🚀 go-pugleaf

[the 3rd attempt at following up .. will it succeed? the world awaits
 with bated breath ..]

In article <5398@brl-smoke.ARPA>, gwyn@brl-smoke.ARPA (Doug Gwyn ) writes:
+ In article <1140@mordred.cs.purdue.edu> avr@mordred.cs.purdue.edu (Andrew V. Royappa) writes:
+ >	Don't know about Most brain-damaged, but the BLIT (dmd 5620, AKA
+ >Bell Labs Intelligent Terminal (I think) -- it should stand for BLAT --
+ >Bell Labs Annoying Terminal). I'm on a BLAT right now, and some of it's
+ >"features" are:
+ >
+ >	1. No known termcap escape for reverse vide -- if someone knows it,
+ >	   please give it to me
+ >	2. Bad mangling of ^S/^Q with emacs (gosling's, anyway).
+ >	3. Most infuriating feature -- if too much text is being printed,
+ >	it just hangs and beeps forever. Makes it tough to page through
+ >	large files. I usually feel like putting my fist through the
+ >	screen when this happens.
+ 
+ This terminal is an AT&T/Teletype 5620 Dot-Mapped Display (DMD) terminal,
+ not a Blit.  The Blit was in effect the DMD's prototype but is not
+ interchangeable with it.  "Blit" is NOT an acronym.
+ 
+ The real problem here is operator stupidity.  He has disabled flow control
+ and is complaining about the consequences.  On recent models (firmware
+ revision 2.0 or later), one can disable DC3/DC1 (^S/^Q, XOFF/XON) flow
+ control safely, but ONLY within a layer, not in standalone mode.  This
+ is explained in the operator's manual.  (One thing that isn't explained
+ is that, on older models, it was important to follow RS-232-C specs in
+ hooking up the terminal.  It was because of this that for a while I had
+ the only DMD at BRL that flow-controlled properly.)  You should also note
+ that the DMD is really intended for operation in layers mode, with
+ standalone use limited pretty much to just getting logged in before
+ starting up layers operation.
+ 

OK .. so I'm stupid. As it turns out, when I do "stty everything" it
shows " ... -tandem ..," even though I don't explicitly set -tandem
(all I set myself are -ctlecho, tostop, and stuff like erase/intr chars).
Is this what you mean by "has disabled flow control ?" Well, stty tandem
doesn't seem to solve the problem either. And if stty tandem IS supposed to
solve the problem, perhaps you could tell me why I don't need to do this
on the four or five other terminals I happen to use or have used.

I'll ask our hardware technician about whether the proper method of installation
was used.

+ Use of ^S/^Q as keyboard commands in EMACS was ALWAYS a mistake.  Too
+ many terminals, particularly fancy models, really do require that these
+ characters be reserved for hardware flow control for correct operation,
+ particularly at high baud rates.  (The DEC VT-100 was probably the first
+ terminal in widespread use for which this was important.)  One should
+ have bound the search and quote commands to other keys than ^S/^Q, for
+ use with all terminals, rather than develop such a bad habit.  There are
+ undoubtedly implementations of EMACS and other programs that don't honor
+ flow control even when one has instructed the OS terminal handler to
+ support it ("stty ixon -ixany" on UNIX), but that's a bug, not a feature.

Well, I've become too used to ^S/^Q, and I don't think I'll change until
I'm forced to. I don't use emacs on a 5620 now, so it's not critical.
In any case, I have this vague notion that I should ideally be able to use
any character that I can generate from the keyboard as an editing character 
(i.e.  I should be able to make the terminal use some other character instead)
but this isn't a particularly strong feeling, I guess.

+ Here are our 5620 termcap entries; recent DMDs support reverse video
+ and underlining directly in their resident terminal emulator firmware,
+ although a downloaded terminal emulator could have whatever features it
+ wanted:

The so string doesn't produce reverse video on my terminal, so as other
people have already pointed out to me it must be the case that our
5620s around here need a firmware upgrade.

			Thanks,

					Andrew Royappa
					ihnp4!purdue!avr