🚀 go-pugleaf

Date:     8 Jul 84 2:29:39-EDT (Sun)
From: uw-beaver!Peterr%toronto.csnet@csnet-relay.arpa
To: info-mac@sumex-aim.arpa
Subject:  Scripts for Mac/Lisa

About a year ago, perhaps a bit more, Apple sent some of their Lisa technical
people on a university speaking tour.  While they were here, I asked them
about "scripts" (i.e. a batch or "SUBMIT" facility) for the Lisa.  They said,
quite definitely, that they were working on it.  I've heard nothing since.

About 4 months ago, an Apple Canada sales rep. spoke to a local IEEE chapter.
It was a very disappointing talk, but I did get to ask my question again--
what about a script or batch facility for the Lisa or Mac?  This time, the
question was greeted with "why would you ever want to do that?" and
dismissed.

But I don't think it will go away so easily.  There are at least two general
areas where one needs "scripts":
- automation of a task that is performed identically each time, but takes
  quite a while to complete
- description of a class of tasks, parameterized in some way

The first is not hard, and much could be gained simply by having a "remember"
mode that dribbled all the mouse movements and presses and keyboard presses
into a file and a "play" command that fed the contents of such a file into
the input buffers.

The second is harder, and seems pretty much an open question for point-and-
press systems.

One interesting observation can be made, however:  we often use scripts to
maintain invariants-- if the invariants were maintained automatically, we
would not need the scripts.  Assume we have a weekly report to produce,
formulated as a MacWrite document incorporating a MacPaint picture.  Each
week, the picture has to be changed.  So, each week, we have to paste the
new version of the picture into the MacWrite document, after deleting the
old version of the picture.  We would want a script to do this.  However,
if we could state the invariant "this MacP picture is always inserted into
this MacW document at this place" (this could be done with something no more
sophisticated than the #include of cpp, for a start), we would not need a
script.  As this is rather more declarative than procedural, it also seems
to fit into the Apple paradigm better than scripts do.

If someone at Apple would say what happened to scripts for the Lisa, I'm
sure quite a few people would be interested in reading it.

p. rowley, University of Toronto