Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > alt.folklore.computers > #213748 > unrolled thread

When did mainframes lose commercial dominance?

Back to article view | Back to alt.folklore.computers

Contents

  When did mainframes lose commercial dominance? Thomas Koenig <tkoenig@netcologne.de> - 2020-09-05 10:12 +0000
    Re: When did mainframes lose commercial dominance? Dan Espen <dan1espen@gmail.com> - 2020-09-05 06:55 -0400
      Re: When did mainframes lose commercial dominance? Bob Eager <news0073@eager.cx> - 2020-09-05 11:19 +0000
        Re: When did mainframes lose commercial dominance? Niklas Karlsson <anksil@yahoo.se> - 2020-09-05 12:31 +0000
          Re: When did mainframes lose commercial dominance? Dan Espen <dan1espen@gmail.com> - 2020-09-05 09:26 -0400
            Re: When did mainframes lose commercial dominance? J. Clarke <jclarke.873638@gmail.com> - 2020-09-05 12:04 -0400
              Re: When did mainframes lose commercial dominance? Dan Espen <dan1espen@gmail.com> - 2020-09-05 12:26 -0400
                Re: When did mainframes lose commercial dominance? J. Clarke <jclarke.873638@gmail.com> - 2020-09-05 12:57 -0400
                  Re: When did mainframes lose commercial dominance? Dan Espen <dan1espen@gmail.com> - 2020-09-05 14:03 -0400
                    Re: When did mainframes lose commercial dominance? J. Clarke <jclarke.873638@gmail.com> - 2020-09-05 15:15 -0400
                      Re: When did mainframes lose commercial dominance? Dan Espen <dan1espen@gmail.com> - 2020-09-05 16:34 -0400
                        Re: When did mainframes lose commercial dominance? J. Clarke <jclarke.873638@gmail.com> - 2020-09-05 17:02 -0400
                          Re: When did mainframes lose commercial dominance? Dan Espen <dan1espen@gmail.com> - 2020-09-05 17:27 -0400
                            Re: When did mainframes lose commercial dominance? J. Clarke <jclarke.873638@gmail.com> - 2020-09-05 20:02 -0400
                              Re: When did mainframes lose commercial dominance? Dan Espen <dan1espen@gmail.com> - 2020-09-05 21:50 -0400
                Re: When did mainframes lose commercial dominance? Dave Garland <dave.garland@wizinfo.com> - 2020-09-05 12:03 -0500
                  Re: When did mainframes lose commercial dominance? Dan Espen <dan1espen@gmail.com> - 2020-09-05 14:07 -0400
              Re: When did mainframes lose commercial dominance? Peter Flass <peter_flass@yahoo.com> - 2020-09-05 17:53 -0700
                Re: When did mainframes lose commercial dominance? J. Clarke <jclarke.873638@gmail.com> - 2020-09-05 20:57 -0400
            Re: When did mainframes lose commercial dominance? Peter Flass <peter_flass@yahoo.com> - 2020-09-05 17:38 -0700
              Re: When did mainframes lose commercial dominance? Dan Espen <dan1espen@gmail.com> - 2020-09-05 21:58 -0400
              Re: When did mainframes lose commercial dominance? Ahem A Rivet's Shot <steveo@eircom.net> - 2020-09-06 07:10 +0100
              Re: When did mainframes lose commercial dominance? scott@slp53.sl.home (Scott Lurndal) - 2020-09-06 20:05 +0000
              Re: When did mainframes lose commercial dominance? Charlie Gibbs <cgibbs@kltpzyxm.invalid> - 2020-09-08 21:10 +0000
          Re: When did mainframes lose commercial dominance? Bob Eager <news0073@eager.cx> - 2020-09-05 13:30 +0000
          Re: When did mainframes lose commercial dominance? Jorgen Grahn <grahn+nntp@snipabacken.se> - 2020-09-05 19:17 +0000
            Re: When did mainframes lose commercial dominance? J. Clarke <jclarke.873638@gmail.com> - 2020-09-05 16:20 -0400
              Re: When did mainframes lose commercial dominance? Jorgen Grahn <grahn+nntp@snipabacken.se> - 2020-11-19 07:15 +0000
            Re: When did mainframes lose commercial dominance? Dan Espen <dan1espen@gmail.com> - 2020-09-05 16:41 -0400
        Re: When did mainframes lose commercial dominance? Jorgen Grahn <grahn+nntp@snipabacken.se> - 2020-09-05 18:59 +0000
          Re: When did mainframes lose commercial dominance? Bob Eager <news0073@eager.cx> - 2020-09-05 20:01 +0000
            Re: When did mainframes lose commercial dominance? Ahem A Rivet's Shot <steveo@eircom.net> - 2020-09-05 22:06 +0100
              Re: When did mainframes lose commercial dominance? Bob Eager <news0073@eager.cx> - 2020-09-06 08:21 +0000
                Re: When did mainframes lose commercial dominance? Ahem A Rivet's Shot <steveo@eircom.net> - 2020-09-06 10:26 +0100
            Re: When did mainframes lose commercial dominance? Jorgen Grahn <grahn+nntp@snipabacken.se> - 2020-11-14 08:58 +0000
        Re: When did mainframes lose commercial dominance? Peter Flass <peter_flass@yahoo.com> - 2020-09-05 17:38 -0700
          Re: When did mainframes lose commercial dominance? Bob Eager <news0073@eager.cx> - 2020-09-06 08:20 +0000
        Re: When did mainframes lose commercial dominance? Charlie Gibbs <cgibbs@kltpzyxm.invalid> - 2020-09-08 21:10 +0000
          Re: X et al, When did mainframes lose commercial dominance? John Levine <johnl@taugh.com> - 2020-09-08 22:29 +0000
          Re: When did mainframes lose commercial dominance? Niklas Karlsson <anksil@yahoo.se> - 2020-09-09 19:20 +0000
      Re: When did mainframes lose commercial dominance? J. Clarke <jclarke.873638@gmail.com> - 2020-09-05 09:44 -0400
      Re: When did mainframes lose commercial dominance? Ahem A Rivet's Shot <steveo@eircom.net> - 2020-09-05 16:59 +0100
    Re: When did mainframes lose commercial dominance? Jason Evans <jsevans@mailfence.com> - 2020-09-05 14:20 +0000
      Re: When did mainframes lose commercial dominance? Thomas Koenig <tkoenig@netcologne.de> - 2020-09-05 16:27 +0000
        Re: When did mainframes lose commercial dominance? Dan Espen <dan1espen@gmail.com> - 2020-09-05 12:58 -0400
          Re: When did mainframes lose commercial dominance? J. Clarke <jclarke.873638@gmail.com> - 2020-09-05 13:01 -0400
        Re: When did mainframes lose commercial dominance? Jorgen Grahn <grahn+nntp@snipabacken.se> - 2020-09-05 19:38 +0000
          Re: When did mainframes lose commercial dominance? J. Clarke <jclarke.873638@gmail.com> - 2020-09-05 16:24 -0400
    Re: When did mainframes lose commercial dominance? Rick Umali <rickumali@gmail.com> - 2020-09-05 19:40 +0000
    Re: When did mainframes lose commercial dominance? hancock4@bbs.cpcn.com - 2020-09-08 11:22 -0700
      Re: When did mainframes lose commercial dominance? Niklas Karlsson <anksil@yahoo.se> - 2020-09-09 16:02 +0000
        Re: When did mainframes lose commercial dominance? Thomas Koenig <tkoenig@netcologne.de> - 2020-09-09 16:56 +0000
    Re: When did mainframes lose commercial dominance? massimo79m@gmail.com - 2020-09-12 01:39 -0700
      Re: When did mainframes lose commercial dominance? hancock4@bbs.cpcn.com - 2020-09-17 13:28 -0700

Page 1 of 3  [1] 2 3  Next page →

#213748 — When did mainframes lose commercial dominance?

Clearly, in the 1980s, mainframes ruled the commercial world, at
least for large companies.

By that time, they had mostly lost the crown of scientific
calculation to the minis (with the exception of vector computers).

So, when did they lose their dominant commercial position?

At the company I used to work for, they had a home-grown accounting
system running on mainframes, which they then replaced with an ERP
system in the late 1990s.  That system didn't need the mainframes
any more, so they were deactivated.

(You could argue that a company which does that replaces a hardware
vendor lock with the mother of all software vendor locks.)

Today, some SAP products seem to run on a zSystem, but there is
no certified hardware for HANA, their high-speed in-memory database,
for that architecture.  It's only Intel and POWER.

So, when did the mainframes start losing the commercial computing
centers?  Is around 2000-ish the right timeframe?

