undo window move/resize?

[J. P. Gilliver (John)]

(I'm on 7, but the problem applies to XP and 10 too, and the solution -
if there is one! - may be universal.)

Occasionally, I accidentally either move a window, or drag one of its
sides, by mistake - while doing (or intending to) something else with
the mouse pointer, it catches on an edge, or more likely a title bar.
(Unlike a lot of people, I _don't_ operate most of my windows
full-screen most of the time, so their relative sizes and positions are
important to me.)

I discovered fairly recently that _if I notice while the movement is in
process_ (I just see the _outline_ move, i. e. I've still got the "mouse
button pressed" [actually I use a touchpad, but similar applies]), I can
cancel the move/resize by - keeping the button pressed and - pressing
the Esc key. But if I don't - anyone know of a way of undoing a window
move/resize? (Ctrl-Z doesn't seem to work.)
--
J. P. Gilliver. UMRA: 1960/<1985 MB++G()AL-IS-Ch++(p)Ar@T+H+Sh0!:`)DNAf

to sing like an injured seagull takes real skill. - Andrew Collins on Meryl
Streep as Florence Foster Jenkins, RT 2018/8/4-10

[Mayayana]

"J. P. Gilliver (John)" <G6...@255soft.uk> wrote

| I discovered fairly recently that _if I notice while the movement is in
| process_ (I just see the _outline_ move, i. e. I've still got the "mouse
| button pressed" [actually I use a touchpad, but similar applies]), I can
| cancel the move/resize by - keeping the button pressed and - pressing
| the Esc key. But if I don't - anyone know of a way of undoing a window
| move/resize? (Ctrl-Z doesn't seem to work.)

Seems very unlikely. Move and resize are not up to
Windows. Programs handle it. A program may remember
where it was last and resume that position if the window
has been minimized or the program closed. But that's
up to the program author to write the code for storing
coordinates and then retrieving the data.

I've never heard of an Undo.
That would require a program to store an undo history of
window coordinates, on the off chance that someday,
someone would care. If they went to that trouble then also,
of course, there would be a menu item, such as View ->
Last Window Position.

[J. P. Gilliver (John)]

On Mon, 23 May 2022 at 08:13:58, Mayayana <maya...@invalid.nospam>
wrote (my responses usually FOLLOW):
[]

> Seems very unlikely. Move and resize are not up to
>Windows. Programs handle it. A program may remember

Interesting point.

>where it was last and resume that position if the window
>has been minimized or the program closed. But that's
>up to the program author to write the code for storing
>coordinates and then retrieving the data.

Now you mention it, they _do_ remember (their position and size) when
minimised - or maximised for that matter. (At least, I can't think for
the moment of one that doesn't [obviously some have a fixed size].)

>
> I've never heard of an Undo.
>That would require a program to store an undo history of
>window coordinates, on the off chance that someday,

Well, only one would be needed, not a full history. And a we've just
discussed - though I hadn't thought about it until we did - they _do_
remember that through a minimise/maximise followed by a restore. Or
something does - might be Windows.

>someone would care. If they went to that trouble then also,
>of course, there would be a menu item, such as View ->
>Last Window Position.
>

I agree, I've never seen that. Though not that it'd need a full menu
entry.

>
--
J. P. Gilliver. UMRA: 1960/<1985 MB++G()AL-IS-Ch++(p)Ar@T+H+Sh0!:`)DNAf

Never be led astray onto the path of virtue.

[Big Al]

I've found that cntl-Z, the undo for text editing will sometimes undo a file copy. At least it does on Linux, might be
worth trying on you window move/resize.

Al

[J. P. Gilliver (John)]

On Mon, 23 May 2022 at 08:44:12, Big Al <Be...@invalid.com> wrote (my
responses usually FOLLOW):
[]

>I've found that cntl-Z, the undo for text editing will sometimes undo a
>file copy. At least it does on Linux, might be worth trying on you
>window move/resize.
>
>Al

It certainly does undo a move in Windows - at least to another location
on the same partition; I haven't tried it on a move between
partitions/drives or on a copy. It undoes a rename, too. In both cases,
only if fairly soon after the move/rename - I don't mean in time, but in
other actions. But unfortunately it doesn't for window move/resize.

Assuming Linux has a similar facility for moving/resizing windows (by
dragging the title bar or edges), does ^Z work there?

--
J. P. Gilliver. UMRA: 1960/<1985 MB++G()AL-IS-Ch++(p)Ar@T+H+Sh0!:`)DNAf

Quantum particles: the dreams that stuff is made of - David Moser

[Mayayana]

"J. P. Gilliver (John)" <G6...@255soft.uk> wrote

| they _do_
| remember that through a minimise/maximise followed by a restore. Or
| something does - might be Windows.
|

No, that's what I'm saying. I write software. If I don't track and
store window positions then windows will typically come up in
the upper left, because that's the coordinates 0,0. But nearly all
programs are designed to track their window positions. I code mine
to open center-screen unless I have last-opened coordinates. Then
I use those. Some programs do other things. Firefox, for example,
seems to check open instances and display multiple windows in
cascade fashion.

| >someone would care. If they went to that trouble then also,
| >of course, there would be a menu item, such as View ->
| >Last Window Position.
| >
| I agree, I've never seen that. Though not that it'd need a full menu
| entry.

You might not need it. You might be happy with Ctrl + Alt + 1 or
some such. But then people can't do it with a mouse and there's no
way to advertise the functionality. It would be very dumb to go
to the trouble to code a window position history, when only some
guy in England wants it. It would be far more dumb to go to that
trouble and then not advertise the functionality through the menu.
No one would ever find out about it! You could put it in the help,
but ono one reads help. The general public lives in a mindset of
TL;DR. Will it fill my stomach or give me an orgasm? Then TL;DR.

There's tradition, though MS have broken a lot of their
own rules in recent years. But the general idea is to put things on
the menu, also put it on the context menu if appropriate, and at least
for standard operations like Copy, also implement standard shortcuts
like Ctrl + C. But those are left over from DOS days. Most people
don't know about them and they're not discoverable.

I use context menu whenever possible, almost never using
keyboard shortcuts, but I have noticed that in
some older programs, like Paint Shop Pro 5, context menu wasn't
implemented, despite the menu listing keyboard shortcuts for most things.
So I guess the historical development was keyboard in DOS, which was
then carried over to Windows for compatibility. Then in Windows, having
a GUI, a menu was introduced. Then perhaps only later, context menu
was made standard.

But even that can vary. One especially maddening detail for me is
Cut and Copy. Those operations should cut or copy the selected
text in the window where text is selected. That text window is the
context. But in some programs, the right-click must be done exactly
on top of the selected text. If it isn't, the right click is treated as a
left click, unselecting all text!

[VanguardLH]

J. P. Gilliver (John) wrote:

> (I'm on 7, but the problem applies to XP and 10 too, and the solution -
> if there is one! - may be universal.)
>
> Occasionally, I accidentally either move a window, or drag one of its
> sides, by mistake - while doing (or intending to) something else with
> the mouse pointer, it catches on an edge, or more likely a title bar.
> (Unlike a lot of people, I _don't_ operate most of my windows
> full-screen most of the time, so their relative sizes and positions are
> important to me.)
>
> I discovered fairly recently that _if I notice while the movement is in
> process_ (I just see the _outline_ move, i. e. I've still got the "mouse
> button pressed" [actually I use a touchpad, but similar applies]), I can
> cancel the move/resize by - keeping the button pressed and - pressing
> the Esc key. But if I don't - anyone know of a way of undoing a window
> move/resize? (Ctrl-Z doesn't seem to work.)

You already know about using the Esc key to abort the operation while it
is *in progress* (the mouse button is still pressed when Esc is hit).
That the window is getting moved or resized should be obvious before you
release the mouse button hence the Esc key is how you abort a
noncommitted operation. As Maya notes, there is no Undo (Ctrl+Z) for a
move/size operation after the operation has been committed. Too late
after releasing the mouse. Once you see the window resizing or moving,
do NOT release the mouse button! With a pending operation, you can
Esc(ape) it. When you have an Oops, keep the mouse button pressed to
let you un-Oops.

You have to break the habit of releasing the mouse button to afterward
seek how to revert the move or resize of a window. Don't release until
you're satisfied with the change, and don't release if you want to use
Esc to abort. Instinct has you back off (release).

Personally I don't rely on absolute positioning or sizing to stay fixed
for windows. Too many times when loading an app, the window opens
somewhere else than before. You cannot guarantee the window will always
open in the same spot unless the program itself lets you configure its
initial X-Y coordinates for the upper left corner along with size (which
could still run into a problem if the screen size is smaller than the
program's fixed window size). Plus, even if a program gave you that
option, it would only get enforced when the program's window was opened,
not prevent you later from moving or resizing that window. Trying to
load windows always in the same spot and always with the same size is an
exercise in futility if the program doesn't track a location history
even if just one history entry of where was the window and its size when
the window was last closed.

If you want windows to always open in the same spot, you need something
more for window management than Windows affords to the end user, like a
better window manager. I've heard of, but not used Desksoft's
WindowManager (https://www.desksoft.com/WindowManager.htm). I'm sure I
could find other window managers if I bothered to hunt for them. That
window manager lets you configure where a window opens (don't know about
sizing), and even lock the window so it cannot be moved or resized.
I've also heard of, but not used, ShellFolderFix
(https://www.majorgeeks.com/files/details/shellfolderfix.html).
ShellFolderFix works with some programs (Firefox, LibreOffice, Revo
Uninstaller), but not with some other programs (Thunderbird, CCleaner,
Irfanview). Up to you to discover with which program's window it can
manage their positioning.

Another option is to use a hotkey macro tool, like AutoHotkey or AutoIt,
that can position a window to a fixed coordinate for the upper left
corner and set a size on the window. You'd define a single hotkey that
has multiple rules depending on the titlebar's string (window name), and
each rule or condition would position that window for that program at
the same spot. I haven't used AutoHotkey for many years, but recall it
can interrogate the string for the titlebar, and it could position and
size the window. You'd have to define multiple conditions, or tests, in
the macro that would set position and size based on the titlebar. Alas,
many programs use the same titlebar string for multiple window opens by
the same program, so there's no way to distinguish one program's window
from another that it opened.

In a similar vein of using a macro with multiple conditions to test on
the titlebar string for where to position and size the currently focused
window, looks like WinSize2 can do that for you; see
http://winsize2.sourceforge.net/en/index.htm. Instead of having to
write a script with multiple conditions testing the titlebar string to
know on which window to effect positioning and sizing actions, you
record a table of windows in WinSize2 to do the same. However, it looks
like this tool triggers when a object.window is created; i.e., when a
window opens is when WinSize2 exercises its rules.

If you have Nirsoft's nircmd.exe tool, you can use it to set the
coordinates and size of an opened window by running:

nircmd.exe win setsize title "<titlebar>" <x> <y> <width> <height>

<titlebar> is the string on which to test in the titlebars of the
currently opened windows. <x> and <y> are the coordinates for the upper
left corner of the window. <width> and <height> are obvious. Values
are in pixels. No, the angle brackets are not part of the arguments.
Alas, Nirsoft didn't bother to document the win setsize arguments at
http://www.nirsoft.net/utils/nircmd.html; however, the nircmd.chm help
file does mention the win arg and its setsize parameter.

