From CJL at REAGAN.AI.MIT.EDU  Tue Mar 29 20:19:00 1988
From: CJL at REAGAN.AI.MIT.EDU (Chris Lindblad)
Date: Mar 29 88 13:19 EST
Subject: Some history questions.
Message-ID: <880329131943.9.CJL@OTIS.AI.MIT.EDU>

I'm trying to track down the history of file version numbers, and of delete
and expunge.

Was it really Tenex that invented the features of file version numbers and
delete-and-expunge?

Or did file versions in the second name of files appear in ITS before Tenex,
or was the idea inspired by Tenex?

Can anyone point me to an early Tenex paper describing these features?  Any
ITS papers?


From MAEDA at AI.AI.MIT.EDU  Sat Mar 12 17:24:38 1988
From: MAEDA at AI.AI.MIT.EDU (Christopher M. Maeda)
Date: Mar 12 88 11:24:38 EST
Subject: UMass ITS's
Message-ID: <340451.880312.MAEDA@AI.AI.MIT.EDU>

I heard of a pretty big Intelligent Tutoring Systems effort 
going on at UMass.  I have the paper at home if you want it.
They aren't really doing anything new, just applying AI to a 
new area.  The most interesting thing about the paper is that
they continually use the acronym ITS to describe their systems.

Chris


From Ed at MEAD.SCRC.Symbolics.COM  Mon Mar  7 21:57:00 1988
From: Ed at MEAD.SCRC.Symbolics.COM (Ed Schwalenberg)
Date: Mar 7 88 15:57 EST
Subject: [ota: SPACE Digest V8 #156]
In-Reply-To: <12380322663.23.SRA@XX.LCS.MIT.EDU>
Message-ID: <880307155748.0.ED@BLACK-BIRD.SCRC.Symbolics.COM>

    Date: Sun 6 Mar 88 23:23:32-EST
    From: Rob Austein <SRA at XX.LCS.MIT.EDU>

	Date: Sun,  6 Mar 88 14:22:49 EST
	From: "Christopher M. Maeda" <MAEDA at AI.AI.MIT.EDU>

	The following was posted to space digest in a discussion about 
	remote logins to the moon.,,
    [Note COMSAT ".." lossage here^^.]

	    From: tektronix!reed!douglas at ucbvax.berkeley.edu  (P Douglas Reeder)
	    Subject: Re: data and long distances

	    The distance problem applies to satelites in geosynchonous
	    orbit, as well.  radio wave take a noticeable fraction of a
	    second to get there and back.  That would play hell with high
	    baud rates if not accounted for.  A comsat expert might know
	    how it's done.

    For interplanetary stuff you'd want to use a batched mail protocol
    like BSMTP over a high bandwidth transmission protocol like NETBLT for
    mail.  You could probably get away with SMTP to Lunagrad (Moonbase
    doesn't look like it's going to be an issue any time soon).

    Remote login will be bad no matter what.  The best you could do would
    be something like RMS's local editing protocol, again using something
    like NETBLT for transmission so that at least screen updates would be
    fast once they arrived.

I actually tried using a computer in Miami from Antarctica via a geosynch
satellite.  It was, er, painful.  But I did a fun experiment too: I did
an analog loopback through the satellite, and watched my characters echo
on the screen.  Then I tried typing a character, and while it was on its
way up and back I digitally looped back my end.  After about 6 passes the
character would begin to decay:
	aaaaaabp~~~~~~<rubout><rubout><rubout>
Another satellite hacker (the gentleman in Miami) ran a PDP-11 to the satellite,
ran the analog downlink back up to another channel on the same satellite, then
to a terminal, for a total delay of about .75 sec.

From SRA at XX.LCS.MIT.EDU  Sun Mar  6 00:00:00 1988
From: SRA at XX.LCS.MIT.EDU (Rob Austein)
Date: Sun 6 Mar 88, 00:00
Subject: [ota: SPACE Digest V8 #156]
In-Reply-To: <336939.880306.MAEDA@AI.AI.MIT.EDU>
Message-ID: <12380322663.23.SRA@XX.LCS.MIT.EDU>

    Date: Sun,  6 Mar 88 14:22:49 EST
    From: "Christopher M. Maeda" <MAEDA at AI.AI.MIT.EDU>

    The following was posted to space digest in a discussion about 
    remote logins to the moon.,,
[Note COMSAT ".." lossage here^^.]

    	From: tektronix!reed!douglas at ucbvax.berkeley.edu  (P Douglas Reeder)
    	Subject: Re: data and long distances

    	The distance problem applies to satelites in geosynchonous
    	orbit, as well.  radio wave take a noticeable fraction of a
    	second to get there and back.  That would play hell with high
    	baud rates if not accounted for.  A comsat expert might know
    	how it's done.

For interplanetary stuff you'd want to use a batched mail protocol
like BSMTP over a high bandwidth transmission protocol like NETBLT for
mail.  You could probably get away with SMTP to Lunagrad (Moonbase
doesn't look like it's going to be an issue any time soon).

Remote login will be bad no matter what.  The best you could do would
be something like RMS's local editing protocol, again using something
like NETBLT for transmission so that at least screen updates would be
fast once they arrived.

You might also want to ask somebody at BBN about the current tuning of
the ARPANET (net 10.0.0.0 itself): one of the three transcontinental
channels is a comsat link, the other two are land lines, and all hell
broke loose for the few days between when they installed the comsat
link and when they got it tuned properly.
-------


From MAEDA at AI.AI.MIT.EDU  Sun Mar  6 20:22:49 1988
From: MAEDA at AI.AI.MIT.EDU (Christopher M. Maeda)
Date: Sun,  6 Mar 88 14:22:49 EST
Subject: [ota: SPACE Digest V8 #156]
Message-ID: <336939.880306.MAEDA@AI.AI.MIT.EDU>


The following was posted to space digest in a discussion about 
remote logins to the moon.,,

	From: tektronix!reed!douglas at ucbvax.berkeley.edu  (P Douglas Reeder)
	Subject: Re: data and long distances

	The distance problem applies to satelites in geosynchonous
	orbit, as well.  radio wave take a noticeable fraction of a
	second to get there and back.  That would play hell with high
	baud rates if not accounted for.  A comsat expert might know
	how it's done.

Chris