[TUHS] early Unix terminals (was: Re: Curly braces: An evolution of UNIX and C)

[Clem Cole via TUHS]

There is great story about this.  Control-T came from Tenex   One of the
items that would come back was the status:

Core Execution States

   - *RUNNING*: The program was currently executing code on the CPU.
   - *IO WAIT*: The program was temporarily paused, waiting for an input or
   output operation to complete (most frequently waiting for the user to type
   something on the terminal).
   - *SLEEP*: The program was intentionally paused for a set duration,
   usually via a programmed delay or timer call.
   - *FORK WAIT*: The parent process was suspended, waiting for a
   subordinate parallel process (a "fork" in TENEX/TOPS-20 terminology) to
   finish execution or change state.
   -

Blocked and Sub-System States
Depending on the specific resource or kernel function blocking the process,
the monitor could also return more specialized wait states:

   - *JSYS WAIT*: Waiting on the completion of a specific *Jump to
System* (JSYS)
   monitor call.
   - *PAGE WAIT*: Blocked while the operating system's demand-paging
   architecture retrieved a required memory page from the disk swap space into
   physical core memory.
   - *LOCK WAIT*: Waiting to acquire a system lock or a shared internal
   resource that was currently in use by another job.
   - *DEVICE WAIT*: Waiting specifically on a non-terminal hardware device,
   such as a magnetic tape drive or line printer



Control States

   - *HALT*: The program had encountered an explicit halt instruction or
   hit a hard breakpoint and was no longer running.
   - *^C DETACHED*: The process had been interrupted by the user typing
   Ctrl-C to escape back to the EXEC command level, leaving the background
   program in a paused, unattached state.


Stanford migrated from IBM 360 systems to DEC systems to provide
timesharing cycles for the undergraduates - Pascal programming and doc prep
mostly. After they started up the first TOPS-20 folks customized (hacked)
OS so that depending what the load average, control-T returned different
messages telling you a bit more than “Running.”  So instead of saying
“Running” it started saying “Jogging”, “Walking” or “Crawling” but the
administration got upset when it was most often stating “Dying.”

On the other hand one of my friends that was an undergraduate there at the
time said, the students were able to use it as evidence that there was not
enough compute power to serve the students.  I believe, like CMU at the
same time, Stanford ended up with 4 2060s to replace their mainframe for
undergrad computes.

Sent from a handheld expect more typos than usual

On Wed, May 20, 2026 at 9:22 PM G. Branden Robinson via TUHS <tuhs at tuhs.org>
wrote:

> At 2026-05-20T17:23:22-0700, Larry McVoy via TUHS wrote:
> > > Everyone's experience was different.  The Unix system I used in a
> > > lab at Princeton only had the ASR-33 TTY.  The one I ran at Yale had
> > > the TTY console and a bunch of unique homebrewed bitmap terminals
> > > that I wrote up in an article in Software Practice and Experience.
> >
> > I don't have a lot to add to this other than there was some terminal,
> > I think a VT-something, that had the standard 80x24 but it had a 25th
> > line that was the status line.
>
> Yup, that was a DEC VT220 or maybe 320 feature.  (Others had them, but
> DEC's VT series wagged a great deal of the dog.)
>
> https://vt100.net/docs/tp83/chapter4.html
>
> It's possible that the status line on the 25th row didn't become
> entrenched until the DEC VT320, or was not host-programmable until then;
> Thomas Dickey's terminfo database exhibits an "hs" capability for the
> "vt320" terminal type description but not for "vt220".
>
> https://invisible-island.net/ncurses/man/terminfo.5.html
> https://invisible-island.net/ncurses/terminfo.src.html
>
> If the database is inaccurate, maybe someone can help Thomas improve it.
>
> > I think BSD had a Ctrl-T that dug info out of the kernel and displayed
> > it on that line.
>
> Yup.  With that factoid, not long ago I surprised a colleague at UNSW
> who was much closer to being Present at the Creation than I could have
> been.  He knew Research Unix backwards and forwards, but not BSD.
>
> > When you are in a terminal room with 30-40 people on a 4MB VAX, Ctrl-T
> > was your friend.
>
> Unless _everybody_ learns that trick and exhibits undergrad patience,
> whacking it several times a second, and all those traps into the kernel
> just make the problem worse.  ;-)
>
> Regards,
> Branden
>