Digital PDFs

EK-TY503-PR-001

December 2000

3 pages

Original

0.3MB

Document:	DECset-8000 A Review
Order Number:	EK-TY503-PR
Revision:	001
Pages:	3
Original Filename:	http://bitsavers.org/pdf/dec/pdp8/decset-8000/DECset-8000_A_Review.pdf

OCR Text

Text F o r m a t t i n g
Section Editorl
Russ Abbott
Department o f Computer Science
C a l i f o r n i a State U n i v e r s i t y ,
Northridge
Northrid~e, California

DECset

8000

-- A

Review

91324
D E C s e t - 8 0 0 0 System User's Guide
Vol #I, #2
Dec #EK-TY800-OP-002, EK-TY808-PR-001
vincent Manis

and

Peter van den Bosch
vision
set
-- and yet this is precisely
what is
expected
of
them
by computer
system designers.

When
it comes to designing software
zor
people
not
used
to
dealing
with
computers,
manufacturers of hardware are
f reguently
the
worst
offenders
in
producing complex, inconsistent and frustrating systems. ~he ~roblem is that they
~end
to think of all people as potential
programmers. Now, a programmer, though he
may
grumble, will put up with almost any
z n c o n v e n i e n c e thrown in his
way,
but
a
~ypist is not a programmer, nor is a bank
clerk, nor is a
journalist.
Yet
these
people,
and
others
in every aspect of
business and industry,
are
expected
to
deal,
in
increasing numbers, with software designed for programmers
and
other
members of the computer culture. They tco
will adjust themselves, since it is their
Aivelihcod,
but
they
will not grumble,
Decause they too
often
do
not
realize
that
matters
could be better. When "the
computer" sakes a mistake it will
afford
them
a
moment of bitter pleasure during
c o f f e e - b r e a k s , but in general
they
will
feel
frustrated,
inadequate
and
oppressed, and their working
lives
will
~e
made
just
a little more unpleasant
because of it.

Text
formatting
is no simple task:
typesetting is a skilled discipline which
combines
a sense
of
aesthetics with a
deep knowledge of
the
available
tools:
even
the
typing
of
a
business letter
requires a considerable amount of experience
before
it is done well. A computer
is
no
substitute
for
a
sense
of
aesthetics.
It
is
merely
a tool
to
lighten the load of other, less difficult
and
therefore
often
duller
tasks.
However,
when
the
use
of
a computer
merely
replaces
one
set
of
wellunderstoed but dull
tasks
with
another
set,
far less well-understood, the human
function, the
sense
of
aesthetics,
is
thwarted
and
the computer
loses
its
effectiveness as a tool. A case in
point
is the DECset-8000 typesetting system.
It
is
worth pointing out, first of
all, that a typesetting system like
this
one
is
generally
a small
operating
system: there are facilities for managing
story
files,
displaying formatted text,
editing,
sending
material
off
to
the
photocomposition
machine,
etc. This is,
in principle,
an
improvement
over
the
sort of support generally provided by the
photocompositor manufacturer,
where
the
user
produces a paper tape (usually at a
Teletype) and is left only with
a crude
paper
tape
editor
to make changes. But
punching and editing paper tapes are
two
simple
tasks which one can learn without
undue effort; learning to use an operating system may well be another matter.

Why do computer system designers get
away
with
it?
If an architect designed
unusable buildings he would very
quickly
~un
out of clients (unless he were first
proclaimed a genius, in which case public
inertia would keep him very busy indeed);
more than likely he would be drummed
out
of
his professional association. This is
also true of surgeons whose patients were
put
back
together in a biologically unusual way, engineers whose
bridges
only
went
half-way across the river, and lawyers whose contracts were full
of loopholes.

The
DECset-8000 system is a turnkey
PDP-8 c o n f i g u r a t i o n
designed
for
newspaper
typesetting. It comprises editing,
filing,
proofing,
justification
and
hyphenation
facilities,
as
well as the
ability to generate tapes
for
a phototypesetter.
It includes
a Wire Service
Subsystem for dealing
with
AP/UPI
wire
service
input,
a subsystem for handling
classified
ads,
and
the
ability
to
generate reports on operator performance.

