From ALAN at MIT-MC.ARPA  Wed Oct 30 02:47:43 1985
From: ALAN at MIT-MC.ARPA (Alan Bawden)
Date: Oct 29 85 20:47:43 EST
Subject: Silly lineprinter devices again.
Message-ID: <[MIT-MC.ARPA].697383.851029.ALAN>

Remember the 7LP: and 7LR: devices?  Well, now that we have a third such
device on the 9th floor, and another one coming soon to the 8th, it would
make sense for me to tell you that I have installed 8LP: and 9LP: devices,
right?  WRONG!  Since a clear numbering trend is emerging it makes sense to
change the naming scheme so that the number comes -last-.  So the official
names are now:

	LP7:	The 7th floor QMS 2400
	LP8:	The 8th floor QMS 2400
	LP9:	The 9th floor QMS 1200

	LR7:	The 7th floor laserwriter

The two old names 7LP: and 7LR: continue to work for compatibility.

From ALAN at MIT-MC.ARPA  Mon Oct 28 04:20:53 1985
From: ALAN at MIT-MC.ARPA (Alan Bawden)
Date: Oct 27 85 22:20:53 EST
Subject: echoing weirdness
In-Reply-To: Msg of Sun 27 Oct 85 20:08:43 EST from Ian Macky <GRENNY at MIT-MC.ARPA>
Message-ID: <[MIT-MC.ARPA].694733.851027.ALAN>

    Date: Sun, 27 Oct 85 20:08:43 EST
    From: Ian Macky <GRENNY at MC>
    *fr^K!a l
    e
    [There goes GZ]
    cTo OAF.G.BRIDGET at MIT-OZ:
    ittle pkish, perhaps... <End>

    [the "ec" from "peckish" got echoed before anything else... how WEIRD!!
    (well, not before everything, but you see)]

Just another wierd case with echoing input at interrupt level.  ITS decides
how to do interrupt level echoing by looking at what job currently has
control of the TTY.  In this case DDT had control of the TTY for a few
moments while it typed the Argus message, so ITS echoed the two characters
you typed during those moments.  (I guess FR is looking at %TXPIE bits in
character to determine how to echo the user's typeahead.)

I'm suprised that FR and REPLY aren't handling %PIATY interrupts.  If they
were you wouldn't have seen this happen.  SEND seems to be hacking %PIATY;
why should SENDER vary this aspect of its behavior for replying vs. sending?

From GRENNY at MIT-MC.ARPA  Mon Oct 28 02:08:43 1985
From: GRENNY at MIT-MC.ARPA (Ian Macky)
Date: Oct 27 85 20:08:43 EST
Subject: No subject
Message-ID: <[MIT-MC.ARPA].694678.851027.GRENNY>

echoing weirdness:

*fr^K!a l
e
[There goes GZ]
cTo OAF.G.BRIDGET at MIT-OZ:
ittle pkish, perhaps... <End>

[the "ec" from "peckish" got echoed before anything else... how WEIRD!!
(well, not before everything, but you see)]

From ALAN at MIT-MC.ARPA  Fri Oct 25 04:38:25 1985
From: ALAN at MIT-MC.ARPA (Alan Bawden)
Date: Oct 24 85 23:38:25 EDT
Subject: Any takers?
Message-ID: <[MIT-MC.ARPA].692189.851024.ALAN>

With all of these parity errors MC gets these days, and with all of these
naive users who haven't the slightest idea what to do when they have
detached job trees, the amount of disowned junk is getting to be a pain.
Now there is really no good reason why a program couldn't be written that
would would look around in the system for jobs with your XUNAME and offer
some intelligent (and clearly explained) choices to you.  Anybody
interested?  I'm not up for thinking about the problem right now, but I'd
be happy to brainstorm in person with anyone else who would like to take a
crack at it.

From ALAN at MIT-MC.ARPA  Tue Oct 22 19:29:28 1985
From: ALAN at MIT-MC.ARPA (Alan Bawden)
Date: Oct 22 85 14:29:28 EDT
Subject: super-image mode/slaved ttys
In-Reply-To: Msg of Tue 22 Oct 85 00:14:31 EDT from David Vinayak Wallace <GUMBY at MIT-MC.ARPA>
Message-ID: <[MIT-MC.ARPA].688804.851022.ALAN>

    Date: Tue, 22 Oct 85 00:14:31 EDT
    From: David Vinayak Wallace <GUMBY at MIT-MC.ARPA>
    ...
    1> I know he's on the same kind of terminal I am, so I want whatere
       goes down the wire to his terminal to go down the wire to mine.