You could have a "Restore Windows" folder with a bunch of shortcuts, or
a toolbar in the Windows Taskbar, or other means of storing a multitude
of nircmds for the windows you are likely to open and might need to
reposition and resize.

[Ken Blake]

On Mon, 23 May 2022 11:27:18 +0100, "J. P. Gilliver (John)"
<G6...@255soft.uk> wrote:

>(I'm on 7, but the problem applies to XP and 10 too, and the solution -
>if there is one! - may be universal.)
>
>Occasionally, I accidentally either move a window, or drag one of its
>sides, by mistake - while doing (or intending to) something else with
>the mouse pointer, it catches on an edge, or more likely a title bar.
>(Unlike a lot of people, I _don't_ operate most of my windows
>full-screen most of the time,

Same for me. I almost never run anything full screen.

>so their relative sizes and positions are
>important to me.)
>
>I discovered fairly recently that _if I notice while the movement is in
>process_ (I just see the _outline_ move, i. e. I've still got the "mouse
>button pressed" [actually I use a touchpad, but similar applies]), I can
>cancel the move/resize by - keeping the button pressed and - pressing
>the Esc key. But if I don't - anyone know of a way of undoing a window
>move/resize? (Ctrl-Z doesn't seem to work.)

Can't you just drag it back where it was? Or is that not precise
enough for you? That's what I do.

[Ken Blake]

On Mon, 23 May 2022 13:35:15 +0100, "J. P. Gilliver (John)"
<G6...@255soft.uk> wrote:

>On Mon, 23 May 2022 at 08:13:58, Mayayana <maya...@invalid.nospam>
>wrote (my responses usually FOLLOW):
>[]
>> Seems very unlikely. Move and resize are not up to
>>Windows. Programs handle it. A program may remember
>
>Interesting point.
>
>>where it was last and resume that position if the window
>>has been minimized or the program closed. But that's
>>up to the program author to write the code for storing
>>coordinates and then retrieving the data.
>
>Now you mention it, they _do_ remember (their position and size) when
>minimised - or maximised for that matter. (At least, I can't think for
>the moment of one that doesn't [obviously some have a fixed size].)

For some strange reason, one of the programs I always run (Directory
Opus) doesn't remember where it was if I reboot.

It's not a big deal, but it's a minor annoyance.

[R.Wieser]

Mayayana,

> Firefox, for example, seems to check open instances and display
> multiple windows in cascade fashion.

Try using 0x8000, 0x8000 for the windows/dialogs initial position.

Regards,
Rudy Wieser

[Big Al]

On 5/23/22 09:02, this is what J. P. Gilliver (John) wrote:
> On Mon, 23 May 2022 at 08:44:12, Big Al <Be...@invalid.com> wrote (my responses usually FOLLOW):
> []
>> I've found that cntl-Z, the undo for text editing will sometimes undo a file copy.   At least it does on Linux, might
>> be worth trying on you window move/resize.
>>
>> Al
>
> It certainly does undo a move in Windows - at least to another location on the same partition; I haven't tried it on a
> move between partitions/drives or on a copy. It undoes a rename, too. In both cases, only if fairly soon after the
> move/rename - I don't mean in time, but in other actions. But unfortunately it doesn't for window move/resize.
>
> Assuming Linux has a similar facility for moving/resizing windows (by dragging the title bar or edges), does ^Z work there?

No it does not restore if you snap a window to the side.

[Mayayana]

"Ken Blake" <K...@invalid.news.com> wrote

> For some strange reason, one of the programs I always run (Directory
Opus) doesn't remember where it was if I reboot.
>

It may be that it's just not designed to bother. But if it's old
it could also be a permission issue. For example, if it's
trying to save a config file in the program folder, or trying
to write to HKLM, that might happen. I suppose you could
try procmon (or regmon/filemon) to see whether it's failing
such an operation. Then you could change permissions for just
that file or reg key.

[Mayayana]

"R.Wieser" <add...@not.available> wrote

| > Firefox, for example, seems to check open instances and display
| > multiple windows in cascade fashion.
|
| Try using 0x8000, 0x8000 for the windows/dialogs initial position.
|

I don't understand. Are you saying FF has a setting to spec
the opening position? Is the heat getting to you, Rudy? :)

[R.Wieser]

Mayayana,

> I don't understand. Are you saying FF has a setting to spec
> the opening position?

Nope. I'm telling you /the OS/ gouverns the cascading of windows - if you
let it and not force them to a specific location.

Just try it.

> Is the heat getting to you, Rudy? :)

Not yet, but it is starting to bother me. :-|

Regards,
Rudy Wieser

[Mayayana]

"R.Wieser" <add...@not.available> wrote

| Nope. I'm telling you /the OS/ gouverns the cascading of windows - if you
| let it and not force them to a specific location.
|
| Just try it.
|

I still don't know where I'd spec H8000. But I think
I know what you mean. If I open one window and set it, then
subsequent windows would automatically cascade? I don't
think I've ever written anytyhing with multiple instances before,
so I've never thought about how to handle that.

So FF is setting the first window and then leaving it to
Windows if there's already one showing? I just tried
Notepad. It opens to its last position but then every
subsequent instance opens at the same spot, on top of
each other.

[R.Wieser]

Mayayana,

> I still don't know where I'd spec H8000.

After posting I realized by explanation-fu left something to be desired :-\

When I create a dialog in a resourcefile I have something like this :

- - - - - - - - - - - - - - - - - -
#define IDD_BASEWIN 1000

IDD_BASEWIN DIALOG DISCARDABLE 300, 0, 200, 36
begin
....
end
- - - - - - - - - - - - - - - - - -

The only thing you need to do is to replace the, in this case, "300, 0" with
"0x8000, 0x8000"

To be honest, I have not tried to do that with CreateWindow (seldom use it
anymore), but as far as I understood it works the same way there.

> But I think I know what you mean. If I open one window and
> set it, then subsequent windows would automatically cascade?

Not quite. It works for any program, and they do not even need to be the
same. IOW, it works for any child of the desktop. You can take it as the
OS'es effort to keep all (newly) placed windows visible (not one hidden
behind another).

> So FF is setting the first window and then leaving it to
> Windows if there's already one showing?

Its not about FF. Its about dialogs/windows in general. As you said
yourself, its ultimatily the window which decides where it gets placed.

I'm not sure if its fully comparable, but a quick test just now opening a
number of "file explorer" dialogs shows that the first one got create at its
old place, but the subsequent ones started to form in a cascading wise from
the top-left of the screen.

The "first one at its old place and the subsequent ones cascading from the
top-left" *might* have something to do with

HKEY_CURRENT_USER\Software\Microsoft\Windows\ShellNoRoam\Bags

(I was looking for an answer to the OPs question).

> I just tried Notepad. It opens to its last position but then
> every subsequent instance opens at the same spot, on top of
> each other.

Same here. It probably uses (the equivalent of) DS_CENTER as one of its
styles.

Regards,
Rudy Wieser

[Mayayana]

"R.Wieser" <add...@not.available> wrote

| When I create a dialog in a resourcefile I have something like this :
|
| - - - - - - - - - - - - - - - - - -
| #define IDD_BASEWIN 1000
|
| IDD_BASEWIN DIALOG DISCARDABLE 300, 0, 200, 36

I see. VC++? I've only dabbled in that and never with GUIs.
I use VB6, generally. That does all the grunt GUI work behind the
scenes. When I load the window, before making it visible, I
look up where it should be and position it.

I use CreateWindowEx, but only for self-subclassing
userControls.

I'm going to miss VB6. I'd become fairly expert over the years,
easily creating self-subclassing control windows for maximum
speed and flexibility. At the same time it's easy to code. And
the software runs on virtually any existing Windows box with
no support files needed. Real Win32 software, not bloated
"apps" pasted together with Python, QT, .Net, Java, RT, or
worse.
But I don't code much now. I just don't need things that I
don't have. And it won't be long, I'm sure, before non-corporate
non-spyware is considered to be "illegal sideloading" and Windows
sends out a SWAT team to take down my EXE.... Strange to look
back and realize the age of popular computing lasted a mere
30-odd years.

[DanS]

Ken Blake <K...@invalid.news.com> wrote in
news:ctdn8hp1hh8sjar1m...@4ax.com:

> On Mon, 23 May 2022 11:27:18 +0100, "J. P. Gilliver (John)"
> <G6...@255soft.uk> wrote:
>
>>(I'm on 7, but the problem applies to XP and 10 too, and
>>the solution - if there is one! - may be universal.)
>>
>>Occasionally, I accidentally either move a window, or drag
>>one of its sides, by mistake - while doing (or intending
>>to) something else with the mouse pointer, it catches on an
>>edge, or more likely a title bar. (Unlike a lot of people,
>>I _don't_ operate most of my windows full-screen most of
>>the time,
>
>
> Same for me. I almost never run anything full screen.

I don't think that many people run everything 'full screen' (more likely meaning
'maximized').

[Char Jackson]

When I worked in a cube farm at a large corporation, as I walked around
the place I noticed that virtually everyone ran their stuff maximized,
with the only exception being one woman who spent much of her time
resizing windows and moving them around on her screen. Admittedly, that
was just over a decade ago, so things may have changed now that bigger
screens and/or higher resolutions are available.

I still run almost everything maximized, with just a few exceptions. I
don't like unnecessary vertical/horizontal scrolling.

[R.Wieser]

Mayayana,

> | When I create a dialog in a resourcefile I have something like this :
> |
> | - - - - - - - - - - - - - - - - - -
> | #define IDD_BASEWIN 1000
> |
> | IDD_BASEWIN DIALOG DISCARDABLE 300, 0, 200, 36
>
> I see. VC++? I've only dabbled in that and never with GUIs.

Nope, hand-crafted and to be compiled into a .RES file using brc32.exe
(Borland Resource Compiler). I'm the guy writing Windows programs in
Assembly, remember.

> I use VB6, generally

I have used VB5 myself (for work), and as far as I can remember you could
specify the starting coordinates of the dialog if you wanted. If you want
to use the "auto placement" of Windows it would not hurt to take a look.

It could even mean you would not need to add extra code in the create event
of the window to move it to its desired-by-you position anymore.

> Real Win32 software, not bloated "apps" pasted together with
> Python, QT, .Net, Java, RT, or worse.

:-) As an assembly programmer any language above it feels needlesly
bloated. Including good-old C and it derivatives.

> But I don't code much now. I just don't need things that I
> don't have.

To me it was and has again become a hobby. Instead of figuring sudokos or
jigsaw puzzles I try to "puzzle" programs together.

> And it won't be long, I'm sure, before non-corporate
> non-spyware is considered to be "illegal sideloading" and
> Windows sends out a SWAT team to take down my EXE....

:-) Thats pretty-much why I don't yearn for any of those blasted
"smartphones". Any OS which I am not the lord-and-master over isn't welcome
in my house. Its also is the reason why I still have XP as my major OS.

> Strange to look back and realize the age of popular computing
> lasted a mere 30-odd years.

Same for hobby electronics. For a number of years I designed circuits,
turned them into layouts and etched PCBs for them, drilled the holes and
soldered the components in. Nowerdays the miniaturisation of the components
doesn't allow for that anymore. A gone era.