[toc] | [next] | [standalone]

#213749

Thomas Koenig <tkoenig@netcologne.de> writes:

> Clearly, in the 1980s, mainframes ruled the commercial world, at
> least for large companies.
>
> By that time, they had mostly lost the crown of scientific
> calculation to the minis (with the exception of vector computers).
>
> So, when did they lose their dominant commercial position?
>
> At the company I used to work for, they had a home-grown accounting
> system running on mainframes, which they then replaced with an ERP
> system in the late 1990s.  That system didn't need the mainframes
> any more, so they were deactivated.
>
> (You could argue that a company which does that replaces a hardware
> vendor lock with the mother of all software vendor locks.)
>
> Today, some SAP products seem to run on a zSystem, but there is
> no certified hardware for HANA, their high-speed in-memory database,
> for that architecture.  It's only Intel and POWER.
>
> So, when did the mainframes start losing the commercial computing
> centers?  Is around 2000-ish the right timeframe?

We've all probably got our own little peephole that we look at computing
history through.

I remember discussion among my peers about automating dentist offices,
gas stations, etc. there was  a huge untapped market there, but S/360
was priced way out of that market.  Then in the 90s PCs started
to enter that market.  That's when I saw the future of the mainframe.

I'd guess 2000 is about right for the start of significant
movement off the mainframe.  But of course there are plenty of
large companies that are still on mainframes and will remain there
for the foreseeable future.


-- 
Dan Espen

[toc] | [prev] | [next] | [standalone]

#213750

On Sat, 05 Sep 2020 06:55:08 -0400, Dan Espen wrote:

> We've all probably got our own little peephole that we look at computing
> history through.
> 
> I remember discussion among my peers about automating dentist offices,
> gas stations, etc. there was  a huge untapped market there, but S/360
> was priced way out of that market.  Then in the 90s PCs started to enter
> that market.  That's when I saw the future of the mainframe.
> 
> I'd guess 2000 is about right for the start of significant movement off
> the mainframe.  But of course there are plenty of large companies that
> are still on mainframes and will remain there for the foreseeable
> future.

Around 1990 (in a university computer centre) one of the senior tech guys 
gave a presentation. In it, he said that PCs were never going to catch on 
in a big way (i.e. one on every desk) but that the future was X 
terminals. He demonstrated how wonderful they were.

The only reason there weren't many PCs in that place was the director. 
Nice guy, but he'd taken against PCs for soem reason. He retired a year 
later, and the tech guy suddenly found PCs everywhere. Not many X 
terminals, even as PCs.

(I have to say that his idea of an X terminal's usefulness was how many 
command prompt windows it could display).



-- 
Using UNIX since v6 (1975)...

Use the BIG mirror service in the UK:
 http://www.mirrorservice.org

[toc] | [prev] | [next] | [standalone]

#213754

On 2020-09-05, Bob Eager <news0073@eager.cx> wrote:
>
> (I have to say that his idea of an X terminal's usefulness was how many 
> command prompt windows it could display).

That does seem to be how most hardcore *nix geeks use X.

I'm not really a fan of X, especially what it's become today with a
bunch of inconsistent window managers and other systems, umpteen
different ways copy/paste works depending on which window you're in,
etc. Perhaps Wayland is an improvement somehow - I've not really used it
yet.

Windows 10 and macOS have their flaws, to be sure, but their GUIs are
pretty agreeable these days. And with PuTTY (or something fancier that
keeps an easily accessible database of machines you connect to) or ssh
in Terminal.app, you might as well be in *nix anyhow. And of course, in
the case of macOS, you actually ARE.

Niklas
-- 
 "... I've seen Sun monitors on fire off the side of the multimedia lab.
 I've seen NTU lights glitter in the dark near the Mail Gate.
 All these things will be lost in time, like the root partition last week.
 Time to die...". - Peter Gutmann in alt.sysadmin.recovery 

[toc] | [prev] | [next] | [standalone]

#213755

Niklas Karlsson <anksil@yahoo.se> writes:

> On 2020-09-05, Bob Eager <news0073@eager.cx> wrote:
>>
>> (I have to say that his idea of an X terminal's usefulness was how many 
>> command prompt windows it could display).
>
> That does seem to be how most hardcore *nix geeks use X.

Where I worked, we mostly used Sun Workstations but we had a few X
terminals.  The X terminals were just as useful as the workstations.
(You could go to any number of other machines for computing.)
I got the impression the X terminals weren't cheap enough to catch on.

> I'm not really a fan of X, especially what it's become today with a
> bunch of inconsistent window managers and other systems, umpteen
> different ways copy/paste works depending on which window you're in,
> etc. Perhaps Wayland is an improvement somehow - I've not really used it
> yet.

As a big fan of Fvwm and a past X windows networked user I think Wayland
is a bad joke.  A typical case of someone wanting to blaze a new path
rather than enhancing what's already working.

I watched the Wayland demos and every one of them talked about
"tearing".  I've got years of using X windows, I have no idea what
tearing they're talking about.  Likewise I thought their arguments about
no one needing networking were (well I don't need it so you don't
either).

So, not a fan of Wayland.  Fortunately, from what I'm reading,
Wayland has yet to show it's superiority.  I wish it would just die and
go away.

As for consistency, I notice all the mainstream desktop managers
change with every release.  My WM, Fvwm only changes when I change
my configuration file.  Fvwm gets new features, it just doesn't force
change with every release.  My desktop is pretty much the same as it
was 20 years ago.  That's consistent.

> Windows 10 and macOS have their flaws, to be sure, but their GUIs are
> pretty agreeable these days. And with PuTTY (or something fancier that
> keeps an easily accessible database of machines you connect to) or ssh
> in Terminal.app, you might as well be in *nix anyhow. And of course, in
> the case of macOS, you actually ARE.

Back when I had to connect to many different machines at work
I had a GUI I wrote hooked up to my Fvwm configuration.
If I had a new server say "treerat", I'd go into my GUI and pick a
shortcut letter for the server, probably "t" if it wasn't already used.
I'd enter the "t", the server host name and some arguments to the
terminal program (for example a background color pattern).

The command line to start the terminal used the host name as a -name
argument to the terminal program.  That made it possible for Fvwm
to tell one terminal from another.

Then I'd be able to type "shift t" to launch a new terminal on that
host, or "meta t" to do the same thing iconified, or just "t" to warp to
an existing terminal on treerat.

That made it possible to access all my servers with a single root window
keystroke.

The GUI also built an Fvwm menu for server access so if I didn't want
to devote a hot key I could still get to the server.

I tried to create a similar workflow on Windows.  Not a chance.

I also tried to hot key frequently used menu entries on Windows.
I think I was renaming things like "control panel" as '&C. Control Panel".
I seem to remember that getting close but quickly fell apart.

-- 
Dan Espen

[toc] | [prev] | [next] | [standalone]

#213764

On Sat, 05 Sep 2020 09:26:51 -0400, Dan Espen <dan1espen@gmail.com>
wrote:

>Niklas Karlsson <anksil@yahoo.se> writes:
>
>> On 2020-09-05, Bob Eager <news0073@eager.cx> wrote:
>>>
>>> (I have to say that his idea of an X terminal's usefulness was how many 
>>> command prompt windows it could display).
>>
>> That does seem to be how most hardcore *nix geeks use X.
>
>Where I worked, we mostly used Sun Workstations but we had a few X
>terminals.  The X terminals were just as useful as the workstations.
>(You could go to any number of other machines for computing.)
>I got the impression the X terminals weren't cheap enough to catch on.

This has been the problem with GUI terminals right along.  A terminal
that can display a GUI has all the pieces to just be a computer so why
not just put an OS on it and use it as such?

And before anybody says "Chrombook", ChromeOS is just Linux that
somebody shot in the leg.

