> > > *Particularly irritating was emacs's use of <ctrl-s> for* > *"search" because it conflicted with this flow-control, meaning* > *that you had to either reconfigure your tty settings or the**emacs > keybindings.*
I still remember my first experience with Emacs: 1. Open file, edit buffer 2. Undo a mistake using <CTRL+Z> 3. Wonder where the hell everything suddenly went And that, folks, is how I configured my first keybinding in Emacs, simultaneously discovering what ^Z does to a process. > *Back in those days, terminals ran at 30-240 characters per second. Not > all * > *that fast. Actually 300 characters per second, i.e. 300 baud, was slowww!**I > remember being blown away by 9600 baud terminals.* Surely 300 baud was a more refreshing[1] experience than an electromechanical teletypewriter, right? [1] Pun intended. On Wed, 3 Jul 2019 at 07:49, Tadziu Hoffmann <[email protected]> wrote: > > > > Some terminals, the Tek 401x series especially, could > > > be configured to tell the host to stop sending text on > > > a "page full" condition. Some sent the proper RS-232 > > > hardware signals, some sent <ctrl-s>/<ctrl-x>. > > > Really? That's interesting. What did <ctrl-s> do? On the > > terminal emulators I have on hand at the moment, none of them > > are responding or behaving differently. > > I remember these as <ctrl-s> and <ctrl-q> (ASCII DC1 and DC3). > > Typing "stty -a" gives (among other stuff): > > start = ^Q; stop = ^S > > and it works as intended (<ctrl-s> stops output to the terminal > and <ctrl-q> enables it again). > > Wikipedia has a page on this: > > https://en.wikipedia.org/wiki/Software_flow_control > > Particularly irritating was emacs's use of <ctrl-s> for > "search" because it conflicted with this flow-control, meaning > that you had to either reconfigure your tty settings or the > emacs keybindings. (Vi did not use <ctrl-s> or <ctrl-q> > and was therefore "safe".) > > > >
