes. What it is doing is being impacted by those of you with SDS. That is a YOU issue... nothing to do with me.
--
It's impossible for someone who is at war with themselves to be at peace with you.
--- Synchronet 3.22a-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

On 11 May 2026 06:44:28 GMT, Brock McNuggets wrote:

On May 10, 2026 at 11:19:10 PM MST, "Lawrence D´Oliveiro" wrote <10trscu$tabj$2@dont-email.me>:

On 11 May 2026 05:50:32 GMT, Brock McNuggets wrote:

If you think in terms of applications it is intuitive.

This is all just speculation on your part. Nobody from Apple’s UI
group has ever explained the rationale behind this.

It is Application centric. If you think in terms of applications it
makes sense.

No, only if you think of it the way your rationalization as an Apple
apologist was laid out, then it seems a plausible hypothesis, nothing
more. But to claim that is the actual reason behind it is just a
circular argument, with no actual evidence behind it.
--- Synchronet 3.22a-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

On 11 May 2026 22:53:04 GMT, Brock McNuggets wrote:

On May 11, 2026 at 3:40:52 PM MST, "Lawrence D´Oliveiro" wrote <10ttltk$1g1he$4@dont-email.me>:

On 11 May 2026 19:25:23 GMT, Brock McNuggets wrote:

Asked another AI. To me this is interesting.

——-
### A Third Path: Linux Environments
Interestingly, some Linux desktop environments (like GNOME) allow
users to toggle between these behaviors. User preference studies
there suggest that people who do **"heavy" multitasking** (10+
windows) prefer the macOS style for stability, while **"light"
multitaskers** (3-4 windows) prefer the Windows style for speed.

Did your AI say that neither Windows nor Mac offer the optimum
situation -- dynamically switching between the two paradigms as
needed -- but Linux does?

What makes you think that is the optimum solution? If that is all
that is changing it likely offers more cognitive overhead than
giving back and forth between Mac and Windows.

“Optimum” because you can do it all on one system, without having to
have two machines on your desk. Or rebooting a single machine between
two modes. Or figuring out how to virtualize one within the other, or
both under a third. Or other such hang-a-bag-on-the-side bandaids.

“Optimum” because you can do it within the same user session.

“Optimum”, in short, because it requires the least work on the part of
the user.

As I think about this I do see the value more to how macOS does it.
With the single menu and not having apps be tied to any given window
(they can even have zero documents open), and with the use of tabs I
sorta get it.

But then, you could have the situation where the frontmost window does
not belong to the frontmost app -- the menu bar says one thing, the
visible window says something else.

Resiling from what your AI said, that both approaches have their
benefits?
--- Synchronet 3.22a-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

Lawrence D´Oliveiro <ldo@nz.invalid> wrote:

On 11 May 2026 06:44:28 GMT, Brock McNuggets wrote:

On May 10, 2026 at 11:19:10 PM MST, "Lawrence D´Oliveiro" wrote
<10trscu$tabj$2@dont-email.me>:

On 11 May 2026 05:50:32 GMT, Brock McNuggets wrote:

If you think in terms of applications it is intuitive.

This is all just speculation on your part. Nobody from Apple’s UI
group has ever explained the rationale behind this.

It is Application centric. If you think in terms of applications it
makes sense.

No, only if you think of it the way your rationalization as an Apple apologist was laid out, then it seems a plausible hypothesis, nothing
more. But to claim that is the actual reason behind it is just a
circular argument, with no actual evidence behind it.

How do you figure? It works in a consistent, app-focused way. The single
menu and tabs would make it in some ways awkward to work like Windows —
which is not to say the Windows way does not have advantages. Given my use
of tabs are rarely multiple Windows per app it’s not much of an issue for
me.
--
Personal attacks from those who troll show their own insecurity. They
cannot use reason to show the message to be wrong so they try to feel
somehow superior by attacking the messenger.