>> I'm not really a fan of X, especially what it's become today with a
>> bunch of inconsistent window managers and other systems, umpteen
>> different ways copy/paste works depending on which window you're in,
>> etc. Perhaps Wayland is an improvement somehow - I've not really used it
>> yet.
>
>As a big fan of Fvwm and a past X windows networked user I think Wayland
>is a bad joke.  A typical case of someone wanting to blaze a new path
>rather than enhancing what's already working.
>
>I watched the Wayland demos and every one of them talked about
>"tearing".  I've got years of using X windows, I have no idea what
>tearing they're talking about.  Likewise I thought their arguments about
>no one needing networking were (well I don't need it so you don't
>either).
>
>So, not a fan of Wayland.  Fortunately, from what I'm reading,
>Wayland has yet to show it's superiority.  I wish it would just die and
>go away.
>
>As for consistency, I notice all the mainstream desktop managers
>change with every release.  My WM, Fvwm only changes when I change
>my configuration file.  Fvwm gets new features, it just doesn't force
>change with every release.  My desktop is pretty much the same as it
>was 20 years ago.  That's consistent.
>
>> Windows 10 and macOS have their flaws, to be sure, but their GUIs are
>> pretty agreeable these days. And with PuTTY (or something fancier that
>> keeps an easily accessible database of machines you connect to) or ssh
>> in Terminal.app, you might as well be in *nix anyhow. And of course, in
>> the case of macOS, you actually ARE.
>
>Back when I had to connect to many different machines at work
>I had a GUI I wrote hooked up to my Fvwm configuration.
>If I had a new server say "treerat", I'd go into my GUI and pick a
>shortcut letter for the server, probably "t" if it wasn't already used.
>I'd enter the "t", the server host name and some arguments to the
>terminal program (for example a background color pattern).
>
>The command line to start the terminal used the host name as a -name
>argument to the terminal program.  That made it possible for Fvwm
>to tell one terminal from another.
>
>Then I'd be able to type "shift t" to launch a new terminal on that
>host, or "meta t" to do the same thing iconified, or just "t" to warp to
>an existing terminal on treerat.
>
>That made it possible to access all my servers with a single root window
>keystroke.
>
>The GUI also built an Fvwm menu for server access so if I didn't want
>to devote a hot key I could still get to the server.
>
>I tried to create a similar workflow on Windows.  Not a chance.
>
>I also tried to hot key frequently used menu entries on Windows.
>I think I was renaming things like "control panel" as '&C. Control Panel".
>I seem to remember that getting close but quickly fell apart.

Interesting question.  Never tried to do anything like that with
Windows--might involve writing some real code.

[toc] | [prev] | [next] | [standalone]

#213766

J. Clarke <jclarke.873638@gmail.com> writes:

> On Sat, 05 Sep 2020 09:26:51 -0400, Dan Espen <dan1espen@gmail.com>
> wrote:
>
>>Niklas Karlsson <anksil@yahoo.se> writes:
>>
>>> On 2020-09-05, Bob Eager <news0073@eager.cx> wrote:
>>>>
>>>> (I have to say that his idea of an X terminal's usefulness was how many 
>>>> command prompt windows it could display).
>>>
>>> That does seem to be how most hardcore *nix geeks use X.
>>
>>Where I worked, we mostly used Sun Workstations but we had a few X
>>terminals.  The X terminals were just as useful as the workstations.
>>(You could go to any number of other machines for computing.)
>>I got the impression the X terminals weren't cheap enough to catch on.
>
> This has been the problem with GUI terminals right along.  A terminal
> that can display a GUI has all the pieces to just be a computer so why
> not just put an OS on it and use it as such?
>
> And before anybody says "Chrombook", ChromeOS is just Linux that
> somebody shot in the leg.

Yes an X terminal is significantly simpler than a full blown PC
but it still has to have a lot of the same parts.

>>> I'm not really a fan of X, especially what it's become today with a
>>> bunch of inconsistent window managers and other systems, umpteen
>>> different ways copy/paste works depending on which window you're in,
>>> etc. Perhaps Wayland is an improvement somehow - I've not really used it
>>> yet.
>>
>>As a big fan of Fvwm and a past X windows networked user I think Wayland
>>is a bad joke.  A typical case of someone wanting to blaze a new path
>>rather than enhancing what's already working.
>>
>>I watched the Wayland demos and every one of them talked about
>>"tearing".  I've got years of using X windows, I have no idea what
>>tearing they're talking about.  Likewise I thought their arguments about
>>no one needing networking were (well I don't need it so you don't
>>either).
>>
>>So, not a fan of Wayland.  Fortunately, from what I'm reading,
>>Wayland has yet to show it's superiority.  I wish it would just die and
>>go away.
>>
>>As for consistency, I notice all the mainstream desktop managers
>>change with every release.  My WM, Fvwm only changes when I change
>>my configuration file.  Fvwm gets new features, it just doesn't force
>>change with every release.  My desktop is pretty much the same as it
>>was 20 years ago.  That's consistent.
>>
>>> Windows 10 and macOS have their flaws, to be sure, but their GUIs are
>>> pretty agreeable these days. And with PuTTY (or something fancier that
>>> keeps an easily accessible database of machines you connect to) or ssh
>>> in Terminal.app, you might as well be in *nix anyhow. And of course, in
>>> the case of macOS, you actually ARE.
>>
>>Back when I had to connect to many different machines at work
>>I had a GUI I wrote hooked up to my Fvwm configuration.
>>If I had a new server say "treerat", I'd go into my GUI and pick a
>>shortcut letter for the server, probably "t" if it wasn't already used.
>>I'd enter the "t", the server host name and some arguments to the
>>terminal program (for example a background color pattern).
>>
>>The command line to start the terminal used the host name as a -name
>>argument to the terminal program.  That made it possible for Fvwm
>>to tell one terminal from another.
>>
>>Then I'd be able to type "shift t" to launch a new terminal on that
>>host, or "meta t" to do the same thing iconified, or just "t" to warp to
>>an existing terminal on treerat.
>>
>>That made it possible to access all my servers with a single root window
>>keystroke.
>>
>>The GUI also built an Fvwm menu for server access so if I didn't want
>>to devote a hot key I could still get to the server.
>>
>>I tried to create a similar workflow on Windows.  Not a chance.
>>
>>I also tried to hot key frequently used menu entries on Windows.
>>I think I was renaming things like "control panel" as '&C. Control Panel".
>>I seem to remember that getting close but quickly fell apart.
>
> Interesting question.  Never tried to do anything like that with
> Windows--might involve writing some real code.

As far as I can tell, the Windows OS owns the root window.
I'm not a windows expert but I couldn't find a way to write my own key bindings
and Windows doesn't expose it for cusomization.

I'm a big fan of root window key bindings.

Typically the root window brings up menus and you navigate the menus
with the arrow keys.  So, I use the arrow keys to launch root window menus.

-- 
Dan Espen

[toc] | [prev] | [next] | [standalone]

#213768

On Sat, 05 Sep 2020 12:26:05 -0400, Dan Espen <dan1espen@gmail.com>
wrote:

>J. Clarke <jclarke.873638@gmail.com> writes:
>
>> On Sat, 05 Sep 2020 09:26:51 -0400, Dan Espen <dan1espen@gmail.com>
>> wrote:
>>
>>>Niklas Karlsson <anksil@yahoo.se> writes:
>>>
>>>> On 2020-09-05, Bob Eager <news0073@eager.cx> wrote:
>>>>>
>>>>> (I have to say that his idea of an X terminal's usefulness was how many 
>>>>> command prompt windows it could display).
>>>>
>>>> That does seem to be how most hardcore *nix geeks use X.
>>>
>>>Where I worked, we mostly used Sun Workstations but we had a few X
>>>terminals.  The X terminals were just as useful as the workstations.
>>>(You could go to any number of other machines for computing.)
>>>I got the impression the X terminals weren't cheap enough to catch on.
>>
>> This has been the problem with GUI terminals right along.  A terminal
>> that can display a GUI has all the pieces to just be a computer so why
>> not just put an OS on it and use it as such?
>>
>> And before anybody says "Chrombook", ChromeOS is just Linux that
>> somebody shot in the leg.
>
>Yes an X terminal is significantly simpler than a full blown PC
>but it still has to have a lot of the same parts.
>
>>>> I'm not really a fan of X, especially what it's become today with a
>>>> bunch of inconsistent window managers and other systems, umpteen
>>>> different ways copy/paste works depending on which window you're in,
>>>> etc. Perhaps Wayland is an improvement somehow - I've not really used it
>>>> yet.
>>>
>>>As a big fan of Fvwm and a past X windows networked user I think Wayland
>>>is a bad joke.  A typical case of someone wanting to blaze a new path
>>>rather than enhancing what's already working.
>>>
>>>I watched the Wayland demos and every one of them talked about
>>>"tearing".  I've got years of using X windows, I have no idea what
>>>tearing they're talking about.  Likewise I thought their arguments about
>>>no one needing networking were (well I don't need it so you don't
>>>either).
>>>
>>>So, not a fan of Wayland.  Fortunately, from what I'm reading,
>>>Wayland has yet to show it's superiority.  I wish it would just die and
>>>go away.
>>>
>>>As for consistency, I notice all the mainstream desktop managers
>>>change with every release.  My WM, Fvwm only changes when I change
>>>my configuration file.  Fvwm gets new features, it just doesn't force
>>>change with every release.  My desktop is pretty much the same as it
>>>was 20 years ago.  That's consistent.
>>>
>>>> Windows 10 and macOS have their flaws, to be sure, but their GUIs are
>>>> pretty agreeable these days. And with PuTTY (or something fancier that
>>>> keeps an easily accessible database of machines you connect to) or ssh
>>>> in Terminal.app, you might as well be in *nix anyhow. And of course, in
>>>> the case of macOS, you actually ARE.
>>>
>>>Back when I had to connect to many different machines at work
>>>I had a GUI I wrote hooked up to my Fvwm configuration.
>>>If I had a new server say "treerat", I'd go into my GUI and pick a
>>>shortcut letter for the server, probably "t" if it wasn't already used.
>>>I'd enter the "t", the server host name and some arguments to the
>>>terminal program (for example a background color pattern).
>>>
>>>The command line to start the terminal used the host name as a -name
>>>argument to the terminal program.  That made it possible for Fvwm
>>>to tell one terminal from another.
>>>
>>>Then I'd be able to type "shift t" to launch a new terminal on that
>>>host, or "meta t" to do the same thing iconified, or just "t" to warp to
>>>an existing terminal on treerat.
>>>
>>>That made it possible to access all my servers with a single root window
>>>keystroke.
>>>
>>>The GUI also built an Fvwm menu for server access so if I didn't want
>>>to devote a hot key I could still get to the server.
>>>
>>>I tried to create a similar workflow on Windows.  Not a chance.
>>>
>>>I also tried to hot key frequently used menu entries on Windows.
>>>I think I was renaming things like "control panel" as '&C. Control Panel".
>>>I seem to remember that getting close but quickly fell apart.
>>
>> Interesting question.  Never tried to do anything like that with
>> Windows--might involve writing some real code.
>
>As far as I can tell, the Windows OS owns the root window.
>I'm not a windows expert but I couldn't find a way to write my own key bindings
>and Windows doesn't expose it for cusomization.
>
>I'm a big fan of root window key bindings.
>
>Typically the root window brings up menus and you navigate the menus
>with the arrow keys.  So, I use the arrow keys to launch root window menus.