An
excellent
example
of
software
generally
used
by
non-programmers
and
a l m o s t invariably inadeguate in design is
that intended for
text
formatting
(and
this
includes typesetting). Here we have
a class of users whose only interest in a
computer
is
that it will speed up their
p r o c e s s i n g of large amounts of text: they
are
not programmers, nor are they interested in becoming programmers -- any more
than
I
am
interested
in
becoming
an
e l e c t r o n i c engineer when I buy
a tele-

-7-

It
appears, from the d o c u m e n t a t i o n ,
that the D E C s e t - 8 0 0 0 p e r f o r m s many
newspaper
typesetting
activities.
However,
there a p p e a r s to have been little t h o u g h t
taken
regarding
human
factors.
For
example, at the top level, there
are
at
most
26
programs, one per l e t t e r of the
alphaDet. A single
mistyped
letter
can
put
the
operator
into
the
disc p a t c h
r o u t i n e s , where a n g e l s fear to tread. One
program
is
a general
monitor,
which
allows
the
user
to
execute
utility
commands
(which
have
a rather peculiar
syntax). Thus, it a p p e a r s that human
engineerinq
considerations,
such
as
the
provision
of
a
system
which
always
presents
one
face to the user, have not
been met.

p o i n t s out a c o m m c n problem ~ith c o m p u t e r
systems:
it is not e n o u g h to i m i t a t e the
c r u d e m e t h o d s in use before c o m p u t e r s are
introduced
-- it is about time s u p p l i e r s
t o o k care to provide s o m e t h i n g better.
we come now to that aspect
of
text
~ o r m a t t i n g which all text f o r m a t t e r s more
or less have in
common:
the
formatting
language.
This
'.language"
is
almost
~ n v a r i a b l y a set of c o m m a n d s
ranging
in
syntactic
complexity
from
special
character-codes
(i.e.,
the
Teletype
School
of
Programming,
s t r o n g l y influe n c e d by the p r e s e n c e of
the
CTRL-ke7),
through
codes
with
one
or
more
p a r a m e t e r s , to g e n e r a l - p u r p o s e languages,
often
entered
by
means
of
escape
characters
in
the
input
stream.
The
uECset
format
language
falls
into the
m i d d l e range of such l a n g u a g e s : a c o m m a n d
zs
followed
by a list of p a r a m e t e r s sep a r a t e d by commas; in
most
cases
there
are
reasonable
defaults
for
absent
p a r a m e t e r values.

R e g a r d i n g the
programs
themselves,
there
is r e l a t i v e l y l i t t l e to say. There
is an e d i t o r , a " d r i v e r markup"
program,
a hyphenato~,
and
so on. The fact that
each of t h e s e p r o g r a m s
has
a
different
sort of c o m m a n d s y n t a x from all o t h e r s is
r e g r e t t a b l e . The d o c u m e n t a t i o n
makes
it
a p p e a r that each program does all
it
is
e x p e c t e d to.

One c a n n o t fault the f o r m a t l a n g u a g e
ior
lack
of
command
types.
What
i~
l a c k i n g is a s e n s e of
design,
of
princ i p l e s u n d e r l y i n g the e n t i r e language. It
is
admirable
that
the
designers
have
tried
to
come
up with a l a n g u a g e which
will drive a number of
different
photocomposition
machines,
but
instead
of
~ r y i n g to a b s t r a c t some basic
principles
of text f o r m a t t i n g and b u i l d i n g on these,
~hey appear to have s i m p l y
extended
the
language
w h e n e v e r there was a f e a t u r e on
a
given
machine
which
the
language
couldn't
yet
exploit.
The
result
is
redundancy.