Regards,
Rudy Wieser

[Mayayana]

"R.Wieser" <add...@not.available> wrote

| Nope, hand-crafted and to be compiled into a .RES file using brc32.exe
| (Borland Resource Compiler). I'm the guy writing Windows programs in
| Assembly, remember.
|

Interesting. I didn't know that. I decided to learn assemby
at one point, but didn't get far. I'm not really a math person.
I'm sort of a design-builder, an expert at straddling science
and art, but not a master of either. So I write code with a
practical sense. Once it works I like to optimize, but what I
like about VB is that I can just skip all the grunt work and
focus on the main functionality.

I dabbled a bit in C++ and could see the appeal. Compared
to VB it's like riding without a seat belt or a floor. But I just
wasn't ambitious enough to master it. And it's a damned ugly
language.

| > Strange to look back and realize the age of popular computing
| > lasted a mere 30-odd years.
|
| Same for hobby electronics. For a number of years I designed circuits,
| turned them into layouts and etched PCBs for them, drilled the holes and
| soldered the components in. Nowerdays the miniaturisation of the
components
| doesn't allow for that anymore. A gone era.
|

Ah. An oldtimer. I was never around for the early days. I only
discovered computers with Win98, when it became a car to drive
and not just an engineering challenge for greasemonkeys. I think
that crippled my understanding, partly. I'm an auto-didact in
GUI programming. So understanding what really happens underneath
came later and was hard for me to grasp. It's still not second nature
for me, the way it is with fully trained C++ programmers.

[Mayayana]

"DanS" <t.h.i.s....@r.o.a.d.r.u.n.n.e.r.c.o.m> wrote

| I don't think that many people run everything 'full screen' (more likely
meaning
| 'maximized').
|

I certainly don't. I have a 27" monitor. Fullscreen would
have me turning my head back and forth to read. And it
would negate the advantage of being able to switch around
by clicking title bars.

But so many people today use phones as a primary device.
One app at a time. And many use laptops with small screens,
even if they never move it. Very few people consider ergonomic
design. I imagine that someone on a 15" laptop, using almost
exclusively MS Word, would have no reason not to have a
maximized window. They've crippled their functionality back
to 1998, but they don't notice.

[J. P. Gilliver (John)]

On Mon, 23 May 2022 at 09:36:19, Ken Blake <K...@invalid.news.com> wrote
(my responses usually FOLLOW):

>On Mon, 23 May 2022 11:27:18 +0100, "J. P. Gilliver (John)"
><G6...@255soft.uk> wrote:
>
>>(I'm on 7, but the problem applies to XP and 10 too, and the solution -
>>if there is one! - may be universal.)
>>
>>Occasionally, I accidentally either move a window, or drag one of its
>>sides, by mistake - while doing (or intending to) something else with
>>the mouse pointer, it catches on an edge, or more likely a title bar.
>>(Unlike a lot of people, I _don't_ operate most of my windows
>>full-screen most of the time,
>
>
>Same for me. I almost never run anything full screen.
>

Nice to know there are others like me! A lot of both people in general,
and programmers (especially web page designers), are (or act*) very
surprised when you explain this, and treat you/us as if you're being
awkward. (* I think in the case of programmers/designers, they just
_act_ that way because it's more work for them catering to us.)

>
>>so their relative sizes and positions are
>>important to me.)

[]

>Can't you just drag it back where it was? Or is that not precise
>enough for you? That's what I do.

Oh, usually it would be; however, if the move/resize is accidental - i.
e I was trying to do something else with the mouse - I'm not necessarily
sure, even roughly, what the original position/size _was_. (-:

--
J. P. Gilliver. UMRA: 1960/<1985 MB++G()AL-IS-Ch++(p)Ar@T+H+Sh0!:`)DNAf

in the kingdom of the bland, the one idea is king. - Rory Bremner (on
politics), RT 2015/1/31-2/6

[Frank Slootweg]

R.Wieser <add...@not.available> wrote:
[...]

> :-) As an assembly programmer any language above it feels needlesly
> bloated. Including good-old C and it derivatives.

Assembly is needlessly bloated! Real Men Microcode!

And using assembly just means you're too dumb to remember the opcodes.

:-) (for the HI)

[J. P. Gilliver (John)]

On Tue, 24 May 2022 at 17:31:11, Frank Slootweg <th...@ddress.is.invalid>

wrote (my responses usually FOLLOW):

>R.Wieser <add...@not.available> wrote:
>[...]
>
>> :-) As an assembly programmer any language above it feels needlesly
>> bloated. Including good-old C and it derivatives.
>
> Assembly is needlessly bloated! Real Men Microcode!

Entering it with switches, of course. And hoping an insect - or to use
the US term, bug - doesn't end up in the relays, sorry, software ...

>
> And using assembly just means you're too dumb to remember the opcodes.

(-: (I can still remember the opcodes for the first computer I learnt on
- all 8 of them. Z [000], A, S, T, I, J, K, and E [111].) And yes, you
really _did_ program BRENDA with switches, though I don't think it had
relays, in the logic anyway.
>
>:-) (for the HI)

--
J. P. Gilliver. UMRA: 1960/<1985 MB++G()AL-IS-Ch++(p)Ar@T+H+Sh0!:`)DNAf

What's really worth knowing is for the most part unlearnable until you have
enough experience to even recognise it as knowledge, let alone as useful
knowledge. - Wolf K <wol...@sympatico.ca>, in alt.windows7.general, 2017-4-30

[R.Wieser]

Mayayana,

> I'm not really a math person.

Neither am I. Luckily it hasn't stopped me.

> but what I like about VB is that I can just skip all the
> grunt work and focus on the main functionality.

And there I am, thinking that all those (system) DLLs are doing the grunt
work for me. I only have to stich a few functions outof them together and
magic happens. :-)

> I dabbled a bit in C++ and could see the appeal. Compared
> to VB it's like riding without a seat belt or a floor.

I can only imagine that doing Assembly must have been awfull for you, not
even knowing which side was up ...

Using higher programming languages I was always troubled by the distance
between what I wrote and what it actually got compiled into. It didn't
help to read that the compiler could sometimes just "optimize" whole swats
of code away, making it not do what the programmer intended.

And the almost always needed typecasting, because everyone can define their
own type names. Very funny when you have to deal with a couple of
DLLs/externam modules that have been written by different people ...

> Ah. An oldtimer.

You could say that I guess.

> I was never around for the early days. I only discovered computers
> with Win98,

My own first contact with computers was a 8080 processor based (later
upgraded to a Z80 processor) non-name home-build by the leader of the
electronics club I had become a member of (learned my transistors there).
After a number of years and a series of other 8-bitters (TRS80, Apple 2e,
C64 to name a few) my first own PC was a 286 running DOS 3.3 with a *BIG*
harddisk. A whopping 40 Megabyte of it. :-)

> when it became a car to drive and not just an engineering
> challenge for greasemonkeys.

I'm not much of an engineer, but I'm afraid that I am one of those
"greasemonkeys". I have /way/ to much fun in having the 'puter do what /I/
want instead of the other way around.

> I'm an auto-didact in GUI programming.

Although I had some helpfull people around in my early days on that
8080-based 'puter most of the later stuff learned with the help of library
books and later the internet. So, also auto-didact. Certainly whole
Windows part.

> So understanding what really happens underneath came later and was
> hard for me to grasp. It's still not second nature for me, the way
> it is with fully trained C++ programmers.

I've disassembled enough programs make me wonder how many of those "fully
trained C++ programmers" who know what is going on beneath their feet ...

Than again, part of the ugly and rather clumsy code I sometimes see could be
the result of a bad compiler. Who knows.

Regards,
Rudy Wieser

[R.Wieser]

Frank,

> Assembly is needlessly bloated! Real Men Microcode!

And when they need to change it they do it on the chip, using a fine-pointed
needle to flip the bits !

> And using assembly just means you're too dumb to remember the opcodes.

Ackkk... You got me there I'm afraid. Never been able to remember just a
meager few of them .. :-p

Regards,
Rudy Wieser

[nospam]

In article <t6j82u$uno$1...@gioia.aioe.org>, R.Wieser

<add...@not.available> wrote:

> Using higher programming languages I was always troubled by the distance
> between what I wrote and what it actually got compiled into. It didn't
> help to read that the compiler could sometimes just "optimize" whole swats
> of code away, making it not do what the programmer intended.

optimizing means make it faster. it does *not* mean change what it does.

> And the almost always needed typecasting, because everyone can define their
> own type names. Very funny when you have to deal with a couple of
> DLLs/externam modules that have been written by different people ...

typecasting is a very good thing. it's much easier to find errors at
compile time rather than at run time, surprising the user, which is
never a good thing.

> > when it became a car to drive and not just an engineering
> > challenge for greasemonkeys.
>
> I'm not much of an engineer, but I'm afraid that I am one of those
> "greasemonkeys". I have /way/ to much fun in having the 'puter do what /I/
> want instead of the other way around.

who intentionally writes code that does something they don't want?

> > So understanding what really happens underneath came later and was
> > hard for me to grasp. It's still not second nature for me, the way
> > it is with fully trained C++ programmers.
>
> I've disassembled enough programs make me wonder how many of those "fully
> trained C++ programmers" who know what is going on beneath their feet ...

they know far more about what's going on than assembly programmers, who
can't see the big picture.

put another way, they deal in concepts, not micromanaging bits.

> Than again, part of the ugly and rather clumsy code I sometimes see could be
> the result of a bad compiler. Who knows.

not you.

[DanS]

"Mayayana" <maya...@invalid.nospam> wrote in
news:t6j08a$c6h$1...@dont-email.me:

You know, after I posted my comment, I started to think about it a little more about it...

The first thing that came to mind when I read the OP, was Facebook. If you run
Facebook maximized, on say a 27" monitor, there are HUGE, huge areas of wasted
white space on both sides of the screen.

But the more I thought about it, I do realize I do run a lot of stuff maximized, but only
programs where being maximized improves the s/w use.

I do CAD work, always maximized. The more you see, the better. Spreadsheets too,
sometimes, as the boss is really hip on "just add another column!" after there's already
50 columns in his spreadsheets. Using VS is often maximized. I can pull several code
windows of tabs into separate windows to reference back and forth between them.

Audio and video editing is always maximized...the more you can see at once, the
better.

I'm writing this post right now, from work, as I'm on break and RDP'd into my home
machine, in fullscreen.

On a side/related note...

When Windows 11 came out, and listed one of the new 'features' is that you set the
start button to be in the middle of the tasbar, instead of all the way to the left bottom
corner. I balked at that, likw wow!!!! that's a feature, you need to tout?!?!...

Fast-forward in time, and I'm in front of my studio PC with Pro Tools maximized, the
the 34" ultra-wide I'm using for that, and need to get tot he Start Menu...

...it was at that point, I realized that perhaps that is a cool feature, because I found I had
to move my head diagonally quite a bit to focus almost two feet away from what I was
looking at to 'get to' the Start button way the hell down over there!

I used to be like you, Mayayana...I hated anything new that came out, especially
Windows, as each version would get more bloated with all this junk I'll never need or
use, and be super-cncerned about *everything*...I mean, I even wrote my own
Windows shell to replace Explorer as a shell, so it would work l;ike I wanted it to, and
be efficient like I thought it should be.