    2> I want the system to hack ^P codes for me so that I can poke around
       even if our terminals don't match.

    In my previous letter I was asking for 1> above, but two would be more
    useful in most cases.  By "escape code" I meant some ^_ code to tell
    ITS just to pump the characters through.
    ...

Getting any scheme like 2 to work is probably impossible.  What do you do
if the guy's terminal supports insert and delete character and yours
doesn't?  What about if his terminal is 5 lines longer than yours?  Even
virtual terminals have characteristics that are visible to the programs
that use them.

1 could be made to work I guess.  You would have to do it at the level of
%TD codes rather than ^P codes (since that is the only level at which there
is really a universal language).  There is a bunch of other things you have
to worry about I guess, like making sure the characters you type get echoed
in the way his program desires.  Given that it can only work some of the
time, and that it is pretty hard to get right, I can think of better things
to do with our time.

From GUMBY at MIT-MC.ARPA  Tue Oct 22 05:14:31 1985
From: GUMBY at MIT-MC.ARPA (David Vinayak Wallace)
Date: Oct 22 85 00:14:31 EDT
Subject: super-image mode/slaved ttys
In-Reply-To:         Msg of Mon, 21 Oct 85 13:48:07 EDT from Alan Bawden <ALAN at MIT-MC.ARPA>
	Msg of Mon, 21 Oct 85 18:23:37 EDT from Ed Schwalenberg <ED at MIT-MC.ARPA>,
Message-ID: <[MIT-MC.ARPA].687999.851022.GUMBY>

    Date: Mon, 21 Oct 85 13:48:07 EDT
    From: Alan Bawden <ALAN at MIT-MC.ARPA>

	Date: Mon, 21 Oct 85 06:11:40 EDT
	From: David Vinayak Wallace <GUMBY at MIT-MC.ARPA>

	Is there any way to get ITS to put your tty in super-image mode via
	escape codes?  Otherwise when you slave a tty you get ghubbish (making
	slaved ttys almost useless.  Are they a pre-TS3TTY relic?).

    I can't figure out what you are asking from this.  First please distinguish
    between super-image input and super-image output (two quite unrelated
    things despite the similarity in names).  Second please distinguish between
    the various kind of escape codes in the ITS world; are you talking about ^P
    codes?  Third, by "slaved ttys" are you refering to the state you can
    achieve in a com-link where one console's type-in is seen by another
    console's job-tree?

    Date: Mon, 21 Oct 85 18:23:37 EDT
    From: Ed Schwalenberg <ED at MIT-MC.ARPA>

    And fourth, is the TTY you are slaving logged-in or free, and what type
    (software, printing, AAA, etc.) is your TTY and the one you're slaving?

Ok.  Let me start over:

Luser :LUSERs me and tells me he can't run BABYL.  BABYL runs fine for
me on his mail file.  I'd like to ^_e him and poke around in the
debugger in his BABYL.  But all I get is ghubbish on my screen.

Two possible ways that ITS could get around this (and maybe both are
desirable for different circumstances):

1> I know he's on the same kind of terminal I am, so I want whatere
   goes down the wire to his terminal to go down the wire to mine.

2> I want the system to hack ^P codes for me so that I can poke around
   even if our terminals don't match.

In my previous letter I was asking for 1> above, but two would be more
useful in most cases.  By "escape code" I meant some ^_ code to tell
ITS just to pump the characters through.

Free vs logged-in isn't relevent to this, since you can :TCTYP a free
tty after sending a CALL through.

From CENT at MIT-MC.ARPA  Tue Oct 22 04:08:10 1985
From: CENT at MIT-MC.ARPA (Pandora B. Berman)
Date: Oct 21 85 23:08:10 EDT
Subject: file system not fukt
Message-ID: <[MIT-MC.ARPA].687814.851021.CENT>

i should have been using $^R. my mind must be going. sorry to bother you.

From CENT at MIT-MC.ARPA  Tue Oct 22 04:00:01 1985
From: CENT at MIT-MC.ARPA (Pandora B. Berman)
Date: Oct 21 85 23:00:01 EDT
Subject: FILE SYSTEM FUKT
Message-ID: <[MIT-MC.ARPA].687803.851021.CENT>

i was in ddt, moving some ostats files to third: with $$^O. it worked for
the first one i tried no problem -- this was the second most recently
created ostats file.  then i tried
$$^O ostats >,third:
and got the reply that it can't copy to a different directory -- third:
so i tried specifying the file version
$$^O ostats 3636,third:
and it replied
third: .mail.;ostats 3636 - file exists (!)

clearly someone is real confused here, and i don't think it's me.