They cling to their attacks and ignore the message time and time again.
--- Synchronet 3.22a-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

Lawrence D´Oliveiro <ldo@nz.invalid> wrote:

On 11 May 2026 22:53:04 GMT, Brock McNuggets wrote:

On May 11, 2026 at 3:40:52 PM MST, "Lawrence D´Oliveiro" wrote
<10ttltk$1g1he$4@dont-email.me>:

On 11 May 2026 19:25:23 GMT, Brock McNuggets wrote:

Asked another AI. To me this is interesting.

——-
### A Third Path: Linux Environments
Interestingly, some Linux desktop environments (like GNOME) allow
users to toggle between these behaviors. User preference studies
there suggest that people who do **"heavy" multitasking** (10+
windows) prefer the macOS style for stability, while **"light"
multitaskers** (3-4 windows) prefer the Windows style for speed.

Did your AI say that neither Windows nor Mac offer the optimum
situation -- dynamically switching between the two paradigms as
needed -- but Linux does?

What makes you think that is the optimum solution? If that is all
that is changing it likely offers more cognitive overhead than
giving back and forth between Mac and Windows.

“Optimum” because you can do it all on one system, without having to
have two machines on your desk.

By that logic I can run Windows and Linux on top of macOS.

Or rebooting a single machine between
two modes. Or figuring out how to virtualize one within the other, or
both under a third. Or other such hang-a-bag-on-the-side bandaids.

The cost of switching back and forth is likely larger than the cost of
either solution. Curious if when you pick on a system if it changes other things like the top menu and app tabs, and the application switcher. On
both macOS and Windows these things are coordinated and made consistent to benefit usability.

“Optimum” because you can do it within the same user session.

As I said: likely more cognitive overhead than Windows or macOS.

“Optimum”, in short, because it requires the least work on the part of the user.

How do you figure? What you describe — going back and forth — means more mode switching. Of course you can set it and forget it which. If done well, could be a benefit. If not it could add complications most users are not equipped to understand well. Both MS and Apple have usability research to
base decisions on.

As I think about this I do see the value more to how macOS does it.
With the single menu and not having apps be tied to any given window
(they can even have zero documents open), and with the use of tabs I
sorta get it.

But then, you could have the situation where the frontmost window does
not belong to the frontmost app -- the menu bar says one thing, the
visible window says something else.

The window would not be active. And the menu backs that (I wish there was
more visual difference).

Resiling from what your AI said, that both approaches have their
benefits?

They do. The only time I can think when it’s been a detriment to me is closing Finder windows and that is rarely an issue. But it still is one.
Given the consistency with the menu and tabs, and given how I sometimes
close a window and then quickly open a new one, I also see advantages to
the app based paradigm.
--
Personal attacks from those who troll show their own insecurity. They
cannot use reason to show the message to be wrong so they try to feel
somehow superior by attacking the messenger.

They cling to their attacks and ignore the message time and time again.
--- Synchronet 3.22a-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

On 12 May 2026 03:07:20 GMT, Brock McNuggets wrote:

Lawrence D´Oliveiro <ldo@nz.invalid> wrote:

On 11 May 2026 22:53:04 GMT, Brock McNuggets wrote:

On May 11, 2026 at 3:40:52 PM MST, "Lawrence D´Oliveiro" wrote
<10ttltk$1g1he$4@dont-email.me>:

Did your AI say that neither Windows nor Mac offer the optimum
situation -- dynamically switching between the two paradigms as
needed -- but Linux does?

What makes you think that is the optimum solution? If that is all
that is changing it likely offers more cognitive overhead than
giving back and forth between Mac and Windows.

“Optimum” because you can do it all on one system, without having to
have two machines on your desk.

By that logic I can run Windows and Linux on top of macOS.

But ... why would you need to?

Why did your AI suggest that Linux could bridge both worlds, but macOS
could not? After all, macOS is “Unix”, isn’t it? So what does it take
to manage this kind of integrated dual-mode operation, that Linux can
handle, that a mere “Unix” system cannot?
--- Synchronet 3.22a-Linux NewsLink 1.2