...at some point, you start going through so much work to not do/have anything new,
updated, or upgraded, that it just not worth a the hassle. What is gained, is not more
than the effort but forth to do so. (For me anyway.)

I mean, if you've got an older PC being used, that's fine, but to like buy new hardware in
202x and want to use XP on it??

Don't get me wrong...I do do some trimming, and customization, just not anywhere
near to the degree I used to.

[R.Wieser]

nospam,

> optimizing means make it faster.

No, it doesn't. Not by a long shot. Though mostly the /effect/ of those
optimalisations is a faster running program.

> it does *not* mean change what it does.

Do a wild guess why I put the word between quotes ...

> typecasting is a very good thing.

In its intention ? Sure. But not when the programmer gets warnings about
them every second variable he's trying to use. Which /will/ happen when
trying to use third-party stuff - for the reason I already described.

>> I'm not much of an engineer, but I'm afraid that I am one of those
>> "greasemonkeys". I have /way/ to much fun in having the 'puter do what
>> /I/
>> want instead of the other way around.
>
> who intentionally writes code that does something they don't want?

Lol. I suggest you re-read what I said. Focus on the last line you quoted.

If that doesn't help, think about you and a car salesman. What would you
like better : the car salesman taking you for a ride, or the other way
around.

> they know far more about what's going on than assembly programmers,
> who can't see the big picture.

If you say so.

> put another way, they deal in concepts, not micromanaging bits.

Yep. And by that feat they lose sight of the cost of what they write - and
depend on their compilers to optimize whatever they come up with into the
most optimum form possible.

> not you.

Yep, that is what I said. You're not a real brainiac, are you ?

Regards,
Rudy Wieser

[Mayayana]

"DanS" <t.h.i.s....@r.o.a.d.r.u.n.n.e.r.c.o.m> wrote

| But the more I thought about it, I do realize I do run a lot of stuff
maximized, but only
| programs where being maximized improves the s/w use.
|

Yes. I do that for graphic editing, VS, my own coding editor for VBS
and HTML, and my occasional Libre Office Writer usage. But for most
things I'm using windows of maybe 1/2 screen width.

|
| I used to be like you, Mayayana...I hated anything new that came out,
especially
| Windows, as each version would get more bloated with all this junk I'll
never need or
| use, and be super-cncerned about *everything*...I mean, I even wrote my
own
| Windows shell to replace Explorer as a shell, so it would work l;ike I
wanted it to, and
| be efficient like I thought it should be.
|
| ...at some point, you start going through so much work to not do/have
anything new,
| updated, or upgraded, that it just not worth a the hassle. What is gained,
is not more
| than the effort but forth to do so. (For me anyway.)
|
| I mean, if you've got an older PC being used, that's fine, but to like buy
new hardware in
| 202x and want to use XP on it??
|

I think I feel the same way. But the main thing for me is control.
They want to put the Start menu in the middle? OK. But can I move
it if I want to? Each version has made it harder to change things
to the way I find convenient. Ribbon menus? OK. But why can't
they also be switched back? I want to be able to cruise at top
speed, without thinking about it. I don't want to be zipping along
and suddenly get a big window asking whether I want to copy a
file, replace it, or add with a different name. Huh? Shut up! All I
need is a simple message: Overwrite existing file?

With Win7 I spent weeks figuring out how to make it civilized.
I even wrote my own program to remove file restrictions in bulk.
I find it reasonably usable but not optimized. With Win10 I just
avoid using it. It's no
fun. Can I stop it calling home? Not sure. It changes services
settings without asking. It's crazy bloated. Much of it can be
customized, but not like it used to be. And now I just use the
popular tweak programs rather than writing my own. It's not
worth the trouble. So, yeah, it's OK. But with Win10 I'm kind of
drawing a line. It's spyware. I probably can't undo that. So if
I ever have to give up XP I'll probably go to 7 or Linux. If I
have to give up 7 then I'll probably just hold my nose and switch
to Linux. Win10 will never be *my* workshop or kitchen. It's
Microsoft's.

At this point I have a Win10 laptop that I use mostly just
for online retail research where I need a new browser and to enable
javascript. It's a system that lives in the gutter. Win10 is a
lying whore. (No offense to prostitutes.) I wouldn't
put private data on it. I wouldn't run productivity software on it.
I didn't even bother to get a real Win10 computer. Just a laptop
that I keep in the closet. I take it out when I'm shopping for
something and want to be able to go to crappy sites like Home
Depot, department stores, and so on without the webpage being
all messed up. Many of those sites now require script to see their
images and are loading webp files. My browsers on XP can't even
display a webp file. Website compatibility coding is out the window.
Special effects is all they care about.

For me, a computer is a very practical issue. Like my kitchen or
workshop, I want the tools I use in the places I expect them to be.
I want things clean and simple. I want no spyware. If I'm going
to do things more than once I might want to write a script for
that task. So I've ended up with ultra-efficency that would require
a great deal of work on another system.

I can get used to changes, but not inflexibility. I don't care if Bill
Gates thinks my frying pan should be on the left and saucepans on
the right. I don't care if he thinks my hammers should be where my
screwdrivers are and vice versa. I didn't ask him. Nearly all these
problems are not about change but about removing choice.

[Mayayana]

"J. P. Gilliver (John)" <G6...@255soft.uk> wrote

| >Same for me. I almost never run anything full screen.
| >
| Nice to know there are others like me! A lot of both people in general,
| and programmers (especially web page designers), are (or act*) very
| surprised when you explain this

You think? I find that most webpages these days are stuck at
1000px wide. They should self-size, but they don't. So there's
no point using the whole screen. But I think they're also writing
more for cellphones, and they don't actually know what they're
ccoding. It's all automated.

[R.Wieser]

... Never been able to remember *more than* just a meager few of them.

[nospam]

In article <t6jegk$r24$1...@dont-email.me>, Mayayana

<maya...@invalid.nospam> wrote:

> I find that most webpages these days are stuck at
> 1000px wide. They should self-size, but they don't.

i dunno what pages you visit, but nearly all of them reflow with
changes in the viewport, even down to mobile sized.

> So there's
> no point using the whole screen.

there can be for smaller screens (almost always on mobile), but usually
not for larger displays.

> But I think they're also writing
> more for cellphones, and they don't actually know what they're
> ccoding. It's all automated.

mobile is a *significant* amount of traffic, so they definitely are
including that in their design, and for it to work properly, they
*must* know what they're doing.

[wasbit]

"nospam" <nos...@nospam.invalid> wrote in message
news:240520221632044747%nos...@nospam.invalid...

> In article <t6jegk$r24$1...@dont-email.me>, Mayayana
> <maya...@invalid.nospam> wrote:
>
>> I find that most webpages these days are stuck at
>> 1000px wide. They should self-size, but they don't.
>
> i dunno what pages you visit, but nearly all of them reflow with
> changes in the viewport, even down to mobile sized.
>

> Snip <

FYI
There are quite a number of web pages that don't change to reflect the size
of the viewing area.
There are even some that you can't read what is outside the viewing area
because there are no scroll bars.

--
Regards
wasbit

[Paul]

int main () {

int i; /* Defined at the top of Main */

...

int j; /* Define a variable with the intention of local scope */
for( i = 0; i < 20000000; i++ ){
j=j+1;
}
...

return 0;
}

If compiled with optimization turned off, the loop wastes a few milliseconds.

If we use -O3, the compiler uses reasoning, to note that "j" is
not referenced, after the loop. First it throws away j. Then
next, it re-does the analysis and notes that the loop has
no body code. The compiler reasons the loop can be
replaced with

i=19999999
(j removed from scope)

A few more bits of analysis, and the program exits instantly
with the error status hardwired to return 0.

If we make a small change...

for( i = 0; i < 20000000; i++ ){
j=j+1;
}
printf("The loop executed %d times\n", j);

then the compiler at -O3 cannot remove j.
After the loop runs, a static analysis shows

i=19999999
j=20000000
printf("The loop executed 20000000 times\n"); /* printf is invariant */

The loop has been removed. The program has very short runtime.

In the old days, we used to have compute contests with piss-weak
computing devices, where incrementing a counter was the fastest
thing they did. Imagine how annoyed we'd be, if our ability
to craft a "time wasting loop" was thwarted. Well, that's
a constant issue when dealing with life at -O3.

When I do hobby programming here, I never use -O3.
The compiler people are too clever, by a mile.

A professional programmer could likely give you
a lot more examples of where the compiler people are
just... over-reaching. I've forgotten some of the ones
I've read about. They apply various pragmas to prevent
the overreach at critical points.

And this is a whole design paradigm. At work, some people
argued for the languages we used for design (like chip design),
to be "power user". "Design me a parallel CRC generator
with 32 bit busses on input and output." But the people
in charge, saw the danger in this. They wanted no part of
languages which worked that way. Instead, the code is
"design exactly what I tell you". If you put in j=j+x ,
you would expect an ALU in adder mode, having bit width
suited to the width of j and the width of x. The user
was responsible for making sure the resulting hardware
function, was the function that was desired. Depending
on the cell library, you might remove the Carry-Out on
the ALU and reduce the wasted gates in it.

The end result was, tools which required a lot of lines
of input. But the advantage was, the developer writing
the code, could immediately see at his/her own desk, the
impact of stating constructs a certain way. And not having
to consult the "tool jockey" who made up an intelligent
parser instead. So instead of wasting hours going
"now, why did it make the CRC generator this particular way",
you could while away your hours making the thing work
exactly the way you thought it should be done.

At -O3, you can easily have all the code vacuumed out
of your examples.

Paul

[Philip Herlihy]

In article <240520221632044747%nos...@nospam.invalid>, nospam wrote...

Worth noting that there's a whole art of building "responsive" webpages, where
text, images and layout are optimised for the detected width of the viewport.
In particular, the CSS styling notation for the web has a feature called "media
queries", where layout can be quite drastically changed below a specified width
in pixels. A table, for example, can be re-interpreted as a list with fewer
columns. More on this here, if you're interested in a gentle introduction:
https://www.w3schools.com/css/css3_mediaqueries_ex.asp

If you're not familiar with CSS, or "Cascading Style Sheets", the approach for
some time has been to separate "content" HTML tags (like Image, Header,
Paragraph) from "styling" tags (CSS) like Bold, color, margin, position. This
is often done in two separate but linked files read together by the browser.
So the identical HTML can be re-imagined quite drastically by changing the
style-sheets. For a dramatic demonstration of this, see:
http://www.csszengarden.com/
(Tip - be sure to click the design name, not the designer name, to see the
various renderings.)

I've found CSS easy enough to learn, but very, very hard to master.

--

Phil, London

[Mayayana]

"wasbit" <wasbit...@hotmail.com> wrote

| FYI
| There are quite a number of web pages that don't change to reflect the
size
| of the viewing area.
| There are even some that you can't read what is outside the viewing area
| because there are no scroll bars.
|

Nospam just likes to argue. He'll tell you the sky isn't
blue. If you prove it is he'll say, "Not when it's cloudy!"
Some days he's worse than others. I can't tell whether
nospam is a real person with well worn synapses or whether
it may perhaps be an ongoing research project by those
nuts out at Stanford trying to make convincing chat bots.