From ALAN at MIT-MC.ARPA  Mon Oct 21 18:48:07 1985
From: ALAN at MIT-MC.ARPA (Alan Bawden)
Date: Oct 21 85 13:48:07 EDT
Subject: super-image mode
In-Reply-To: Msg of Mon 21 Oct 85 06:11:40 EDT from David Vinayak Wallace <GUMBY at MIT-MC.ARPA>
Message-ID: <[MIT-MC.ARPA].687073.851021.ALAN>

    Date: Mon, 21 Oct 85 06:11:40 EDT
    From: David Vinayak Wallace <GUMBY at MIT-MC.ARPA>
    Is there any way to get ITS to put your tty in super-image mode via
    escape codes?  Otherwise when you slave a tty you get ghubbish (making
    slaved ttys almost useless.  Are they a pre-TS3TTY relic?).

I can't figure out what you are asking from this.  First please distinguish
between super-image input and super-image output (two quite unrelated
things despite the similarity in names).  Second please distinguish between
the various kind of escape codes in the ITS world; are you talking about ^P
codes?  Third, by "slaved ttys" are you refering to the state you can
achieve in a com-link where one console's type-in is seen by another
console's job-tree?

From GUMBY at MIT-MC.ARPA  Mon Oct 21 11:11:40 1985
From: GUMBY at MIT-MC.ARPA (David Vinayak Wallace)
Date: Oct 21 85 06:11:40 EDT
Subject: super-image mode
Message-ID: <[MIT-MC.ARPA].686691.851021.GUMBY>

Is there any way to get ITS to put your tty in super-image mode via
escape codes?  Otherwise when you slave a tty you get ghubbish (making
slaved ttys almost useless.  Are they a pre-TS3TTY relic?).

From OAF%MIT-AI.ARPA at MIT-MC.ARPA  Thu Oct 10 08:15:03 1985
From: OAF%MIT-AI.ARPA at MIT-MC.ARPA (Oded Feingold)
Date: Oct 10 85 03:15:03 EDT
Subject: No subject
Message-ID: <[MIT-AI.ARPA].5013.851010.OAF>

I plugged terminals in and out of TTY07, and eventually the DZ wouldn't
answer anymore.  I reloaded ITS, and left the crash dump on
	DZ SILENT.
I doubt there's much interesting there.

From CSTACY at MIT-MC.ARPA  Mon Oct  7 23:12:27 1985
From: CSTACY at MIT-MC.ARPA (Christopher C. Stacy)
Date: Mon,  7 Oct 85 18:12:27 EDT
Subject: some job device losing
Message-ID: <[MIT-MC.ARPA].671384.851007.CSTACY>

I can't use any jobdevs right now because there are a whole pile of
what looks like MLDEVs with the jname "CFTP" and channels open on the
file "OZKS: ;".  I am going to gun them down, preserving one corpse
CRASH;MLDEV OZ.  The pc is/345 (.HANGing).


From TIM at MIT-MC.ARPA  Fri Oct  4 04:49:07 1985
From: TIM at MIT-MC.ARPA (Tim McNerney)
Date: Thu,  3 Oct 85 23:49:07 EDT
Subject: No subject
In-Reply-To: Msg of Thu  3 Oct 85 15:08:07 EDT from Christopher C. Stacy <CSTACY at MIT-MC.ARPA>
Message-ID: <[MIT-MC.ARPA].668221.851003.TIM>

    Date: Thu,  3 Oct 85 15:08:07 EDT
    From: Christopher C. Stacy <CSTACY at MIT-MC.ARPA>
    To:   BUG-ITS at MIT-MC.ARPA

    There is a POOR-MC at MC list now.

What's it for?

From CSTACY at MIT-MC.ARPA  Thu Oct  3 20:08:07 1985
From: CSTACY at MIT-MC.ARPA (Christopher C. Stacy)
Date: Thu,  3 Oct 85 15:08:07 EDT
Subject: No subject
Message-ID: <[MIT-MC.ARPA].667593.851003.CSTACY>

There is a POOR-MC at MC list now.