Systems
such
as
EECset-8000
are
o f t e n d e f e n d e d on the g r o u n d s
that
they
are
turnkey,
d e s i g n e d for those with no
K n o w l e d g e about computers.
There
is
no
general
design
principle
which
states
that
elegance
and
ease
of
use
are
contradictory
-- ask an architect! There
£s no r e a s o n
why
such
turnkey
systems
c a n ' t be i ~ p l e m e n t e d within the f r a m e w o r k
of a s i m p l e
operating
system
such
as
Jnix. Such a system p r o v i d e s named files,
~ a t h e r than n u m b e r e d blocks, and
a
consistent
environment
for
command
e x e c u t i c n . An OS a p p r o a c h also gives
the
sophisticated
user
or
consultant
the
a b i l i t y to add to the r e p e r t o i r e of
p~og r a m s , thus e x t e n d i n g product life.

For example,
there
are
"qua ddin q"
commands
(a t y p e s e t t i n g
term
for
the
process of f i l l i n g a line out with
white
space)
and "output mode" c o m m a n d s (which
are
concerned
with
centring
lines,
setting
them
with
u n j u s t i f i e d right or
left m a r g i n s , or
justifying
them}.
Now
o b v i o u s l y , q u a d d i n g is what the f o r m a t t e r
does when it sets a line a c c o r d i n g
to
a
current
output
mode, so q u a d d i n g is the
more p r i m i t i v e c o n c e p t ;
but
the
manual

It
may
be argued that the PDP-8 in
the D E C s e t - 8 C 0 0 can't s u p p o r t such an amb i t i o u s s o f t w a r e e n v i r o n m e n t . This claim,
e v e n if
true,
is
irrelevant.
It
was,
a f t e r all, DEC who c h o s e the p r o c e s s o r -an 11 would
have
added
little
to
the
overall
system
cost.
The
DECset-8000

Local Groups
New York
Steve Mandell
245 East 19th Street
New York, N.Y. 10003
212/673-5111

San Francisco Bay Area
Tom Griffin
Hewlett Packard
1501 Page Mill Road
B l d g 6 - Upper
Palo Alto, CA 94304

415/.93-15Ol

DUsseldorf
R o b e r t Jones
MANDATA

Unternehmensberatung
GmhH
4 Diisseldorf-lieerdt
Werftsrasse 23
West Germany

(0211) 50 30 10

x2~O2

-8-

Toronto
Richard Nelson
Dept. of Computer Sexwices
York University
t~700 Keels Street
Downsvlew, Ontario, Canada

M3J IP3

416/677-6308

Washlr~ton, D. C.
Christopher H Igglns
Boor Allen & Hamilton
1025 Connecticut Ave,, lq~
Washington, D. C. 20036

presents
the
two
in parallel {in fact,
separated by a number of
unrelated
command groups) as independent concepts.

One further
topic
of
interest
to
readers
of Sigdoc • is the documentation
Atself. This is in many ways
inadequate,
e s p e c i a l l y for users not used to computer
manuals.
Concepts
are
not
clearly
explained,
nor is there a clear overview
of
the
system:
the
many
(too
many)
special
purpose
editors, the formatting
language, the disk system etc., all
flow
together
into
an
unmanageable plate of
~paghetti. The manuals are confusing, and
it
is
hard to figure out where you are,
nor do important
concepts
always
stand
out.
There
is
not
even
a glossary,
although typesetting and computing terminology
freely
intermingle
in the documentation, to the inevitable confusion of
both
kinds
of readers likely to take up
~hese volumes. No
doubt
this
confusion
reflects
the
overall
confusion
of the
system's design.

