Monday, December 2, 1996
Was It Just Smalltalk
or
A Window Into The Future?
By MF Runkle
In my last two Fribbles I wrote about the failure of one company (Commodore), the success of another (Microsoft), and the way another company (Digital Research) missed the business chance of the century. This Fribble will write about the success of an idea rather than a company or product. That idea began to crystallize at Xerox's Palo Alto Research Center (PARC), and ultimately would lead to the Macintosh, OS/2, and Windows. It was a programming language called Smalltalk.
Early in the days of computing, data entry was very difficult. When I was in college (not very long ago), computer science class meant long hours at the Computer Science Center feeding piles of punch cards to the Sperry, praying the program would work. Many times I was stuck there at the University of Maryland, desperately trying to find a bug in my FORTRAN program at 2 a.m. The problem with computing was its lack of interactivity. You wrote your program in a stack of cards, had your data in an additional stack, fed it in as a "job", then two hours later you got a couple pages of paper back with nothing on it but error messages. FORTRAN was very fussy; everything had to be in the right column, and the data you fed in had to be just so, as you specified it in your FORMAT statement. Oh boy!
Along came the personal computer. My first computer, the Compaq portable, was great. Here you had interactive computing. I could write a BASIC or Pascal program (I swore off FORTRAN), and test it immediately. Better yet, I didn't have to stand in line at some university waiting to feed my "job" into a mainframe at an ungodly hour. The computer was now perfect! Or was it?
Well, back in the 60s, Alan Kay had a vision. He had the idea of a computer called the "Dynabook." It was to be the truly personal computer. His first attempt at this was the Flex machine. In the early 1970s, he went on to PARC, and Smalltalk was born. As with many good ideas, this was built on earlier innovations. Smalltalk incorporated many elements of Simula and LISP. All of the technical literature I've read about Smalltalk raves about its "object oriented" design. This whole concept is important for Windows and the Mac, but what is more important was the graphical interface.
The early PCs still imitated that one-line-at-a-time system of the teletype terminal. Sure, with some neat programming tricks you could use the whole screen, and make it look graphic. However, your data still went in one line at a time, and your computer processed it much the same way the mainframe did my punch-cards. Smalltalk went one better. It used a graphical interface, and this thingy called a "mouse." Your program came up on the screen in these things called "windows," and a number of them could be opened at the same time. Pretty nifty idea, huh? Here we were using the screen we had available, not just imitating a teletype machine.
In the spring of 1979, Steve Jobs, the co-founder of Apple, visited Xerox's PARC. He was shown Smalltalk and was converted. The idea of a graphical interface with a mouse impressed him so much that he decided that Apple follow suit. The first attempt at this was the Lisa, which kind of crashed and burned. Then, there was the Macintosh. Andrew Groves, CEO of Intel, saw the Mac for the first time in 1984, and he thought it was a "ridiculous toy." He says he didn't see until later that this was what he calls a "10X" force, one that would change the face of computing. Here all applications had a uniformity of interface. There would be fewer complex commands to learn, and the basic functions of all programs would be the same. No longer would you need to spend an hour studying the user's manual to figure out how to save a file.
The rest is history. OS/2, and then Windows followed. Now the old DOS interface is considered to be almost as quaint as those crummy punch cards at the University of Maryland. As for Smalltalk, it is still going strong. I did some Web surfing when doing the research for this Fribble, and found that this language is still very much in use. Maybe Smalltalk didn't fail because it wasn't a business venture, but an idea, or a vision. It changed how we think of computers in only a few years. Where will it bring us in the future? I don't know; it's hard to see out that window.
Transmitted: 12/2/96