The New 16-Bit Operating Systems, or, The Search for Benützerfreundlichkeit

Editorial

by Chris Morgan, Editor in Chief
BYTE

June 1981

"Benützerfreundlichkeit: (literally 'user friendliness') The philosophy that a system should be constructed with the interests of the user as the chief concern."

—from The Practical Guide to Structured Systems Design by Meilir Page-Jones, Yourdon Press, New York, 1980, page 338.

Sam Goldwyn, the "G" of MGM, was famous for his inside-out logic. He once said, "A verbal agreement isn't worth the paper it's written on." This month's topic prompted me to coin a "Goldwynism" of my own: "The best time to talk about the future is before it happens."

In one sense 16-bit microcomputers are definitely here, yet in another they are strangers to us. The personal-computer community still lives in an 8-bit world, straining all 8 bits of every word to perform miracles.

But all that can and must change. Opponents of 16-bit systems cite cost and software conversion problems as the two main justifications for staying with 8 bits. Yet, how can software keep pace with the increased demand for more sophisticated graphics, to name only one area, unless we can address more than 64 K bytes of memory? How will we be able to access the staggering amounts of information in future memory banks without an increase in word size? And then there are the exciting new languages like Smalltalk that demand 16 bits for their operation. Simply put, 16 bits is the only way to go. The 16-bit operating system, therefore, becomes a critical link in the computing chain.

Doing It Right the Second Time

The operating system is the "master controller" of the computer: it gets us going when we turn on our computers, keeps track of files, lets programs talk to one another, performs input/output tasks, and so on. Put charitably, most operating systems in the 8 bit world have been afterthoughts or compromises in design. Even CP/M, a de facto standard in our field, has been criticized as being awkward for nontechnical users. But CP/M's ubiquitousness is responsible for the development of a lot of valuable software that would otherwise probably not have been written.

The sin of inefficiency is venial compared to the mortal sin of "user-unfriendliness." I'd buy an operating system any day that takes a long time to run a given program but which makes me more productive by communicating with me in useful ways. Let's face it: most of us don't have to worry about real-time process control and its inherent time constraints. And the cost of a line of code is becoming astronomical.

Now we have a chance to start with a clean slate. Software manufacturers are filling their 16-bit tabula rasas with offsprings of UNIX, an operating system developed at Bell Labs in 1969 by Kenneth Thompson and Dennis Ritchie. (See Robert Greenberg's article, 'The UNIX Operating System and the XENIX Standard Operating Environment," page 248.) A software engineer was quoted in a recent issue of Electronics magazine (March 24, 1981, page 119) as saying that UNIX is "like sitting behind the wheel of a well-tuned sports car—when you press the gas, it goes, and when you hit the brakes, it stops. It's the ultimate in responsiveness, and yet all the while you are riding in comfort." UNIX deserves such accolades. Its hierarchical file structure lends much needed order to the chaotic approaches found in many personal computer operating systems; it is designed for truly efficient multiuser operation; the elegant idea of the pipe allows data to flow from program to program efficiently; and the shell program acts as a user-friendly interface to the rest of the operating system. An excellent example of UNIX's versatility, described in Greenberg's article, shows how the user can add a simple spelling correction program to a system, with just one line of code,

New Programs

Several software vendors have taken out licenses to adapt UNIX to 16-bit personal computer systems. These include Microsoft, Whitesmiths, Zilog, and Onyx, the developers of XENIX, Idris, Zeus, and Onix, respectively. Among non-UNIX related 16-bit operating systems, OASIS, developed by Phase One Systems Inc, has received high marks from many professional programmers.

And judging from its past track record with CP/M, Digital Research's new CP/M-86 should also become a major factor in the market. (See "CP/M: A Family of 8- and 16-Bit Operating Systems," by Gary Kildall, page 216.)

Despite the recent relaxation of UNIX licensing fee conditions by Western Electric, the UNIX offspring will not be cheap. Operating system software could sell for more than $2000. However, Lifeboat Associates' version of XENIX will probably retail for less than $1000 by the end of the year.

The 8-bit computer is far from dead. There is too much good 8-bit software around for this to happen. And, for many applications, it's hard to beat the price-performance ratio of the 8-bit machine—at least by today's prices. Sixteen-bit and 8-bit machines will coexist for many years to come. I don't believe in the "mutually exclusive" school of computer punditry. Just as no high-level language has ever supplanted another (can readers give me an example of this?), 8-, 16-, 32-, (etc) bit microcomputers will coexist in the future.

In our field, the future becomes the present overnight, You don't need a crystal ball to state emphatically that we have not seen the end of the 8-bit versus 16-bit debate. But the new operating systems do add a welcomed layer of professionalism to personal computing.