GUIs 'forced' on people (was Re: Here's something to consider.)

From: Allison J Parent <allisonp_at_world.std.com>
Date: Fri Sep 18 08:07:46 1998

< I feel that people also dismissed the first Mac for the same reason. (I
< had limited apps to start with, but wider initial acceptance would've po
< changed that.) If they had gone on ease of use, elegance of interface, a
< technological superiority, they wouldn't have bought slow, clunky DOS-ba
< XT's instead. The Amiga suffered the same fate. Drastically superior
< technically, but perceived by the business world as only a toy. Multimed
< for games and kids until Gates and IBM told them that this new "multimed
< thing they came up with was going to be the wave of the future.

Ok, Boys and Girls. Fetch a cup of your favorite drink and follow along.

There were events and products that had to happen before we could make the
quantum leap of faith to GUI interfaces. One of which was a competent and
reasonably inexpensive platform to execute on. this is about technology
convergence on a set of parameters needed to produce a GUI based system.
I will touch on a few of them here.

At the time people were looking for more serious machines the Apples
were in businesses as were trs80s. S100 crates however were acknowledge
as more potent *IF* the user was willing to get into the guts. At the
time of the 5150 intro S100 crates with 6mhz z80s and 8mhz 8088s were
on the desks of some (compupro 8/16 for example) and setting the standard
for performance that PC would not achieve for several years.

The problem as I saw it back then (1980 -early 82)was the number of
possible systems you could buy and their attractiveness beyond
appearance. Apples were fast, had graphics and were fairly easy to
configure and find tons of cheap software. Many vendors also supported
Apples with hardware add ons like the Softcard extending it's performance
and range of executeable software. TRS80s were popular as they also
offered a soft introduction for the hardware timid. Game software may
have factored in the Apple success at the time as well. BUT, they were
8bit systems. Those that needed raw computing power and couldn't afford
the cost of a decent PDP-11, Nova or HP system would bite the hardware
bullet and go S100, Multibus, STD or VME busses for it.They also eyed the
68000 and 8086 series of 16bit cpus for a bit more computational power and
larger address spaces that their bigger programs would need.

The XT at first really was lackluster. It wasn't until it started
showing up with a 5 or 10 mb hard disks and at least 256-512k of ram
that people could see it as a potential engine of greater performance than
the Apple or the others of the time. It offered the prople that
hit the hardware path something less hardware intensive like the Apple
or TRS80 but with the potential of 16bit performance. That last item
was selling hard at the time as the next place to be and it wasn't IBM
alone saying that. Even the S100 oriented knew that already. You see
the idea of graphics, WYSIWYG text editing and other memory intensive
tasks were already being seen in products like DBASE, Multiplan and
others. This extended to push people from the floppy to larger fixed
drives as well. So in the end it was the need for more performance that
would alter the shape of the early 80s computer market with the outcome
fairly uncertain till the late 80s.

There is no small coincidence that when 8bit systems were common storage
of hundres of kilobytes were adaquate. The 8088 made megabyte a word used
and storage was in the same range. The 80286 and the 68000 would stay in
the megabyte range but the software was using more of it and the storage
formerly in the 5-10mb range was now reaching for the 40-80mb or more
realm. The advent of the 32bit 386 and the faster 68010/020 made memories
also move from the 512k to 2mb range in to the serious 2-8mb range with
disks pushing the 100mb region and growing fast this would push RLL, EDSI
and eventually IDE is the quest for more space. One more notch up.

While GUIs aren't mentioned as yet, they were not the standard.
Computer performance and graphics had to come up a notch or two to
support that and offer acceptable prformance. Its offering of a
less complex user interface would take many years to be realized
and those already knowledgable of command lines saw them as slow and
awkard as they sometimes were. It would take one more notch up to
32bit computing to establish the platform that GUIs needed to succeed.
When it happend those timid people looking for a computer solution
without the languages and hardware jumped on the bandwagon. Since
Apple chose the 68000 series there were there a bit sooner and had
polish off some of the rough edges. They were first, best didn't really
matter. It would take the PC getting to the 386 before it had the
performance perceived to be needed to attract attention.

What was missed is that in the time line GUIs needed 32bit cpus. The
VAX, ECLIPSE, SUN and other worstations already knew that but at $50,000
or so they were not in the running. We are talking in the under $5000
bracket throughout this dicertation. Graphics require big address spaces
or at least larger than a 16bit word could easily provide. Motorola was
already there with the 68000 and Intel was still working their way to it
and the 286 a 16 bit cpu was not it yet! So from the time of the PC
introduction to the day of the GUI is long, nearly seven years and lot's
of other things had to happen along the way, each significant in it own
right and also contributing to system as we know them now.


Allison
Received on Fri Sep 18 1998 - 08:07:46 BST

This archive was generated by hypermail 2.3.0 : Fri Oct 10 2014 - 23:31:34 BST