[fa.editor-p] Z really works - experience

C70:editor-people (06/10/82)

>From Mishkin@YALE Thu Jun 10 01:46:36 1982
Date:  3 Jun 1982 1143-EDT
From: John R. Levine, The INTERACTIVE Electric Calculator Co., Cambridge MA.
Sender: johnl at IMA
Subject: Z really works - experience

     Just to reassure the people who had bad experience with Yale editors
a long time ago, yes the old E editor for the PDP-10 was an awful program
but the new Z editor really works.  I'm one of the few people left who
has used both extensively.

     The old E editor was written in 1971 for a KA10 and Sugarman
terminals, because they were cheap and that was all we had or expected to
get soon.  As far as I can tell, Yale was the only place that ever bought
that particular model of Sugarman.  The editor was about the first PDP-10
code that the guys who wrote it ever wrote, and you can tell.  No useful
comments at all, for example.

     At the time, the Yale department actually turned down a chance to be
on the Arpanet because the senior member of the department thought that
networking was a big waste of time.  (He was right, of course; look how
much time we all spend sending and reading this junk mail.)  Thus Yale's
isolation and original awful editor.  The main thing that can be said in
defense of E was that it was small and fast, which was important on a
192K KA-10.

     Z comes from a more mature environment and from ten more years
of experience.  There are multiple terminal types and multiple Twenexes
at Yale, and the person who wrote most of Z (Steve Wood) is a far more
careful software developer than were the E authors when they wrote
it.

     People with any interest in editors at all and access to a DEC-20
owe it to themselves to try Z, if only to come up with informed reasons
that they don't like it.  I personally like Z a lot, both because it is
an editor style that I like and because on a moderately loaded 20 it
responds faster than other editors.

                John Levine (Levine @ YALE, or decvax!cca!ima!johnl)

     P.S.:  Many of the original Rand editor people later went off to
form Interactive Systems in Santa Monica.  They (Ned Irons in particular)
attacked the sluggish response problems that plague screen editors by
putting as much as possible of the editor in the terminal.

     That project was surprisingly successful and the second generation
smart terminal is now a regular Interactive product.  It really works and
speeds up editing response immeasurably.  An added advantage is that all
messages between the terminal and the host are encoded as lines of
printing characters so that you can use a real screen editor even if the
transmission medium or remote computer is line at a time, e.g. Telenet.
If there is interest I can say more about Interactive's editor and
editing terminals.
-------