Re: Computing Editorial (please read)

[hjohnson@pluto.njcc.com (Herbert R Johnson)]

In article <b2dpvo$3p6$1@freenet9.carleton.ca>,
et472@FreeNet.Carleton.CA (Michael Black) wrote:

*>Yes, thirty years ago many people trying out microprocessor based
*>computers were new to computers, because suddenly they had computers
*>accessible to a far greater extent than previously.  But, I also
*>imagine many starting out with them came from a computer background.
*>....they were in a different space from the average person....
*>
*>..back then the CPUs and computers weren't necessarily simple for beginners,
*>they were simple because that was all there was.... everything has to go
*>through a number of iterations before it becomes complex.
*>
*>Take a KIM-1, really quite simple compared to today's computers.
*>There were things in it helpful for someone trying to learn
*>about such things.  But unless one was going to pay big money
*>for all the extras (memory, video, ascii keyboard), you could
*>only stick with something so simple.  And since you didn't have
*>the hardware to run fancy programs, that too was out of the picture.
*>
*>It only incidentally was helpful in getting people under the hood.
*>
*>  Michael

In my opinion, Michael simply misreads history. His comments above mix
up what was available when, and have many implicit assumptions.

He's simply wrong that the KIM et al were made "incidently" to
get people under the hood (familiar with microprocessors). Buying
a single card all in one computer was the COMMON way you could learn
from experience. The alternatives? - access to a $10K development system,
access to expensive timesharing services or a minicomputer; reading some
books (not hands-on which was critical); access to engineers and some
hobbyists. And NONE of that by email, no downloads, no Web pages,
and only college bookstores (or chip manufacturers) to even get those
books! Most microprocessor manufacturers offered their own "KIM's"
as trainers. They were offered as learning tools and development tools,
as well as for use in one-off control applications. They were CHEAP
compared to alternative systems!

Briefly, he also seems to have the opinion that you could not do
anything complicated with such simple microcomputers; and that
microcomputers needed to
get to a level of "complexity" (or you had to have a lot of technology)
before you could do anything complex. He also seems to make the modern error
that a computer without video and full keyboard is not useful or
sufficient, at least for the kinds of task he sees TODAY as signifigant.
He does not say these things literally but his remarks suggest
such an attitude. In any event such an attitude is prevalent today.

I'll grant that simplicity is relative to the times, the person, and
the available technology. But you can - and we DID - useful and complex
systems "back then". One should keep in mind that 20-30 years ago, there was
no need and no great value for audio/visual interfaces (GUI) in many settings
where they are in use today. And the design and development tools of
that time certainly did not use such capabilities. The fact that in
earlier times the technology to do so was not available or only in
Xerox PARC certainly is a sufficent reason for that. But even after
such capabilities were available, they were not used for several years
by many users and developers. GUI's use a lot of computing resources,
while not necessarily contributing to the performance OF THE DEVELOPMENT
TOOLS. I'm not arguing about the performance of the developer, as the
issue is not efficiency but results.

Technically proficient people use the tools available to produce
extraordinary results. Hey, "that's why they are paid the big bucks."
That is precisely the nature of expertise in production.

However, in brief, the proliferation of powerful, graphic-interfaced systems
allowed untrained and inexperienced people to use computers. And
the availability of software to support what were non-computerized but
information-driven tasks, gave these non-computer-literate people reasons
to use computers. This is the argument for performance and general
access by the "average person".

All of that does not mean that one cannot do reasonably useful things
with a computer like the KIM-1: a card with some memory, a HEX keyboard
and a hex display, with audio cassette tape file storage. People of the
time made control systems with such cards, just as today people control
things with BASIC Stamps and PIC processors and other similar one-card
microcontrollers/computers. A PIC replaces that hex keyboard and display
with a PC's screen and keyboard, and I'd casually say they are "comparable"
in terms of function, use, and resource. (the PIC is a lot cheaper and
smaller which makes it more "popular" in some sense.)

As regards the "average person"; that person almost by definition is one
who has no particular skills or knowledge base. I would no more expect
the "average person" of 1978 (about when the KIM was available) to be
able to use a KIM, or even imagine WHY they would want to; than I would
expect the average person of 2003 to know how to program a VCR to record
a program every Wed at 8PM for an hour. A relevant difference between 1978
technology and 2003 technology is that the "average" person now has MUCH
MORE ACCESS, MUCH CHEAPER to comparable technolgies. So that "average
person" TODAY can probably go around his/her home and find several things
that they could, if they wish, learn to "program" from thermostats to VCR's
to of course their Web-surfing computer.

I'm not closely following this discussion, more-or-less centered on how
people USED to learn about computers versus how they CAN or COULD learn
about them today. But I'd like previous history to be described better
than what I"ve read here from some correspondents. Seems to me it could
be relevant to how one might make a "KIM" (or another BASIC stamp) today
if that is also a goal; or how those old computers were used yesterday.

ONe closing comment. Brian Kernighan (of Kernighan and Richie C language)
spoke in a recent lecture of the "tyranny of the GUI" or words to that
effect. He considers a GUI (mouse and icon) based program very hard to
automate. And since he can't automate it, his productivity is limited
by his ability to point and click. Whereas he likes and prefers languages
like TKL and TK which provide means to automate tasks and string together
programs. He considers the "pipe" feature of Unix (connections between
otherwise independent programs) a critical innovation and a key to its
success and usefulness. That allowed collections of programs to work
together, AUTOMATICALLY (without user intervention) yet be written
and developed seperately and independently.

In other words, sometimes "complexity", and especially a GUI,
gets in the way.

Herb Johnson

Herbert R. Johnson                     voice 609-771-1503, New Jersey USA
                http://njcc.com/~hjohnson  is  my web site
          email address: hjohnson@njcc.SPAMcom <--removespam
                  good used Apple Macs, SGI's, printers
S-100 IMSAI Altair type computers, drives, manuals, parts from "Dr. S-100"
-- 

Herbert R. Johnson              http://pluto.njcc.com/~hjohnson
hjohnson@pluto.njcc.com         voice 609-771-1503, New Jersey USA
             amateur astronomer and telescope tinkerer
   reseller of classic Macs & accessories from Plus to PowerMac
   S-100 & 8-inch drive manuals and parts, call for "Dr. S-100"