The established way to do this in Windows is with third party
software.  "Autohotkey" is the most common.  There are also
brute-force methods--any decent gamer keyboard lets you set up
keyboard macros that are run on the keyboard's internal processor.

[toc] | [prev] | [next] | [standalone]

#213774

J. Clarke <jclarke.873638@gmail.com> writes:

> On Sat, 05 Sep 2020 12:26:05 -0400, Dan Espen <dan1espen@gmail.com>
> wrote:
>
>>J. Clarke <jclarke.873638@gmail.com> writes:
>>
>>> On Sat, 05 Sep 2020 09:26:51 -0400, Dan Espen <dan1espen@gmail.com>
>>> wrote:
>>>
>>>>Niklas Karlsson <anksil@yahoo.se> writes:
>>>>
>>>>> On 2020-09-05, Bob Eager <news0073@eager.cx> wrote:
>>>>>>
>>>>>> (I have to say that his idea of an X terminal's usefulness was how many 
>>>>>> command prompt windows it could display).
>>>>>
>>>>> That does seem to be how most hardcore *nix geeks use X.
>>>>
>>>>Where I worked, we mostly used Sun Workstations but we had a few X
>>>>terminals.  The X terminals were just as useful as the workstations.
>>>>(You could go to any number of other machines for computing.)
>>>>I got the impression the X terminals weren't cheap enough to catch on.
>>>
>>> This has been the problem with GUI terminals right along.  A terminal
>>> that can display a GUI has all the pieces to just be a computer so why
>>> not just put an OS on it and use it as such?
>>>
>>> And before anybody says "Chrombook", ChromeOS is just Linux that
>>> somebody shot in the leg.
>>
>>Yes an X terminal is significantly simpler than a full blown PC
>>but it still has to have a lot of the same parts.
>>
>>>>> I'm not really a fan of X, especially what it's become today with a
>>>>> bunch of inconsistent window managers and other systems, umpteen
>>>>> different ways copy/paste works depending on which window you're in,
>>>>> etc. Perhaps Wayland is an improvement somehow - I've not really used it
>>>>> yet.
>>>>
>>>>As a big fan of Fvwm and a past X windows networked user I think Wayland
>>>>is a bad joke.  A typical case of someone wanting to blaze a new path
>>>>rather than enhancing what's already working.
>>>>
>>>>I watched the Wayland demos and every one of them talked about
>>>>"tearing".  I've got years of using X windows, I have no idea what
>>>>tearing they're talking about.  Likewise I thought their arguments about
>>>>no one needing networking were (well I don't need it so you don't
>>>>either).
>>>>
>>>>So, not a fan of Wayland.  Fortunately, from what I'm reading,
>>>>Wayland has yet to show it's superiority.  I wish it would just die and
>>>>go away.
>>>>
>>>>As for consistency, I notice all the mainstream desktop managers
>>>>change with every release.  My WM, Fvwm only changes when I change
>>>>my configuration file.  Fvwm gets new features, it just doesn't force
>>>>change with every release.  My desktop is pretty much the same as it
>>>>was 20 years ago.  That's consistent.
>>>>
>>>>> Windows 10 and macOS have their flaws, to be sure, but their GUIs are
>>>>> pretty agreeable these days. And with PuTTY (or something fancier that
>>>>> keeps an easily accessible database of machines you connect to) or ssh
>>>>> in Terminal.app, you might as well be in *nix anyhow. And of course, in
>>>>> the case of macOS, you actually ARE.
>>>>
>>>>Back when I had to connect to many different machines at work
>>>>I had a GUI I wrote hooked up to my Fvwm configuration.
>>>>If I had a new server say "treerat", I'd go into my GUI and pick a
>>>>shortcut letter for the server, probably "t" if it wasn't already used.
>>>>I'd enter the "t", the server host name and some arguments to the
>>>>terminal program (for example a background color pattern).
>>>>
>>>>The command line to start the terminal used the host name as a -name
>>>>argument to the terminal program.  That made it possible for Fvwm
>>>>to tell one terminal from another.
>>>>
>>>>Then I'd be able to type "shift t" to launch a new terminal on that
>>>>host, or "meta t" to do the same thing iconified, or just "t" to warp to
>>>>an existing terminal on treerat.
>>>>
>>>>That made it possible to access all my servers with a single root window
>>>>keystroke.
>>>>
>>>>The GUI also built an Fvwm menu for server access so if I didn't want
>>>>to devote a hot key I could still get to the server.
>>>>
>>>>I tried to create a similar workflow on Windows.  Not a chance.
>>>>
>>>>I also tried to hot key frequently used menu entries on Windows.
>>>>I think I was renaming things like "control panel" as '&C. Control Panel".
>>>>I seem to remember that getting close but quickly fell apart.
>>>
>>> Interesting question.  Never tried to do anything like that with
>>> Windows--might involve writing some real code.
>>
>>As far as I can tell, the Windows OS owns the root window.
>>I'm not a windows expert but I couldn't find a way to write my own key bindings
>>and Windows doesn't expose it for cusomization.
>>
>>I'm a big fan of root window key bindings.
>>
>>Typically the root window brings up menus and you navigate the menus
>>with the arrow keys.  So, I use the arrow keys to launch root window menus.
>
> The established way to do this in Windows is with third party
> software.  "Autohotkey" is the most common.  There are also
> brute-force methods--any decent gamer keyboard lets you set up
> keyboard macros that are run on the keyboard's internal processor.

I see, both the keyboard solution and autohotkeys are a macro
approach.

As I said, I had a bunch of terminals only distinguished by the host
they were logged into and I could warp to them by the host they were logged into.

I wonder what kind of macro would know how to do that?

In X windows, each window is assigned a name class and resource which
external programs and the window manager can use to access the window.
The command line to start the terminal lets you assign a name.

I've been looking at replacing my keyboard.  It's a mechanical model
with backlit keys.  It still works but some of the lights are no longer
working and the manufacturer bit the dust so I can no longer
buy their LEDs.

I was looking at a neat Logitech model but the pandemic hit and they
pulled it from production.  There are lots of LED models but most of
them have serious light leakage around the outside of the key.

The ones I'm considering now all have RGB, any light on the keyboard can
be any color.  Sort of nuts but there are Linux drivers for them.
Many of them support macros, but I don't really need that.