In
a
similar
vein,
the
language
gives the user limited ability to
define
commands
in
terms of existing commands.
"Limited,"
because
these
user-defined
commands are not permitted any parameters
themselves, nor can they stand
for
any
common
sequence of two or more commands.
In other words, in place of the
sequence
"X,a,b,c"
it
is
possible to define the
shorter string .'Y", but
"K,a,b,c;L,d,e;M,f,g',
cannot be replaced by "Z". There is also
a facility (called "hold qroup commands")
for defininq common text sequences
as
a
shorter
string.
There are eight "registers" for this purpose. (Why eight?
Why
not
more?
Don't ask. It probably looked
like a good number to
deal
with
on
an
octal
machine, and there may even be devious machine language tricks which
make
this
an ideal number.) Thus the commonly
used sequence "The
party of
the
first
part', can
be
defined
as,
say, "X,3".
Again, there are no mechanisms for
parameters,
so that
I cannot
have a "The
party of the {n-th{ part" reqister
which
I could
call
up as, say, "X,3,fourth".
These two facilities are enough alike
to
merit, combining into a single cencept -one that has been around for a very
long
time -- the concept of the macro. But not
eight macros, and certainly not
numbered
macros. Macros are at their best when unlimited, named,
and
given
a
parameter
mechanism
(it should,
it
almost
goes
without
saying,
be
possible
to
nest
macro-calls
inside
macros). Add to this
the natural concept
of
a
number
as
a
string
of digits, and macros can be made
to do duty as counters.
(There
is
only
one
counter
in
the DECset formatter: I
don't know
what
one
would
do
if
one
wanted
a page-counter
9dl~ a footnotecounter.)
It
is simple
considerations
like
these, applied at all levels, which
make for a cleaner design, and
that,
in
turn,
makes
the
system easier to learn
and use.

However, we are
skirting
the
real
~ssue.
As stated earlier in this review,
~ext formatting is an activity
requiring
the
strong intervention of human aesthetics. Since the results of this
activity
are
judged on their merits largely visually, it is reasonable to conclude
that
the
production of formatted text is best
a c c g m p l i s h e d by visual means.
Formatting
uools,
like
this
system, and all other
systems of its
kind,
will
fail
to
be
adequate
tools,
no
matter
how well we
r e f i n e them,
until
visual,
interactive
aethods
using
high
quality display and
feedback equipment
(already
technologically
feasiDle) are developed. This must
and eventually will be the
direction
cf
such systems, and when that step is taksn
all such quibblings (over the
design
cf
~etter
dinosaurs) as this review will be
merely

There are, here and there, some good
ideas. For example, the facility for tabular material makes it possible not
only
to
split
the
page into columns, but to
split it into p r o p o r t i o n a t e l y sized
columns.
The
user also
gets c o n s i d e r a b l e
control over the parameters which
govern
the
formatting
process,
and
there are
means, though not always the best
means,
for dealing with a wide range of typesetKing
situations. In many ways the DECset
~ormatting
language
compares
favorably
with
such
languages
as
CypherText and
Harris
Composition
Language.
However,
there
is
the
taste
of
assembly-level
L a n g u a g e about it: the D~Cset
formatting
Language
lacks
even
such
high-level
c o n c e p t s as the "blocks" provided by
HCL
(not
that
HCL
goes
far enough in this
respect).

The T e x t F o r m a t t i n g S u r v e y
Continues
The r e s u l t s o f t h e T e x t F o r m a t t i n g S u r v e y a r e
b e i n g t a b u l a t e d b u t more s u r v e y forms a r e s t i l l
being received.
I f you w i s h t o s u b m i t a form you
are still
e n c o u r a g e d t o do s o .
T h i s r e v i e w i s an i n f o r m a l p a r t o f the s u r v e y z i t
I s a r e v i e w b a s e d s o l e l y on the d o c u m e n t a t i o n , and
clone by someone who h a s n e v e r u s e d the f o r m a t t e r
under s c r u t i n y ,
The p u r p o s e i s t o g e t t h e p e r s p e c t i v e
o f the new u s e r .
Reviews of other formatters will
be p u b l i s h e d i n l a t e r i s s u e s .
Many t h a n k s t o t h e s u r v e y r e s p o n d e n t s .
Your comments
and manuals a r e b e i n g c a r e f u l l y e x a m i n e d .

-9-