[Paul]

When you did these things, the machines were simple enough you
could use a fold out programming card.

https://pbs.twimg.com/media/EzhObIaWEAA9znx?format=jpg&name=small

According to the 2020-11-01 entry here, the Intel instruction
set is now up to 2000 instructions. My guess is, HLL still use
300 of the instructions, meaning the other 1700 are never used.

(Agner Fox is a proponent of "sanity in instruction sets")

https://www.agner.org/optimize/blog/read.php?i=25

And some of those extensions, for the most part they
are hand coded. Like the Prime95 guy might use them.
He's the person that detected that some AVX option
wasn't working so well, when he introduced it into
some hand-rolled assembler. When he released his modified
version, some end users noted problems.

Kens processor likely has the AVX512 pinned off now, by
a microcode update. Because (apparently), Intel did not
intend for it to be enabled on the 12th gen. And processors
produced from this date, no longer have it factory enabled
(so changing microcode patch files won't change the behavior
on them).

I don't think anybody has even read all 2000 instructions,
let alone analyzed them for "useful programming patterns".

Paul

[nospam]

In article <t6jdrp$1hn7$1...@gioia.aioe.org>, R.Wieser

<add...@not.available> wrote:

>
> > optimizing means make it faster.
>
> No, it doesn't. Not by a long shot.

it doesn't?

> Though mostly the /effect/ of those
> optimalisations is a faster running program.

but it does?

you just said it doesn't.

you're contradicting yourself.

> > it does *not* mean change what it does.
>
> Do a wild guess why I put the word between quotes ...

there's no need for any guesses when it's obvious that you have no idea
what you're talking about.

> > typecasting is a very good thing.
>
> In its intention ? Sure.

it fulfills that intention, in several different ways.

> But not when the programmer gets warnings about
> them every second variable he's trying to use. Which /will/ happen when
> trying to use third-party stuff - for the reason I already described.

that's a lot of words to say you haven't used 'third party stuff' much,
if you have at all, with a deliberately vague description because you
do not understand anything about it.

> >> I'm not much of an engineer, but I'm afraid that I am one of those
> >> "greasemonkeys". I have /way/ to much fun in having the 'puter do what
> >> /I/ want instead of the other way around.
> >
> > who intentionally writes code that does something they don't want?
>
> Lol. I suggest you re-read what I said. Focus on the last line you quoted.
>
> If that doesn't help, think about you and a car salesman. What would you
> like better : the car salesman taking you for a ride, or the other way
> around.

i'd like it better (as would others) if you didn't pretend to know more
than you actually do and maybe even try to learn something, but that's
wishful thinking.

> > they know far more about what's going on than assembly programmers,
> > who can't see the big picture.
>
> If you say so.

i do, as does anyone who does even casual programming, let alone
professionally.

> > put another way, they deal in concepts, not micromanaging bits.
>
> Yep. And by that feat they lose sight of the cost of what they write

no they very definitely don't.

in fact, it's the opposite, they have a much better idea of what's
going on.

> - and
> depend on their compilers to optimize whatever they come up with into the
> most optimum form possible.

compilers produce *much* better code than humans can in almost every
case, particularly with modern processors.

the exceptions are very rare and in very specific and well understood
scenarios (just not by you).

you're also ignoring the huge cost of writing in assembly, which is
*very* substantial, along with the limitation of it being non-portable.

[nospam]

In article <t6kq46$gg$1...@dont-email.me>, wasbit

<wasbit...@hotmail.com> wrote:

> There are quite a number of web pages that don't change to reflect the size
> of the viewing area.

some web designers are incompetent and do not test their work (or if
they do, they ignore that it doesn't work). however, that is the
exception, not the rule.

> There are even some that you can't read what is outside the viewing area
> because there are no scroll bars.

i've never seen that. do you have an example?

[DanS]

"J. P. Gilliver (John)" <G6...@255soft.uk> wrote in
news:eNo30kyGG2iiFwAq@a.a:

[Top-posted for brevity]

I don't quite understand...

Why can't you move it back, or "un"- resize it ?

> (I'm on 7, but the problem applies to XP and 10 too, and
> the solution - if there is one! - may be universal.)
>
> Occasionally, I accidentally either move a window, or drag
> one of its sides, by mistake - while doing (or intending
> to) something else with the mouse pointer, it catches on an
> edge, or more likely a title bar. (Unlike a lot of people,
> I _don't_ operate most of my windows full-screen most of

> the time, so their relative sizes and positions are
> important to me.)
>
> I discovered fairly recently that _if I notice while the
> movement is in process_ (I just see the _outline_ move, i.
> e. I've still got the "mouse button pressed" [actually I
> use a touchpad, but similar applies]), I can cancel the
> move/resize by - keeping the button pressed and - pressing
> the Esc key. But if I don't - anyone know of a way of
> undoing a window move/resize? (Ctrl-Z doesn't seem to
> work.)

[Zaidy036]

On 5/23/2022 6:27 AM, J. P. Gilliver (John) wrote:
> (I'm on 7, but the problem applies to XP and 10 too, and the solution -
> if there is one! - may be universal.)
>
> Occasionally, I accidentally either move a window, or drag one of its
> sides, by mistake - while doing (or intending to) something else with
> the mouse pointer, it catches on an edge, or more likely a title bar.
> (Unlike a lot of people, I _don't_ operate most of my windows
> full-screen most of the time, so their relative sizes and positions are
> important to me.)
>
> I discovered fairly recently that _if I notice while the movement is in
> process_ (I just see the _outline_ move, i. e. I've still got the "mouse
> button pressed" [actually I use a touchpad, but similar applies]), I can
> cancel the move/resize by - keeping the button pressed and - pressing
> the Esc key. But if I don't - anyone know of a way of undoing a window
> move/resize? (Ctrl-Z doesn't seem to work.)

make a batch with a shortcut key to it for each window you are concerned
about. In the batch only need NirCmd to set size and position
<https://nircmd.nirsoft.net/win.html>

[R.Wieser]

Paul,

> When you did these things, the machines were simple enough you
> could use a fold out programming card.

Never had one of those, but yes, the printed-out ones where nice and short.
That already started to change with the 286 I started with - which also
started to become less logically ordered (with the new commands placed
here-and-there)

... aaaand I see you ninja-ed me with that next link. :-)

> I don't think anybody has even read all 2000 instructions,
> let alone analyzed them for "useful programming patterns".

I certainly haven't. Double for all the added special groups. Heck, I only
started to look at the FPU instructions when I tried to create some 3D
scenes using OpenGL.

Regards,
Rudy Wieser

[R.Wieser]

nospam,

>> > optimizing means make it faster.
>>
>> No, it doesn't. Not by a long shot.
>
> it doesn't?
>
>> Though mostly the /effect/ of those
>> optimalisations is a faster running program.
>
> but it does?
>
> you just said it doesn't.

Look up the word "means". I'm pretty sure that it doesn't mean what you
think it does.

Kiddo, you have a serious comprehension problem - or that is just the way
you play your game.

Regards,
Rudy Wieser

[J. P. Gilliver (John)]

On Tue, 24 May 2022 at 12:09:05, Mayayana <maya...@invalid.nospam>

wrote (my responses usually FOLLOW):

>"R.Wieser" <add...@not.available> wrote
>
>| Nope, hand-crafted and to be compiled into a .RES file using brc32.exe
>| (Borland Resource Compiler). I'm the guy writing Windows programs in
>| Assembly, remember.
>|
>
> Interesting. I didn't know that. I decided to learn assemby
>at one point, but didn't get far. I'm not really a math person.

Assembly isn't all to do with math(s) - in fact I don't think I ever did
any that was, but then I've not done assembly at all for years. (And
very little on x86 machines - more on earlier processors. [Or in work,
ones - bit-slice - we'd designed ourselves, including the interpreter.])

>I'm sort of a design-builder, an expert at straddling science
>and art, but not a master of either. So I write code with a
>practical sense. Once it works I like to optimize, but what I
>like about VB is that I can just skip all the grunt work and
>focus on the main functionality.

VB - or any other high-level language - isn't going to be that
efficient, even if you do optimisation by rewriting it. As (I think it
was) Paul pointed out, a good compiler (or whatever the right word is
for what processes VB code) can do amazing things, but it's likely never
to approach real hand-coded code. However, it's a darn sight easier
(especially not to miss something that will break something); I suspect
if I ever do any coding again, it won't be in assembler.

>
> I dabbled a bit in C++ and could see the appeal. Compared
>to VB it's like riding without a seat belt or a floor. But I just
>wasn't ambitious enough to master it. And it's a damned ugly
>language.

(I never got into the plusses. I did learn raw C. Yes, there's a
definite appeal to it. [And definite cliquiness too, which I've been
guilty of too, though I hope less than some.])

>
>| > Strange to look back and realize the age of popular computing
>| > lasted a mere 30-odd years.
>|
>| Same for hobby electronics. For a number of years I designed circuits,
>| turned them into layouts and etched PCBs for them, drilled the holes and
>| soldered the components in. Nowerdays the miniaturisation of the
>components
>| doesn't allow for that anymore. A gone era.
>|

When I got into electronics - late 70s, I think (and in UK) - it was
still possible to save money by building your own - if you used modules.
By then, it wasn't really money-saving to put the discrete components
together yourself, so the only people who did were those after
ultra-good hifi, or other things not available by using pre-built
modules. Not too long after that, it became cheaper to buy most things
ready-made, which killed the hobby really, apart from as I said the
hi-fi enthusiasts and other specialised requirements.

With PCs, when I came in - probably 1980s, I can't remember - you could
definitely save money by selecting and assembling the various component
modules yourself, and also get something very much more what you wanted;
I think the economic side faded out about the start of XP (not least
because the cost of Windows shot up about then). Over the years after
that, the performance of the various bits improved to the extent that,
although some people do still assemble their own for the ultimate in
either performance or some other parameter (including appearance!),
they're very much in the minority: if I go into a branch of PC World
here now, I'm not sure if I can even find things like memory,
processors, graphics cards - only ready-built PCs.

(Sorry, I've deviated from what you were talking about which was
software coding rather than hardware, but I think similar applies: what
more and more of us are using if we do any "coding" uses higher and
higher languages.)

>
> Ah. An oldtimer. I was never around for the early days. I only
>discovered computers with Win98, when it became a car to drive
>and not just an engineering challenge for greasemonkeys. I think
>that crippled my understanding, partly. I'm an auto-didact in

Yes, a bit. Though I very much doubt I'll ever do any assembler again,
having done some does give me a bit of knowledge of what's going on. In
much the same way as having started out in an old-fashioned engineering
firm where I did proper workshop training - learning to use lathe,
milling machine, welder, fitting tools (files etc.), wiring, and so on -
although I'm unlikely now to actually do these, gives me a bit more
understanding of what can be _done_.

>GUI programming. So understanding what really happens underneath
>came later and was hard for me to grasp. It's still not second nature
>for me, the way it is with fully trained C++ programmers.
>
>

On the other hand, you could knock something up in VB very quickly,
whereas I'd have a steep learning curve.