-- 
Dan Espen

[toc] | [prev] | [next] | [standalone]

#213782

On Sat, 05 Sep 2020 14:03:38 -0400, Dan Espen <dan1espen@gmail.com>
wrote:

>J. Clarke <jclarke.873638@gmail.com> writes:
>
>> On Sat, 05 Sep 2020 12:26:05 -0400, Dan Espen <dan1espen@gmail.com>
>> wrote:
>>
>>>J. Clarke <jclarke.873638@gmail.com> writes:
>>>
>>>> On Sat, 05 Sep 2020 09:26:51 -0400, Dan Espen <dan1espen@gmail.com>
>>>> wrote:
>>>>
>>>>>Niklas Karlsson <anksil@yahoo.se> writes:
>>>>>
>>>>>> On 2020-09-05, Bob Eager <news0073@eager.cx> wrote:
>>>>>>>
>>>>>>> (I have to say that his idea of an X terminal's usefulness was how many 
>>>>>>> command prompt windows it could display).
>>>>>>
>>>>>> That does seem to be how most hardcore *nix geeks use X.
>>>>>
>>>>>Where I worked, we mostly used Sun Workstations but we had a few X
>>>>>terminals.  The X terminals were just as useful as the workstations.
>>>>>(You could go to any number of other machines for computing.)
>>>>>I got the impression the X terminals weren't cheap enough to catch on.
>>>>
>>>> This has been the problem with GUI terminals right along.  A terminal
>>>> that can display a GUI has all the pieces to just be a computer so why
>>>> not just put an OS on it and use it as such?
>>>>
>>>> And before anybody says "Chrombook", ChromeOS is just Linux that
>>>> somebody shot in the leg.
>>>
>>>Yes an X terminal is significantly simpler than a full blown PC
>>>but it still has to have a lot of the same parts.
>>>
>>>>>> I'm not really a fan of X, especially what it's become today with a
>>>>>> bunch of inconsistent window managers and other systems, umpteen
>>>>>> different ways copy/paste works depending on which window you're in,
>>>>>> etc. Perhaps Wayland is an improvement somehow - I've not really used it
>>>>>> yet.
>>>>>
>>>>>As a big fan of Fvwm and a past X windows networked user I think Wayland
>>>>>is a bad joke.  A typical case of someone wanting to blaze a new path
>>>>>rather than enhancing what's already working.
>>>>>
>>>>>I watched the Wayland demos and every one of them talked about
>>>>>"tearing".  I've got years of using X windows, I have no idea what
>>>>>tearing they're talking about.  Likewise I thought their arguments about
>>>>>no one needing networking were (well I don't need it so you don't
>>>>>either).
>>>>>
>>>>>So, not a fan of Wayland.  Fortunately, from what I'm reading,
>>>>>Wayland has yet to show it's superiority.  I wish it would just die and
>>>>>go away.
>>>>>
>>>>>As for consistency, I notice all the mainstream desktop managers
>>>>>change with every release.  My WM, Fvwm only changes when I change
>>>>>my configuration file.  Fvwm gets new features, it just doesn't force
>>>>>change with every release.  My desktop is pretty much the same as it
>>>>>was 20 years ago.  That's consistent.
>>>>>
>>>>>> Windows 10 and macOS have their flaws, to be sure, but their GUIs are
>>>>>> pretty agreeable these days. And with PuTTY (or something fancier that
>>>>>> keeps an easily accessible database of machines you connect to) or ssh
>>>>>> in Terminal.app, you might as well be in *nix anyhow. And of course, in
>>>>>> the case of macOS, you actually ARE.
>>>>>
>>>>>Back when I had to connect to many different machines at work
>>>>>I had a GUI I wrote hooked up to my Fvwm configuration.
>>>>>If I had a new server say "treerat", I'd go into my GUI and pick a
>>>>>shortcut letter for the server, probably "t" if it wasn't already used.
>>>>>I'd enter the "t", the server host name and some arguments to the
>>>>>terminal program (for example a background color pattern).
>>>>>
>>>>>The command line to start the terminal used the host name as a -name
>>>>>argument to the terminal program.  That made it possible for Fvwm
>>>>>to tell one terminal from another.
>>>>>
>>>>>Then I'd be able to type "shift t" to launch a new terminal on that
>>>>>host, or "meta t" to do the same thing iconified, or just "t" to warp to
>>>>>an existing terminal on treerat.
>>>>>
>>>>>That made it possible to access all my servers with a single root window
>>>>>keystroke.
>>>>>
>>>>>The GUI also built an Fvwm menu for server access so if I didn't want
>>>>>to devote a hot key I could still get to the server.
>>>>>
>>>>>I tried to create a similar workflow on Windows.  Not a chance.
>>>>>
>>>>>I also tried to hot key frequently used menu entries on Windows.
>>>>>I think I was renaming things like "control panel" as '&C. Control Panel".
>>>>>I seem to remember that getting close but quickly fell apart.
>>>>
>>>> Interesting question.  Never tried to do anything like that with
>>>> Windows--might involve writing some real code.
>>>
>>>As far as I can tell, the Windows OS owns the root window.
>>>I'm not a windows expert but I couldn't find a way to write my own key bindings
>>>and Windows doesn't expose it for cusomization.
>>>
>>>I'm a big fan of root window key bindings.
>>>
>>>Typically the root window brings up menus and you navigate the menus
>>>with the arrow keys.  So, I use the arrow keys to launch root window menus.
>>
>> The established way to do this in Windows is with third party
>> software.  "Autohotkey" is the most common.  There are also
>> brute-force methods--any decent gamer keyboard lets you set up
>> keyboard macros that are run on the keyboard's internal processor.
>
>I see, both the keyboard solution and autohotkeys are a macro
>approach.
>
>As I said, I had a bunch of terminals only distinguished by the host
>they were logged into and I could warp to them by the host they were logged into.
>
>I wonder what kind of macro would know how to do that?
>
>In X windows, each window is assigned a name class and resource which
>external programs and the window manager can use to access the window.
>The command line to start the terminal lets you assign a name.
>
>I've been looking at replacing my keyboard.  It's a mechanical model
>with backlit keys.  It still works but some of the lights are no longer
>working and the manufacturer bit the dust so I can no longer
>buy their LEDs.
>
>I was looking at a neat Logitech model but the pandemic hit and they
>pulled it from production.  There are lots of LED models but most of
>them have serious light leakage around the outside of the key.
>
>The ones I'm considering now all have RGB, any light on the keyboard can
>be any color.  Sort of nuts but there are Linux drivers for them.
>Many of them support macros, but I don't really need that.

Right now, I'm finding that I don't have any of my systems set up for
remote access so I can't experiment until I do some reconfiguring. But
AutoHotkey does have a provision to activate a window by window
handle.

With regard to keyboards, my three favorites are all Logitech.  The
G910 is the one I use most of the time at home (I use a different one
at work but it's company-provided and the best thing I can say about
it is that it works).  Before the G910 I had a K800--enough keys on it
finally lost the ability to retain a keycap that I replaced it with
the 910.  I have a G915 which is a wonderful keyboard if you're
working on a desk or tabletop but the lack of any kind of rest for the
heel of the hand makes it less comfortable for use in one's lap, which
is now my normal working mode.  All are backlit.  The 800 is white
backlit, the other two are RGB.  The 800 and 915 are wireless--the 800
needs a dongle, the 915 has a dongle but it's run what you brung--it
can connect using the dongle, Bluetooth, or I believe there's a way to
use it with wifi.  They should all work with Linux, but making the 910
and 915 do tricks may require a specialized driver.

[toc] | [prev] | [next] | [standalone]

