Re: not good proposal: "C-z <letter>" reserved for users

[Jean Louis]

* Emanuel Berg via Users list for the GNU Emacs text editor 
<help-gnu-emacs@gnu.org> [2021-02-14 11:34]:
> > tmux, screen or nohup are not related to shell job control.
> > They run processes and continue running them even if user
> > logs off. That feature is not related to job control of
> > a process, but it can be helpful to keep the suspended job
> > in a shell even if user logs off.
> 
> Of course a terminal multiplexer (read tmux) it is not the
> same as job control. It is better, that's why it has
> superseded the C-z/fg practice.

Does it do job control?

>From screen manual:

,----
|       suspend
| 
|       Suspend screen.   The windows  are in the  `detached' state,
|       while  screen is suspended. This feature relies on the shell
|       being able to do job control.
`----

It maybe cooperates with job control, relies on it, but `screen' does
not do job control.

Of course you may compare chicken and apples, sure, it is freedom, but
chicken will never grow on the tree.

> To experience one immediate advantage - literally immediate -
> compare opening a new tmux pane to doing C-z. With tmux, you
> still have interactive control over Emacs, or whatever other
> of a dozen or so applications and interfaces you might have
> open, visible at the same time, with tmux panes all over
> a huge monitor. Including, if you wish, a dedicated pane, just
> to do job control! Compare this to suspending Emacs and then
> 'fg' it back and forth.

So you prefer to compare processes running in parallel to compare to
processes that are stopped. It is fine with me, but that neither
tmux/screen/nohup suspend the process, they run them.

Running is not equal to suspend.

Stopping is not equal to running.

Running ≠ Stopping

Control-Z in shell does Stopping of a process. It does not let process
run in background. `bg' command could run it in the background, but
that is not same as stopping.

Tmux and screen let processes run in background, they are not tools
designed to stop the process, they let job control to the shell.

> You also mention the 'persistent IRC' stunt that can be done
> with tmux:
> 
>   tmux attach [-t X]
>   tmux kill-session -t X
>   tmux list-sessions
>   tmux new -s X-s 'tmux set remain-on-exit on; X'

I use screen and surely I know what it means in tmux, I have long
running, days and weeks running processes on remote servers that I may
control by using screen. But again, making things persistent is not
equal to making things stop. It is not comparable.

> Also, you seem to say this is such a fundamental practice,
> however with zsh, it isn't even enabled by default. (IME
> people seem to use bash, zsh, and - the OpenBSD people - ksh
> (which is actually rksh, for legal reasons). FTR it _is_
> enabled by default on bash and ksh.)
> 
> So it isn't even enabled by default on one of the most
> commonly used shells.
> 
> tmux should be many magnitudes more powerful for several
> reasons, in this aspect and others. Use it and get into the
> game :)

Sure, you win.

We can also join to it, that X Window system is more powerful to shell
job control, even the water boiler is more powerful to it. But is not
relevant.

Jean