--
J. P. Gilliver. UMRA: 1960/<1985 MB++G()AL-IS-Ch++(p)Ar@T+H+Sh0!:`)DNAf

Remembrance of things past is not necessarily the remembrance of things as
they were. - Marcel Proust

[J. P. Gilliver (John)]

On Tue, 24 May 2022 at 16:14:43, Mayayana <maya...@invalid.nospam>

wrote (my responses usually FOLLOW):

[]

> I think I feel the same way. But the main thing for me is control.
>They want to put the Start menu in the middle? OK. But can I move
>it if I want to? Each version has made it harder to change things
>to the way I find convenient. Ribbon menus? OK. But why can't
>they also be switched back? I want to be able to cruise at top
>speed, without thinking about it. I don't want to be zipping along
>and suddenly get a big window asking whether I want to copy a
>file, replace it, or add with a different name. Huh? Shut up! All I
>need is a simple message: Overwrite existing file?

I think Dan and some others would say that the "control" you want is the
right to change it _back_ - i. e. they'd say you aren't willing to spend
the time learning the new way, even if it is actually better.

I'm with you (-: - my Win7 screen looks very like my 98SE one did.
Basically, I've developed "muscle memory" - and similar phrases - for
how to do things, in some cases very quickly, and I get cross when a
change is _imposed_ on me. But I do concede that _sometimes_ "the new
way" is better, once I've learnt it. I do say "but" a lot though. And
I'm unlikely to change much, especially now: I operate on the "trailing
edge", though I do _try_ to keep up with _some_ aspects of the new.

>
> With Win7 I spent weeks figuring out how to make it civilized.

Yup. Classic Shell, turning off services, and so on ... (-:

>I even wrote my own program to remove file restrictions in bulk.
>I find it reasonably usable but not optimized. With Win10 I just
>avoid using it. It's no
>fun.

Yes. Fighting the OS _can_ be fun - up to a point. But one becomes weary
- especially when you have to do it all over again, just slightly
differently.

> Can I stop it calling home? Not sure. It changes services
>settings without asking. It's crazy bloated. Much of it can be
>customized, but not like it used to be. And now I just use the
>popular tweak programs rather than writing my own. It's not
>worth the trouble. So, yeah, it's OK. But with Win10 I'm kind of
>drawing a line. It's spyware. I probably can't undo that. So if
>I ever have to give up XP I'll probably go to 7 or Linux. If I
>have to give up 7 then I'll probably just hold my nose and switch
>to Linux. Win10 will never be *my* workshop or kitchen. It's
>Microsoft's.
>
> At this point I have a Win10 laptop that I use mostly just
>for online retail research where I need a new browser and to enable
>javascript. It's a system that lives in the gutter. Win10 is a
>lying whore. (No offense to prostitutes.) I wouldn't
>put private data on it. I wouldn't run productivity software on it.
>I didn't even bother to get a real Win10 computer. Just a laptop

I have one I was given for parish council work (lowest level of
government in UK, unpaid). I use it for that - it's just _easier_, for
interacting with others using one (it's set up with Office365, Outlook,
and so on). I occasionally use it if I want to find out how to do
something in a recent Word that I know how to do in my 2003 version - so
I can tell someone else how to do it. It's a nice enough machine - just
no _fun_. When I started in computing, even in PCs, it _was_ fun - and I
still _enjoy_ using this customised 7 machine.

>that I keep in the closet. I take it out when I'm shopping for
>something and want to be able to go to crappy sites like Home
>Depot, department stores, and so on without the webpage being
>all messed up. Many of those sites now require script to see their
>images and are loading webp files. My browsers on XP can't even
>display a webp file.

(I don't even know what webp is, though I'm sure I use such pages.)

> Website compatibility coding is out the window.
>Special effects is all they care about.

Boy are you right there.
[]

> I can get used to changes, but not inflexibility. I don't care if Bill
>Gates thinks my frying pan should be on the left and saucepans on
>the right. I don't care if he thinks my hammers should be where my
>screwdrivers are and vice versa. I didn't ask him. Nearly all these
>problems are not about change but about removing choice.
>

They would argue that if they let you choose, you'd always choose the
old way because it's less effort than learning the new way, for a
benefit that's not always obvious - and thus you'd never benefit. I can
see that argument; I even agree with it in theory; but like you, I want
the _option_ of not changing, even if it means I'm inefficient in the
long run.

[J. P. Gilliver (John)]

On Wed, 25 May 2022 at 09:41:36, wasbit <wasbit...@hotmail.com> wrote
(my responses usually FOLLOW):

>"nospam" <nos...@nospam.invalid> wrote in message
>news:240520221632044747%nos...@nospam.invalid...
>> In article <t6jegk$r24$1...@dont-email.me>, Mayayana
>> <maya...@invalid.nospam> wrote:
>>
>>> I find that most webpages these days are stuck at
>>> 1000px wide. They should self-size, but they don't.

Ah, you may be right at the top end; my main gripe is at the bottom end.

>>
>> i dunno what pages you visit, but nearly all of them reflow with
>> changes in the viewport, even down to mobile sized.

Hm. We obviously visit very different sites! I find reflow to be a
rarity: I have to scroll horizontally a lot more than I once did.
Scripting is _presumably_ the reason; reflow is intrinsic to basic HTML,
so for a page/box/whatever _not_ to reflow means a deliberate decision -
and action - has been taken to stop it happening. Many webpages now are
more like PDF documents - you'll take the width _we_ want, and like it.

>>
>> Snip <
>
>FYI
>There are quite a number of web pages that don't change to reflect the
>size of the viewing area.
>There are even some that you can't read what is outside the viewing
>area because there are no scroll bars.
>

Yes, I've seen those. And/or where an important button - such as the
"next" button - is outside my viewing width, so I don't immediately
realise it exists (and in the case where they've killed off the scroll
bars, can't get to it anyway).

--
J. P. Gilliver. UMRA: 1960/<1985 MB++G()AL-IS-Ch++(p)Ar@T+H+Sh0!:`)DNAf

"Windows 10 is the gift that keeps on giving." Like the clap.
- Paul in alt.comp.os.windows-10 and comp.sys.ibm.pc.hardware.storage and
alt.windows7.general, 2020-10-13

[J. P. Gilliver (John)]

On Wed, 25 May 2022 at 08:24:26, DanS
<t.h.i.s....@r.o.a.d.r.u.n.n.e.r.c.o.m> wrote (my responses
usually FOLLOW):

>"J. P. Gilliver (John)" <G6...@255soft.uk> wrote in
>news:eNo30kyGG2iiFwAq@a.a:
>
>[Top-posted for brevity]
>
>I don't quite understand...
>
>Why can't you move it back, or "un"- resize it ?

[]
I can - but usually, if I've done a move/resize _by mistake_ (because a
mouse movement caught an edge or title bar), I don't know where to
move/resize it back _to_.

Yes, you can say "if you don't know where it was, does it matter?"; to
which I can answer yes - just because I couldn't _tell_ you where it
was, (a) doesn't mean I'm happy with it changing, (b) I may
subconsciously rely on it being where it was, even if I couldn't tell
you.

[J. P. Gilliver (John)]

On Wed, 25 May 2022 at 11:02:49, Zaidy036 <Er...@Bloch.com> wrote (my
responses usually FOLLOW):

Seems like an awful lot of work! I also don't necessarily always put the
windows I use a lot in the same place - but once I've opened them, I
don't want to move/resize them unless I do it deliberately. So
predefined sizes wouldn't help.

[nospam]

In article <t6liej$86g$1...@gioia.aioe.org>, R.Wieser

<add...@not.available> wrote:

>
> >> > optimizing means make it faster.
> >>
> >> No, it doesn't. Not by a long shot.
> >
> > it doesn't?
> >
> >> Though mostly the /effect/ of those
> >> optimalisations is a faster running program.
> >
> > but it does?
> >
> > you just said it doesn't.
>
> Look up the word "means". I'm pretty sure that it doesn't mean what you
> think it does.

i'm pretty sure it does.

i'm also pretty sure that once again, you don't know what you're
talking about, which is why you snipped everything and resorted to
semantic games, as you usually do, rather than discuss the various
points made.

> Kiddo, you have a serious comprehension problem - or that is just the way
> you play your game.

projection.

[nospam]

In article <tVW7Szf6gljiFw+3@a.a>, J. P. Gilliver (John)

<G6...@255soft.uk> wrote:

>
> VB - or any other high-level language - isn't going to be that
> efficient, even if you do optimisation by rewriting it.

maybe not vb, but high level languages with optimizing compilers will
generate *more* efficient and better optimized code than hand coded
assembly in nearly every case, for all sorts of reasons.

> As (I think it
> was) Paul pointed out, a good compiler (or whatever the right word is
> for what processes VB code) can do amazing things, but it's likely never
> to approach real hand-coded code.

that is simply false.

[Mayayana]

"J. P. Gilliver (John)" <G6...@255soft.uk> wrote

| VB - or any other high-level language - isn't going to be that
| efficient, even if you do optimisation by rewriting it. As (I think it
| was) Paul pointed out, a good compiler (or whatever the right word is
| for what processes VB code) can do amazing things, but it's likely never
| to approach real hand-coded code. However, it's a darn sight easier
| (especially not to miss something that will break something); I suspect
| if I ever do any coding again, it won't be in assembler.

Actually it can be made very efficient. By optimizing
I mean finding more efficient methods and cutting out
wrappers. I might do things the easy way but then
gradually make it more lean. It's slow to use an ActiveX
control that's wrapping a system window, but then one
can use the system window with API calls to make it more
efficient. I can use the GDI API or byte arrays to process
images, but then display them in a VB picture box control
for convenience.

VB compiles to native code. It just provides a lot of optional
training wheels. It's often interesting what ends up being
most efficient. But for most things it's not important. Few things
really need speed. Yet I like to keep it lean and avoid dependencies.
So in general, anything I write runs without support files on
any Windows system. (It needs system libraries and the VB
runtime, but nothing else. No ActiveX controls. No secondary
libraries. No .Net, Java, Python, etc.)

In my own code editor I started out using a VB RichEdit
control, so that I could do color syntax highlighting. Over the
years I tried various improvements. I reduced the references
made in a call to RichTextBox.SelColor, for example, when I
wanted to do something like select "<DIV>" and color it blue.
But that was still a wrapper calling SendMessage on a system
window.
Then over time I began using the EM_ SendMessage API calls
to call the wrapped RTB directly. Eventually I took out the Ax
control and started loading the Windows RichEdit directly, so
that I could handle the message loop myself and have more
functionality than the VB wrapper provided.

So I was using a RichEdit just as any Windows program would
use it -- directly, via API calls, in compiled software...

Eventually I discovered that the fastest way, by far, to do the
color syntax highlighting was to point an integer array at the
text string of code in the editor window and run it through a
tokenizing routine. That allowed me to parse the text numerically.
(For example, with HTML I'd be looping through the array looking
for 60 (<), 62 (>), etc.
The original text string would then get rebuilt, inserting RTF
encoding directly to color the text. I had just assumed that API
calls were fastest. But a string processed without concatenation
is extremely fast, and a RichEdit loads RTF text extremely fast.

It's possible that hand drawing the entire window would be fastest,
but that would be a whole other level of programming.

I find that kind of challenge fun. But it must be practical or
it doesn't hold my interest. It's hard to write software that I
don't need myself.

| >
| > I dabbled a bit in C++ and could see the appeal. Compared
| >to VB it's like riding without a seat belt or a floor. But I just
| >wasn't ambitious enough to master it. And it's a damned ugly
| >language.
|
| (I never got into the plusses. I did learn raw C. Yes, there's a
| definite appeal to it. [And definite cliquiness too, which I've been
| guilty of too, though I hope less than some.])