#213794

J. Clarke <jclarke.873638@gmail.com> writes:

> On Sat, 05 Sep 2020 14:03:38 -0400, Dan Espen <dan1espen@gmail.com>
> wrote:
>
>>J. Clarke <jclarke.873638@gmail.com> writes:
>>
>>> On Sat, 05 Sep 2020 12:26:05 -0400, Dan Espen <dan1espen@gmail.com>
>>> wrote:
>>>
>>>>J. Clarke <jclarke.873638@gmail.com> writes:
>>>>
>>>>> On Sat, 05 Sep 2020 09:26:51 -0400, Dan Espen <dan1espen@gmail.com>
>>>>> wrote:
>>>>>
>>>>>>Niklas Karlsson <anksil@yahoo.se> writes:
>>>>>>
>>>>>>> On 2020-09-05, Bob Eager <news0073@eager.cx> wrote:
>>>>>>>>
>>>>>>>> (I have to say that his idea of an X terminal's usefulness was how many 
>>>>>>>> command prompt windows it could display).
>>>>>>>
>>>>>>> That does seem to be how most hardcore *nix geeks use X.
>>>>>>
>>>>>>Where I worked, we mostly used Sun Workstations but we had a few X
>>>>>>terminals.  The X terminals were just as useful as the workstations.
>>>>>>(You could go to any number of other machines for computing.)
>>>>>>I got the impression the X terminals weren't cheap enough to catch on.
>>>>>
>>>>> This has been the problem with GUI terminals right along.  A terminal
>>>>> that can display a GUI has all the pieces to just be a computer so why
>>>>> not just put an OS on it and use it as such?
>>>>>
>>>>> And before anybody says "Chrombook", ChromeOS is just Linux that
>>>>> somebody shot in the leg.
>>>>
>>>>Yes an X terminal is significantly simpler than a full blown PC
>>>>but it still has to have a lot of the same parts.
>>>>
>>>>>>> I'm not really a fan of X, especially what it's become today with a
>>>>>>> bunch of inconsistent window managers and other systems, umpteen
>>>>>>> different ways copy/paste works depending on which window you're in,
>>>>>>> etc. Perhaps Wayland is an improvement somehow - I've not really used it
>>>>>>> yet.
>>>>>>
>>>>>>As a big fan of Fvwm and a past X windows networked user I think Wayland
>>>>>>is a bad joke.  A typical case of someone wanting to blaze a new path
>>>>>>rather than enhancing what's already working.
>>>>>>
>>>>>>I watched the Wayland demos and every one of them talked about
>>>>>>"tearing".  I've got years of using X windows, I have no idea what
>>>>>>tearing they're talking about.  Likewise I thought their arguments about
>>>>>>no one needing networking were (well I don't need it so you don't
>>>>>>either).
>>>>>>
>>>>>>So, not a fan of Wayland.  Fortunately, from what I'm reading,
>>>>>>Wayland has yet to show it's superiority.  I wish it would just die and
>>>>>>go away.
>>>>>>
>>>>>>As for consistency, I notice all the mainstream desktop managers
>>>>>>change with every release.  My WM, Fvwm only changes when I change
>>>>>>my configuration file.  Fvwm gets new features, it just doesn't force
>>>>>>change with every release.  My desktop is pretty much the same as it
>>>>>>was 20 years ago.  That's consistent.
>>>>>>
>>>>>>> Windows 10 and macOS have their flaws, to be sure, but their GUIs are
>>>>>>> pretty agreeable these days. And with PuTTY (or something fancier that
>>>>>>> keeps an easily accessible database of machines you connect to) or ssh
>>>>>>> in Terminal.app, you might as well be in *nix anyhow. And of course, in
>>>>>>> the case of macOS, you actually ARE.
>>>>>>
>>>>>>Back when I had to connect to many different machines at work
>>>>>>I had a GUI I wrote hooked up to my Fvwm configuration.
>>>>>>If I had a new server say "treerat", I'd go into my GUI and pick a
>>>>>>shortcut letter for the server, probably "t" if it wasn't already used.
>>>>>>I'd enter the "t", the server host name and some arguments to the
>>>>>>terminal program (for example a background color pattern).
>>>>>>
>>>>>>The command line to start the terminal used the host name as a -name
>>>>>>argument to the terminal program.  That made it possible for Fvwm
>>>>>>to tell one terminal from another.
>>>>>>
>>>>>>Then I'd be able to type "shift t" to launch a new terminal on that
>>>>>>host, or "meta t" to do the same thing iconified, or just "t" to warp to
>>>>>>an existing terminal on treerat.
>>>>>>
>>>>>>That made it possible to access all my servers with a single root window
>>>>>>keystroke.
>>>>>>
>>>>>>The GUI also built an Fvwm menu for server access so if I didn't want
>>>>>>to devote a hot key I could still get to the server.
>>>>>>
>>>>>>I tried to create a similar workflow on Windows.  Not a chance.
>>>>>>
>>>>>>I also tried to hot key frequently used menu entries on Windows.
>>>>>>I think I was renaming things like "control panel" as '&C. Control Panel".
>>>>>>I seem to remember that getting close but quickly fell apart.
>>>>>
>>>>> Interesting question.  Never tried to do anything like that with
>>>>> Windows--might involve writing some real code.
>>>>
>>>>As far as I can tell, the Windows OS owns the root window.
>>>>I'm not a windows expert but I couldn't find a way to write my own key bindings
>>>>and Windows doesn't expose it for cusomization.
>>>>
>>>>I'm a big fan of root window key bindings.
>>>>
>>>>Typically the root window brings up menus and you navigate the menus
>>>>with the arrow keys.  So, I use the arrow keys to launch root window menus.
>>>
>>> The established way to do this in Windows is with third party
>>> software.  "Autohotkey" is the most common.  There are also
>>> brute-force methods--any decent gamer keyboard lets you set up
>>> keyboard macros that are run on the keyboard's internal processor.
>>
>>I see, both the keyboard solution and autohotkeys are a macro
>>approach.
>>
>>As I said, I had a bunch of terminals only distinguished by the host
>>they were logged into and I could warp to them by the host they were logged into.
>>
>>I wonder what kind of macro would know how to do that?
>>
>>In X windows, each window is assigned a name class and resource which
>>external programs and the window manager can use to access the window.
>>The command line to start the terminal lets you assign a name.
>>
>>I've been looking at replacing my keyboard.  It's a mechanical model
>>with backlit keys.  It still works but some of the lights are no longer
>>working and the manufacturer bit the dust so I can no longer
>>buy their LEDs.
>>
>>I was looking at a neat Logitech model but the pandemic hit and they
>>pulled it from production.  There are lots of LED models but most of
>>them have serious light leakage around the outside of the key.
>>
>>The ones I'm considering now all have RGB, any light on the keyboard can
>>be any color.  Sort of nuts but there are Linux drivers for them.
>>Many of them support macros, but I don't really need that.
>
> Right now, I'm finding that I don't have any of my systems set up for
> remote access so I can't experiment until I do some reconfiguring. But
> AutoHotkey does have a provision to activate a window by window
> handle.
>
> With regard to keyboards, my three favorites are all Logitech.  The
> G910 is the one I use most of the time at home (I use a different one
> at work but it's company-provided and the best thing I can say about
> it is that it works).  Before the G910 I had a K800--enough keys on it
> finally lost the ability to retain a keycap that I replaced it with
> the 910.  I have a G915 which is a wonderful keyboard if you're
> working on a desk or tabletop but the lack of any kind of rest for the
> heel of the hand makes it less comfortable for use in one's lap, which
> is now my normal working mode.  All are backlit.  The 800 is white
> backlit, the other two are RGB.  The 800 and 915 are wireless--the 800
> needs a dongle, the 915 has a dongle but it's run what you brung--it
> can connect using the dongle, Bluetooth, or I believe there's a way to
> use it with wifi.  They should all work with Linux, but making the 910
> and 915 do tricks may require a specialized driver.

I was looking at the 810 before they pulled it from production.
I even had it on order.  First they said it would be another month,
then they canceled.  Amazon says they have 2 in stock at $300 apiece
but that's where I had the order first accepted then canceled.

Refurbs are cheaper but I think I'll find something better.

Isn't the 910 the one with no scroll lock LED?
I remember not liking they way it looked much but I was considering it.
Then I think I saw no scroll lock LED.  (I have software that lights
the scroll lock key when I have unread email.  I could have programmed
another key to change color but the missing key annoyed me.)

-- 
Dan Espen

[toc] | [prev] | [next] | [standalone]

#213800

On Sat, 05 Sep 2020 16:34:06 -0400, Dan Espen <dan1espen@gmail.com>
wrote:

>J. Clarke <jclarke.873638@gmail.com> writes:
>
>> On Sat, 05 Sep 2020 14:03:38 -0400, Dan Espen <dan1espen@gmail.com>
>> wrote:
>>
>>>J. Clarke <jclarke.873638@gmail.com> writes:
>>>
>>>> On Sat, 05 Sep 2020 12:26:05 -0400, Dan Espen <dan1espen@gmail.com>
>>>> wrote:
>>>>
>>>>>J. Clarke <jclarke.873638@gmail.com> writes:
>>>>>
>>>>>> On Sat, 05 Sep 2020 09:26:51 -0400, Dan Espen <dan1espen@gmail.com>
>>>>>> wrote:
>>>>>>
>>>>>>>Niklas Karlsson <anksil@yahoo.se> writes:
>>>>>>>
>>>>>>>> On 2020-09-05, Bob Eager <news0073@eager.cx> wrote:
>>>>>>>>>
>>>>>>>>> (I have to say that his idea of an X terminal's usefulness was how many 
>>>>>>>>> command prompt windows it could display).
>>>>>>>>
>>>>>>>> That does seem to be how most hardcore *nix geeks use X.
>>>>>>>
>>>>>>>Where I worked, we mostly used Sun Workstations but we had a few X
>>>>>>>terminals.  The X terminals were just as useful as the workstations.
>>>>>>>(You could go to any number of other machines for computing.)
>>>>>>>I got the impression the X terminals weren't cheap enough to catch on.
>>>>>>
>>>>>> This has been the problem with GUI terminals right along.  A terminal
>>>>>> that can display a GUI has all the pieces to just be a computer so why
>>>>>> not just put an OS on it and use it as such?
>>>>>>
>>>>>> And before anybody says "Chrombook", ChromeOS is just Linux that
>>>>>> somebody shot in the leg.
>>>>>
>>>>>Yes an X terminal is significantly simpler than a full blown PC
>>>>>but it still has to have a lot of the same parts.
>>>>>
>>>>>>>> I'm not really a fan of X, especially what it's become today with a
>>>>>>>> bunch of inconsistent window managers and other systems, umpteen
>>>>>>>> different ways copy/paste works depending on which window you're in,
>>>>>>>> etc. Perhaps Wayland is an improvement somehow - I've not really used it
>>>>>>>> yet.
>>>>>>>
>>>>>>>As a big fan of Fvwm and a past X windows networked user I think Wayland
>>>>>>>is a bad joke.  A typical case of someone wanting to blaze a new path
>>>>>>>rather than enhancing what's already working.
>>>>>>>
>>>>>>>I watched the Wayland demos and every one of them talked about
>>>>>>>"tearing".  I've got years of using X windows, I have no idea what
>>>>>>>tearing they're talking about.  Likewise I thought their arguments about
>>>>>>>no one needing networking were (well I don't need it so you don't
>>>>>>>either).
>>>>>>>
>>>>>>>So, not a fan of Wayland.  Fortunately, from what I'm reading,
>>>>>>>Wayland has yet to show it's superiority.  I wish it would just die and
>>>>>>>go away.
>>>>>>>
>>>>>>>As for consistency, I notice all the mainstream desktop managers
>>>>>>>change with every release.  My WM, Fvwm only changes when I change
>>>>>>>my configuration file.  Fvwm gets new features, it just doesn't force
>>>>>>>change with every release.  My desktop is pretty much the same as it
>>>>>>>was 20 years ago.  That's consistent.
>>>>>>>
>>>>>>>> Windows 10 and macOS have their flaws, to be sure, but their GUIs are
>>>>>>>> pretty agreeable these days. And with PuTTY (or something fancier that
>>>>>>>> keeps an easily accessible database of machines you connect to) or ssh
>>>>>>>> in Terminal.app, you might as well be in *nix anyhow. And of course, in
>>>>>>>> the case of macOS, you actually ARE.
>>>>>>>
>>>>>>>Back when I had to connect to many different machines at work
>>>>>>>I had a GUI I wrote hooked up to my Fvwm configuration.
>>>>>>>If I had a new server say "treerat", I'd go into my GUI and pick a
>>>>>>>shortcut letter for the server, probably "t" if it wasn't already used.
>>>>>>>I'd enter the "t", the server host name and some arguments to the
>>>>>>>terminal program (for example a background color pattern).
>>>>>>>
>>>>>>>The command line to start the terminal used the host name as a -name
>>>>>>>argument to the terminal program.  That made it possible for Fvwm
>>>>>>>to tell one terminal from another.
>>>>>>>
>>>>>>>Then I'd be able to type "shift t" to launch a new terminal on that
>>>>>>>host, or "meta t" to do the same thing iconified, or just "t" to warp to
>>>>>>>an existing terminal on treerat.
>>>>>>>
>>>>>>>That made it possible to access all my servers with a single root window
>>>>>>>keystroke.
>>>>>>>
>>>>>>>The GUI also built an Fvwm menu for server access so if I didn't want
>>>>>>>to devote a hot key I could still get to the server.
>>>>>>>
>>>>>>>I tried to create a similar workflow on Windows.  Not a chance.
>>>>>>>
>>>>>>>I also tried to hot key frequently used menu entries on Windows.
>>>>>>>I think I was renaming things like "control panel" as '&C. Control Panel".
>>>>>>>I seem to remember that getting close but quickly fell apart.
>>>>>>
>>>>>> Interesting question.  Never tried to do anything like that with
>>>>>> Windows--might involve writing some real code.
>>>>>
>>>>>As far as I can tell, the Windows OS owns the root window.
>>>>>I'm not a windows expert but I couldn't find a way to write my own key bindings
>>>>>and Windows doesn't expose it for cusomization.
>>>>>
>>>>>I'm a big fan of root window key bindings.
>>>>>
>>>>>Typically the root window brings up menus and you navigate the menus
>>>>>with the arrow keys.  So, I use the arrow keys to launch root window menus.
>>>>
>>>> The established way to do this in Windows is with third party
>>>> software.  "Autohotkey" is the most common.  There are also
>>>> brute-force methods--any decent gamer keyboard lets you set up
>>>> keyboard macros that are run on the keyboard's internal processor.
>>>
>>>I see, both the keyboard solution and autohotkeys are a macro
>>>approach.
>>>
>>>As I said, I had a bunch of terminals only distinguished by the host
>>>they were logged into and I could warp to them by the host they were logged into.
>>>
>>>I wonder what kind of macro would know how to do that?
>>>
>>>In X windows, each window is assigned a name class and resource which
>>>external programs and the window manager can use to access the window.
>>>The command line to start the terminal lets you assign a name.
>>>
>>>I've been looking at replacing my keyboard.  It's a mechanical model
>>>with backlit keys.  It still works but some of the lights are no longer
>>>working and the manufacturer bit the dust so I can no longer
>>>buy their LEDs.
>>>
>>>I was looking at a neat Logitech model but the pandemic hit and they
>>>pulled it from production.  There are lots of LED models but most of
>>>them have serious light leakage around the outside of the key.
>>>
>>>The ones I'm considering now all have RGB, any light on the keyboard can
>>>be any color.  Sort of nuts but there are Linux drivers for them.
>>>Many of them support macros, but I don't really need that.
>>
>> Right now, I'm finding that I don't have any of my systems set up for
>> remote access so I can't experiment until I do some reconfiguring. But
>> AutoHotkey does have a provision to activate a window by window
>> handle.
>>
>> With regard to keyboards, my three favorites are all Logitech.  The
>> G910 is the one I use most of the time at home (I use a different one
>> at work but it's company-provided and the best thing I can say about
>> it is that it works).  Before the G910 I had a K800--enough keys on it
>> finally lost the ability to retain a keycap that I replaced it with
>> the 910.  I have a G915 which is a wonderful keyboard if you're
>> working on a desk or tabletop but the lack of any kind of rest for the
>> heel of the hand makes it less comfortable for use in one's lap, which
>> is now my normal working mode.  All are backlit.  The 800 is white
>> backlit, the other two are RGB.  The 800 and 915 are wireless--the 800
>> needs a dongle, the 915 has a dongle but it's run what you brung--it
>> can connect using the dongle, Bluetooth, or I believe there's a way to
>> use it with wifi.  They should all work with Linux, but making the 910
>> and 915 do tricks may require a specialized driver.
>
>I was looking at the 810 before they pulled it from production.
>I even had it on order.  First they said it would be another month,
>then they canceled.  Amazon says they have 2 in stock at $300 apiece
>but that's where I had the order first accepted then canceled.
>
>Refurbs are cheaper but I think I'll find something better.
>
>Isn't the 910 the one with no scroll lock LED?

There's definitely a scroll lock LED.

>I remember not liking they way it looked much but I was considering it.
>Then I think I saw no scroll lock LED.  (I have software that lights
>the scroll lock key when I have unread email.  I could have programmed
>another key to change color but the missing key annoyed me.)

[toc] | [prev] | [next] | [standalone]

#213802

J. Clarke <jclarke.873638@gmail.com> writes:

> On Sat, 05 Sep 2020 16:34:06 -0400, Dan Espen <dan1espen@gmail.com>
> wrote:
>>Isn't the 910 the one with no scroll lock LED?
>
> There's definitely a scroll lock LED.


Ah, I see them now.  Not in the regular place but definitely there.

I wasn't too happy with the WASD markings but looking again the
the G1-G5 keys would be perfect for those custom keys I want to use.
Just where Sun put them.

Thanks again.
You sure seem to know your PC hardware.

-- 
Dan Espen

[toc] | [prev] | [next] | [standalone]

#213810

On Sat, 05 Sep 2020 17:27:12 -0400, Dan Espen <dan1espen@gmail.com>
wrote:

>J. Clarke <jclarke.873638@gmail.com> writes:
>
>> On Sat, 05 Sep 2020 16:34:06 -0400, Dan Espen <dan1espen@gmail.com>
>> wrote:
>>>Isn't the 910 the one with no scroll lock LED?
>>
>> There's definitely a scroll lock LED.
>
>
>Ah, I see them now.  Not in the regular place but definitely there.
>
>I wasn't too happy with the WASD markings but looking again the
>the G1-G5 keys would be perfect for those custom keys I want to use.
>Just where Sun put them.