I never knew you were such an oldtimer geek. But actually,
I think both C and C++ are still in big demand. You could probably
get a job, if you could get past the 19 year old HR director. :)

| With PCs, when I came in - probably 1980s, I can't remember - you could
| definitely save money by selecting and assembling the various component
| modules yourself, and also get something very much more what you wanted;
| I think the economic side faded out about the start of XP (not least
| because the cost of Windows shot up about then). Over the years after
| that, the performance of the various bits improved to the extent that,
| although some people do still assemble their own for the ultimate in
| either performance or some other parameter (including appearance!),
| they're very much in the minority: if I go into a branch of PC World
| here now, I'm not sure if I can even find things like memory,
| processors, graphics cards - only ready-built PCs.
|

I'm sitting at a machine I built myself in 2015. About
$400. But hardware prices are way up. And compatibility
is more limited. I'm not sure I'll build again, either. When a
CPU is $250 and a PC is $400, there's no point.

[Mayayana]

"J. P. Gilliver (John)" <G6...@255soft.uk> wrote

| I think Dan and some others would say that the "control" you want is the
| right to change it _back_ - i. e. they'd say you aren't willing to spend
| the time learning the new way, even if it is actually better.
|

Yes, but better is a matter of opinion. I want the option.
And most of the changes are not necessarily improvements.
They're usually changes to make it easier for the average person
and reduce any impression of bugginess. Some changes are
just to make things look new, like the chrome trim on 2022 cars.

Start Menu in the middle? Fine. Make it movable. There's no
excuse for changing it and making it fixed. That used to be the
design logic. All programs should have the same interface, the
same basic menus, and respect user choice of GUI. So you could
get a new program and immediately know what's on the Edit menu,
for example. Over time those norms have been disrespected. When
I insiatlled Chromium on Raspberry Pi (Linux) it was impossible to
show a menu bar. I can only show a flyout menu from 3 dots on
the upper right. But that's phone design, not desktop design. I'm
not being stubborn to want a menu bar designed for a large screen
and mouse.

| (I don't even know what webp is, though I'm sure I use such pages.)
|

I hadn't noticed it until recently. I was trying to
figure out why a picture wouldn't show in a webpage.
I downloaded the JPG. IrfanView said it's not really a
JPG. It's a webp. ! Yet they named it JPG. So I downloaded
the IV plugin for webp. Now I have exactly one program
that recognizes webp. Yet commercial websites are using
them routinely. I don't even use PNG online, in case
someone's browser can't render it.

| > I can get used to changes, but not inflexibility. I don't care if Bill
| >Gates thinks my frying pan should be on the left and saucepans on
| >the right. I don't care if he thinks my hammers should be where my
| >screwdrivers are and vice versa. I didn't ask him. Nearly all these
| >problems are not about change but about removing choice.
| >
| They would argue that if they let you choose, you'd always choose the
| old way because it's less effort than learning the new way, for a
| benefit that's not always obvious - and thus you'd never benefit.

There you go again. :) Why so passive? You're the one using
the product. You bought it. And as I said, there's a lot more that
goes into design than good design. There's fashion, incompetence,
irrelevant priorities... Recently I bought a new shop vac. 6' cord.
The old one had a 15' cord. Is 6' an improvement? Of course
not. It's all but useless for my purposes. So I spliced the old cord
onto the new one and now I have 21'. So I don't have a carry an
extension cord with my shop vac. That's not stubbornness. It's just
adapting the product to what I actually need. And the short cord
is not design. It's just cost-cutting.

[J. P. Gilliver (John)]

On Wed, 25 May 2022 at 14:50:33, Mayayana <maya...@invalid.nospam>

wrote (my responses usually FOLLOW):

[]

> VB compiles to native code. It just provides a lot of optional
>training wheels. It's often interesting what ends up being
>most efficient. But for most things it's not important. Few things
>really need speed.

Indeed.

> Yet I like to keep it lean and avoid dependencies.

I appreciate your efforts in that direction.

>So in general, anything I write runs without support files on
>any Windows system. (It needs system libraries and the VB
>runtime, but nothing else. No ActiveX controls. No secondary
>libraries. No .Net, Java, Python, etc.)

I wish more would do that!
[]

> I find that kind of challenge fun. But it must be practical or
>it doesn't hold my interest. It's hard to write software that I
>don't need myself.
>

Indeed.

>| >
>| > I dabbled a bit in C++ and could see the appeal. Compared
>| >to VB it's like riding without a seat belt or a floor. But I just
>| >wasn't ambitious enough to master it. And it's a damned ugly
>| >language.
>|
>| (I never got into the plusses. I did learn raw C. Yes, there's a
>| definite appeal to it. [And definite cliquiness too, which I've been
>| guilty of too, though I hope less than some.])
>
> I never knew you were such an oldtimer geek. But actually,
>I think both C and C++ are still in big demand. You could probably
>get a job, if you could get past the 19 year old HR director. :)

(-:! I never actually passed any exams in C, and have to refresh my
knowledge occasionally when looking at some code. The computing part of
my electronics degree actually taught us Algol, which I hardly very hear
mention of now - but the _concepts_ are transferrable to C (and some
other languages). I did my degree around 1980.
[]

> I'm sitting at a machine I built myself in 2015. About

Yes, but you're on of those here (wherever "here" is as I started this
in three 'groups - though this probably applies in all three of them).
The average TL;DR Joe Public would never even think of building a PC.
(The younger generation might even not buy a PC as we know it.)

>$400. But hardware prices are way up. And compatibility
>is more limited. I'm not sure I'll build again, either. When a
>CPU is $250 and a PC is $400, there's no point.
>

Indeed.

>
--
J. P. Gilliver. UMRA: 1960/<1985 MB++G()AL-IS-Ch++(p)Ar@T+H+Sh0!:`)DNAf

science is not intended to be foolproof. Science is about crawling toward the
truth over time. - Scott Adams, 2015-2-2

[wasbit]

"nospam" <nos...@nospam.invalid> wrote in message

news:250520220911064307%nos...@nospam.invalid...

Not off hand.
I just swear & move on. It's not something I keep a record of but it's
blasted annoying when you do come across one.

--
Regards
wabit

[Mayayana]

"wasbit" <wasbit...@hotmail.com> wrote

| >> There are even some that you can't read what is outside the viewing
area
| >> because there are no scroll bars.
| >
| > i've never seen that. do you have an example?
|
| Not off hand.
| I just swear & move on. It's not something I keep a record of but it's
| blasted annoying when you do come across one.
|

I've seen it. That's one of the many reasons that I
have a CSS toggle button the the browser toolbar.
I suspect it's another case of trying to make sure the
page won't work without script.

I've even seen pages (Forbes used to do it) where the
basic HTML is put into script in the page. So nothing is
there unless you let the script run. And the script's job
is the write the HTML to the document! That's not an easy
thing to accomplish. The only possible reason is the break the
webpage for people blocking spying and personalized ads.

[Mayayana]

"J. P. Gilliver (John)" <G6...@255soft.uk> wrote

| I think Dan and some others would say that the "control" you want is the
| right to change it _back_ - i. e. they'd say you aren't willing to spend
| the time learning the new way, even if it is actually better.
|

I saw an article this AM that reminded me of this
discussion:

https://www.businessinsider.com/tesla-sued-by-family-of-man-died-model-s-fire-2019-10?op=1

A man crashed his Tesla, the battery caught fire, then he
burned to death because only the car's computer can decide
when to make the door handles accessible, so bystanders
couldn't open the doors to help the man get out. That comes
on top of numerous reports of people killed by faulty auto-pilot
functionality. Yet people are determined to make cars self
driving. Why? Because it's the future, and the future is better.

[Andy Burns]

Mayayana wrote:

> A man crashed his Tesla, the battery caught fire, then he
> burned to death because only the car's computer can decide
> when to make the door handles accessible, so bystanders
> couldn't open the doors to help the man get out.

I thought Elon demonstrated that Tesla windows *are* still breakable?

[Paul]

"A man crashed his Tesla"

The tempered glass will break, if you apply one of
those spring-loaded window breaker tools to it.

But there's a technique. Like operating RAID hard drives,
it's "practice, practice, practice" :-/ You do that in
a corner of the window, rather than dead center.

https://www.youtube.com/watch?v=2TM5r37a3ng

Those tools also include a belt-cutter, for when
you're upside-down and the mechanism is jammed. You want the
tool to be located where you can reach it in the car.
The glove box would be a daft storage location. I don't
expect cutting a belt to be easy.

And the "big ole round boulder" works too. That's how
they got into my car one day, to relieve the car of some tools.
They even left the boulder behind, as a souvenir. Thoughtful.
The tools were out-of-sight, so a passerby, while I had the
trunk open, probably saw the tools and passed the word along.

Paul

[Paul]

I've seen the webpage without the scroll bars too, but not lately.
It could have been a passing fad.

And there is a way to make extra large display surfaces... on Linux.
With an extra large display surface, we don't need stinkin scroll bars.
Because I just tested this yesterday :-) This is an easier procedure
than it used to be (editing Xorg.conf not necessary). The gsettings
steps are only if you want to clean up the desktop for photo work
(fewer distractions).

***************** Copy/Pasta ******************
Tested on Ubuntu 20.04 with Gnome desktop.

xrandr # discover names of outputs

xrandr --output HDMI-0 --panning 1280x10000 # my small screen, panning in Y only

gsettings set org.gnome.desktop.background picture-uri none # remove bkgd
gsettings set org.gnome.desktop.background primary-color '#3ea5a6" # pick a color

firefox http://something.interesting & # Compose the screenshot

sleep 10 ; xwd -root -out xwd.xwdump # delay time to iconify terminal.
# xwdump is to allow verification that
# output is 24-bit. (51,203,179 bytes)
# (32-bit pixels manipulate faster)

Opened with GIMP, it looks pretty good. This is scaled down 8x
and just to show the coverage of Yahoo News graphics is decent.

https://i.postimg.cc/qqYDckcZ/proof-print.jpg
***************** End Copy/Pasta ******************

Paul

[Mayayana]

"Paul" <nos...@needed.invalid> wrote

| The tempered glass will break, if you apply one of
| those spring-loaded window breaker tools to it.
|

I wondered about that. But I suppose it was probably
a time issue. People tried to get him out... no handle... now
what... woops. Too late.

[nospam]

In article <t6npte$kv7$1...@dont-email.me>, Mayayana
<maya...@invalid.nospam> wrote:

> https://www.businessinsider.com/tesla-sued-by-family-of-man-died-model-s-fire-

> 2019-10?op=1
>
> A man crashed his Tesla, the battery caught fire, then he
> burned to death because only the car's computer can decide
> when to make the door handles accessible, so bystanders
> couldn't open the doors to help the man get out.

that sounds like yet another frivolous lawsuit.

the link states that the vehicle veered off a parkway and hit a palm
tree. that sounds like he wasn't paying attention and driving in an
unsafe manner, which means that the door handles aren't the reason he
died.

many serious crashes result in doors becoming inoperable, which is why
there's the 'jaws of life', to extract people from such wrecks.

it is *not* unique to tesla.