And they're programmable.  You might have to program them on a windows
box but the program should stay.

>Thanks again.
>You sure seem to know your PC hardware.

Well, thank you.  I've been building PCs for fun and profit for about
40 years now.  More for fun lately than for profit.  

I was thinking about another one this year, but after being annoyed
one too many times by no usable signal on my cheap 5 year old Chinese
phone (going to be a real problem shortly--they're changing the
authentication system at work so that you have to have a working
cellular connection) I decided to drop the hammer on a high end
Samsung to see if paying that kind of money for a phone actually makes
a difference.  8 cores, 3 GHz, 12 GB RAM, half an effing TERABYTE of
solid state disk (and another half on a microSD card smaller than the
nail on my pinkie finger), more money than I paid for a (used) Lincoln
Continental in 1969, and it's a damned _phone_.  Which is big enough
to have a decent-sized antenna inside.  I know it's an extravagance
but if I'm paying that kind of money anyway, I may as well get _all_
the goodies.

I'm told that von Neumann one chewed out a grad student for wasting
expensive CPU cycles having the machine do things (like compiling
code) that a human could do.  I wonder what he would say about such
beast?

[toc] | [prev] | [next] | [standalone]

#213823

J. Clarke <jclarke.873638@gmail.com> writes:

> On Sat, 05 Sep 2020 17:27:12 -0400, Dan Espen <dan1espen@gmail.com>
> wrote:
>
>>J. Clarke <jclarke.873638@gmail.com> writes:
>>
>>> On Sat, 05 Sep 2020 16:34:06 -0400, Dan Espen <dan1espen@gmail.com>
>>> wrote:
>>>>Isn't the 910 the one with no scroll lock LED?
>>>
>>> There's definitely a scroll lock LED.
>>
>>
>>Ah, I see them now.  Not in the regular place but definitely there.
>>
>>I wasn't too happy with the WASD markings but looking again the
>>the G1-G5 keys would be perfect for those custom keys I want to use.
>>Just where Sun put them.
>
> And they're programmable.  You might have to program them on a windows
> box but the program should stay.

I just checked, looks like I can do almost everything with Linux:

https://github.com/MohamadSaada/LogiGSK

and this package does the G keys:

https://github.com/JSubelj/g910-gkey-macro-support

But I don't want the G keys to emit strings, just a unique key code.
I can get Linux/fvwm/Emacs to do the rest.
I think I can figure it out.


-- 
Dan Espen

[toc] | [prev] | [next] | [standalone]

#213771

On 9/5/2020 11:26 AM, Dan Espen wrote:
> J. Clarke <jclarke.873638@gmail.com> writes:

>>>
>>> I also tried to hot key frequently used menu entries on Windows.
>>> I think I was renaming things like "control panel" as '&C. Control Panel".
>>> I seem to remember that getting close but quickly fell apart.
>>
>> Interesting question.  Never tried to do anything like that with
>> Windows--might involve writing some real code.
> 
> As far as I can tell, the Windows OS owns the root window.
> I'm not a windows expert but I couldn't find a way to write my own key bindings
> and Windows doesn't expose it for cusomization.
> 
> I'm a big fan of root window key bindings.
> 
> Typically the root window brings up menus and you navigate the menus
> with the arrow keys.  So, I use the arrow keys to launch root window menus.
> 

As a user, there are of course various hotkey/scripting programs for 
Windows (I use AutoHotKey, it's pretty slick, GPL2 and the source is 
online). I realize that's not getting down in the weeds and doing it 
yourself.

[toc] | [prev] | [next] | [standalone]

#213775

Dave Garland <dave.garland@wizinfo.com> writes:

> On 9/5/2020 11:26 AM, Dan Espen wrote:
>> J. Clarke <jclarke.873638@gmail.com> writes:
>
>>>>
>>>> I also tried to hot key frequently used menu entries on Windows.
>>>> I think I was renaming things like "control panel" as '&C. Control Panel".
>>>> I seem to remember that getting close but quickly fell apart.
>>>
>>> Interesting question.  Never tried to do anything like that with
>>> Windows--might involve writing some real code.
>> As far as I can tell, the Windows OS owns the root window.
>> I'm not a windows expert but I couldn't find a way to write my own key bindings
>> and Windows doesn't expose it for cusomization.
>> I'm a big fan of root window key bindings.
>> Typically the root window brings up menus and you navigate the menus
>> with the arrow keys.  So, I use the arrow keys to launch root window menus.
>
> As a user, there are of course various hotkey/scripting programs for
> Windows (I use AutoHotKey, it's pretty slick, GPL2 and the source is 
> online). I realize that's not getting down in the weeds and doing it
> yourself.

See my other post, not sure if it could do it at all.
Of course I no longer need that since I don't work anymore.
It might cover the stuff I do now.  Not sure.


-- 
Dan Espen

[toc] | [prev] | [next] | [standalone]

#213817

J. Clarke <jclarke.873638@gmail.com> wrote:
> On Sat, 05 Sep 2020 09:26:51 -0400, Dan Espen <dan1espen@gmail.com>
> wrote:
> 
>> Niklas Karlsson <anksil@yahoo.se> writes:
>> 
>>> On 2020-09-05, Bob Eager <news0073@eager.cx> wrote:
>>>> 
>>>> (I have to say that his idea of an X terminal's usefulness was how many 
>>>> command prompt windows it could display).
>>> 
>>> That does seem to be how most hardcore *nix geeks use X.
>> 
>> Where I worked, we mostly used Sun Workstations but we had a few X
>> terminals.  The X terminals were just as useful as the workstations.
>> (You could go to any number of other machines for computing.)
>> I got the impression the X terminals weren't cheap enough to catch on.
> 
> This has been the problem with GUI terminals right along.  A terminal
> that can display a GUI has all the pieces to just be a computer so why
> not just put an OS on it and use it as such?
> 
> And before anybody says "Chrombook", ChromeOS is just Linux that
> somebody shot in the leg.

Now they say it’s the “do everything” OS. It runs Android apps, Linux
programs, and Windows programs.

-- 
Pete

[toc] | [prev] | [next] | [standalone]

#213818

On Sat, 5 Sep 2020 17:53:45 -0700, Peter Flass <peter_flass@yahoo.com>
wrote:

>J. Clarke <jclarke.873638@gmail.com> wrote:
>> On Sat, 05 Sep 2020 09:26:51 -0400, Dan Espen <dan1espen@gmail.com>
>> wrote:
>> 
>>> Niklas Karlsson <anksil@yahoo.se> writes:
>>> 
>>>> On 2020-09-05, Bob Eager <news0073@eager.cx> wrote:
>>>>> 
>>>>> (I have to say that his idea of an X terminal's usefulness was how many 
>>>>> command prompt windows it could display).
>>>> 
>>>> That does seem to be how most hardcore *nix geeks use X.
>>> 
>>> Where I worked, we mostly used Sun Workstations but we had a few X
>>> terminals.  The X terminals were just as useful as the workstations.
>>> (You could go to any number of other machines for computing.)
>>> I got the impression the X terminals weren't cheap enough to catch on.
>> 
>> This has been the problem with GUI terminals right along.  A terminal
>> that can display a GUI has all the pieces to just be a computer so why
>> not just put an OS on it and use it as such?
>> 
>> And before anybody says "Chrombook", ChromeOS is just Linux that
>> somebody shot in the leg.
>
>Now they say it’s the “do everything” OS. It runs Android apps, Linux
>programs, and Windows programs.

I've never seen a claim that it runs Windows programs.  It runs
Microsoft cloud applications, which is a rather different thing.

[toc] | [prev] | [next] | [standalone]

#213813

Dan Espen <dan1espen@gmail.com> wrote:
> Niklas Karlsson <anksil@yahoo.se> writes:
> 
> As for consistency, I notice all the mainstream desktop managers
> change with every release.  My WM, Fvwm only changes when I change
> my configuration file.  Fvwm gets new features, it just doesn't force
> change with every release.  My desktop is pretty much the same as it
> was 20 years ago.  That's consistent.
> 

People say the plethora of window managers is a plus, I tend to doubt it.
As far as frequent changes, I thought that Ubuntu moving by default from
Gnome2 to Gnome3 was a big mistake. I installed a new release and it was
hate at first sight. I hunted around and found Mate, which is still Gnome2,
although I guess there are several, like Mint, that are as well. Mate seems
to stay the dame from release to release.

Developers seem to like to change things, maybe not “just because they
can”, but maybe just because someone thinks they have a better idea without
considering how users actually work.
 
-- 
Pete

[toc] | [prev] | [next] | [standalone]

Page 1 of 3  [1] 2 3  Next page →

Back to top | Article view | alt.folklore.computers

csiph-web