> That comes
> on top of numerous reports of people killed by faulty auto-pilot
> functionality.

now compare that number versus the numerous reports of people killed by
faulty human drivers, including those who are drunk, distracted and
incompetent.

those crashes don't make the news because there are so many of them to
report. people just accept the fact that humans fuck up and crash into
things.

normalized to vehicle miles traveled, autonomous vehicles are safer
than human drivers.

there are approximately 40k vehicle fatalities in the usa per year,
which is more than 100 deaths *per* *day*. many, many more are injured.

keep in mind that some of those crashes are the fault of new and
inexperienced drivers, some of whom do not even have a license and are
driving with a learner's permit, still learning what to do and not do.

here's just a few examples:
<https://www.dispatch.com/story/news/disaster/2013/06/28/teen-killed-in-
car-wreck/23668046007/>
Jacob had just passed his driverąs license test early this morning,
only a few hours before the Chevrolet Cavalier he was driving hit a
utility pole, Cvetan said.

Jacob was thrown from the car and died at the scene. His passengers ‹
sisters who are Jacobąs cousins, Renee N. Farmer, 15, and Janay E.
Farmer, 16, both of Columbus ‹ also were thrown from the car.

<https://nypost.com/2022/02/10/girl-killed-by-driver-with-learners-permi
t-ided-as-davina-afokoba/>
The distraught mother of the 10-girl-old girl who was fatally struck
by a motorist with only a learnerąs permit tearfully described her
youngest daughter Thursday as her łbaby˛ and her łmini-me.˛

<https://gothamist.com/news/teen-driver-with-learners-permit-kills-4-fri
ends-in-southern-state-parkway-crash>
Yesterday, a Subaru swerved off the road on the westbound Southern
State Parkway between exits 17 and 18 and then crashed into the
woods, killing four passengers. The driver survived and has been
identified as Queens resident Joseph Beer, 17‹and he only had a
learner's permit.

> Yet people are determined to make cars self
> driving. Why?

because it will greatly reduce the number of injuries and fatalities
compared to human drivers.

the number of crashes will never be zero, nor does it need to be.

crashes will still occur with autonomous vehicles, just not as many nor
as severe, which is a significant benefit that will save many lives and
reduce and eliminate many injuries.

it only needs to be less than with human drivers, which sadly, is not
that difficult.

> Because it's the future, and the future is better.

it is. try it sometime. the water's warm. join in the fun.

[J. P. Gilliver (John)]

On Thu, 26 May 2022 at 07:56:30, Mayayana <maya...@invalid.nospam>

wrote (my responses usually FOLLOW):

Normally, I strongly dislike people who say things like "straw man", but
... "numerous reports of people killed by faulty auto-pilot
>functionality" is not the same as "all self-driving cars should be
banned". (As another has pointed out, in terms of passenger-miles,
they've probably passed human drivers already.)

The car door handles thing was indeed something that needed looking
into. I presume it has been, and recalls (for reprogramming) issued if
necessary.

--
J. P. Gilliver. UMRA: 1960/<1985 MB++G()AL-IS-Ch++(p)Ar@T+H+Sh0!:`)DNAf

... a series about a grumpy old man who lives in a phone box is unlikely to
have been commissioned these days. 798 episodes later ...

[DanS]

"J. P. Gilliver (John)" <G6...@255soft.uk> wrote in
news:Tq9EtBhVyljiFw7N@a.a:

> On Tue, 24 May 2022 at 16:14:43, Mayayana
> <maya...@invalid.nospam> wrote (my responses usually
> FOLLOW): []
>> I think I feel the same way. But the main thing for me
>> is control.
>>They want to put the Start menu in the middle? OK. But can
>>I move it if I want to? Each version has made it harder to
>>change things to the way I find convenient. Ribbon menus?
>>OK. But why can't they also be switched back? I want to be
>>able to cruise at top speed, without thinking about it. I
>>don't want to be zipping along and suddenly get a big
>>window asking whether I want to copy a file, replace it, or
>>add with a different name. Huh? Shut up! All I need is a
>>simple message: Overwrite existing file?
>
> I think Dan and some others would say that the "control"
> you want is the right to change it _back_ - i. e. they'd
> say you aren't willing to spend the time learning the new
> way, even if it is actually better.

That is one thing I may have been trying to say.

You mention Classic Shell below, because, of course the 'Live Tiles' menu is
horrificaly nasty.

...but, some people like it? (They do. The question mark is my disbelief.)

So, I can cry about it and suffer thoruhg it like a martyr, I can write my own (which I'm
not going to do, again) or I can use free Classic Shell, or pay $5 for the Stardock Startx
offering.

In fact, I chose to buy the Stardock Suite, because for $50 (at the time) I found several
of the included titles of use. At one time, I would just have procurred a cracked copy,
but, now can afford, to some extent, software I find useful.)

>
> I'm with you (-: - my Win7 screen looks very like my 98SE
> one did. Basically, I've developed "muscle memory" - and
> similar phrases - for how to do things, in some cases very
> quickly, and I get cross when a change is _imposed_ on me.
> But I do concede that _sometimes_ "the new way" is better,

<SNIP>

Except with the Windows 10 GUI. WindowBlinds is anohter very useful utility out of that
suite I use. It works very, very well (for me) and is nothing like that buggy, crash
inducing cr*p that was introduced a tthe end of the last millennium.

I just feel the flat, square, "window decorations" are a huge step backwards from
Windows 7. And, as you can see, I called them "window decorations" because that's
what the *nix term is. In *nix, it;s just built in that you can change all that stuff to suite
your liking.

Many people like the flat, ugly look of Windows 3.x. Not me. Sure, it doesn't add any
productivity, but it doesn't cause any loss of productivity either.

One thing that is a HUUUUUUUUUUGE improvement in Win10 over 7, is the boot time.

Sure, it doesn't matter THAT much, because all my home PCs and work PCs are on
24/7, but when I do need to reboot....

[DanS]

"Mayayana" <maya...@invalid.nospam> wrote in
news:t6lvbb$khe$1...@dont-email.me:

> "J. P. Gilliver (John)" <G6...@255soft.uk> wrote
>
>| I think Dan and some others would say that the "control"
>| you want is the right to change it _back_ - i. e. they'd
>| say you aren't willing to spend the time learning the new
>| way, even if it is actually better.
>|
>
> Yes, but better is a matter of opinion. I want the
> option.
> And most of the changes are not necessarily improvements.
> They're usually changes to make it easier for the average
> person and reduce any impression of bugginess. Some changes
> are just to make things look new, like the chrome trim on
> 2022 cars.

...perhaps Linux is what you are looking for ?

Nearly cpletely customizable, with almost all source code avaliable for you to customize it
exactly how you want it.

[Mayayana]

"DanS" <t.h.i.s....@r.o.a.d.r.u.n.n.e.r.c.o.m> wrote

| ...perhaps Linux is what you are looking for ?
|
| Nearly cpletely customizable, with almost all source code avaliable for
you to customize it
| exactly how you want it.

It might come to that. But I have a lot of expertise
invested in Windows. And now Linux is starting to go
the other way.

The way I used to see it was that Macs were a sportscar
with the hood welded shut. Good as fashion accessories or
to pick up groceries, but not so practical. And way too cute
and sleazy. Linux was like a car kit for greasemonkeys. Windows
was in between, providing tools for any level of expertise.
A basic, useful car. It could be what I wanted and there was
loads of software. MS were catering to corporate workers, so
they focused on usability, customizing, good tools for writing
software, and amazing backward compatibility.

Now Windows is going south fast. MS are trying to sneak into
my driveway and replace my car with a taxi. But Linux is going
in the direction of Mac, with a kind of split personality. On one
extreme it's still an unfriendly car kit, with a culture of unsocialized
geeks who squeal at the very idea of having to write documentation.
At the other extreme it's becoming locked down. Software installs
and updates have become automated. I'm increasingly faced with
either having no options or digging down into /etc.

Awhile back I repeated an occasional experiment: I download
various Linux installers and try them out. I have a basic usability
test: Can I install an easy-to-use firewall to block outgoing
processes? Can I set it up the way I want it without resorting to
console windows? Is WINE perhaps good enough to run my
Windows software? For years now my test fails quickly. The designers
are people who only see engineers and dummies. Programmers
and grandmas. Last time I tried, I think it was Red Hat that
overwrote my boot manager without asking and had no option
to install without doing that. I asked in a reddit group. I was told
that it's supposed to do that. "No one understands multibooting.
Too much trouble." Huh?! Another version demanded a password
during setup, but I hadn't configured a password. It was a nightmare.
In no time I was back into commandline and online research; in
some case before I even got the thing installed! Then, of course,
the Linux freaks screamed at me when I complained on Reddit.
It's always their way or the highway... Yes, you can have it whatever
way you like, if you want to rewrite it yourself.

I remember a story about Richard Stallman. It seems that he
was at some kind of conference, giving a talk. A man got up and
said he was having trouble getting Stallman's compiler to work.
Stallman answered, "Then write your own." :)

[Mayayana]

"DanS" <t.h.i.s....@r.o.a.d.r.u.n.n.e.r.c.o.m> wrote

| I just feel the flat, square, "window decorations" are a huge step
backwards from
| Windows 7.

That's an interesting thing to my mind. Design fads seem to
go in patterns. Pre-industrial cultures often celebrate bright
colors because they're rare. In the 60s we had day-glo. Today
you can have pretty much any color you like. So what happens?
People fret over tiny differences between off-white and dustball
hues. Similarly, GUIs were big on 3-D and curves, until that
became easy. Then flat with subtle colors started being the thing.
We always want what we don't have.

Similarly, maple cabinets were the only thing to buy for decades,
as tech became tendy and people wanted slick. Maple is among
the least "woody" woods. These days the trend is all over, with
dark stains and painted colors being popular. But oak is making a
slight comeback, as we worry about losing natural landscapes.

[DanS]

"Mayayana" <maya...@invalid.nospam> wrote in
news:t6tfdg$lfb$1...@dont-email.me:

LOL!!! ***THAT'S*** what you remeber of Richard Stallman?

First thing that comes to mind when I hear his name...is the time he was giving a
lecture about FOSS, I think it was, and he proceeded to take off one of his shoes, then
the sock, then dug something from between his toes....and ate it, right there on stage.

It made me almost want to hurl....

[nospam]

In article <XnsAEA58B3F3EF86th...@216.166.97.131>, DanS

<t.h.i.s....@r.o.a.d.r.u.n.n.e.r.c.o.m> wrote:

> > I remember a story about Richard Stallman. It seems that
> > he
> > was at some kind of conference, giving a talk. A man got up
> > and said he was having trouble getting Stallman's compiler
> > to work. Stallman answered, "Then write your own." :)
>
> LOL!!! ***THAT'S*** what you remeber of Richard Stallman?
>
> First thing that comes to mind when I hear his name...is the time he was
> giving a
> lecture about FOSS, I think it was, and he proceeded to take off one of his
> shoes, then
> the sock, then dug something from between his toes....and ate it, right there on stage.

stallman would sexually harass women at mit (and elsewhere).

as it turns out, he hates plants, so women kept lots of plants in their
offices to repel him.

there is also the 'anyone but stallman' license:
<https://github.com/benlk/misc-licenses/blob/master/stallman-license.md>
1 The licensed work may not be used by GNU RMS or his descendants.