I've been around technology all my life.

I've developed a pragmatic view of the computer technologies that I've dealt with for most of my life. It might help you to learn how I got there and how I see the future unfolding. There are lots of computer consultants that don't share my views, but then there are a lot of computer consultants that started in this business at the time of the IBM PC. Today's PC revolution has been a long time coming, largely due to some of the stories of the failures in the early years. It's my contention that there are still a lot of things that need to be resolved before we can truly say we have won the revolution.

I started taking old tube radios apart before I hit my teens. In my grade 9 electronics lab I learned how to make things like amplifiers and such using tubes like the venerable 6L6, still used in some of today's guitar amps because of the sweet sound such amps can have. By grade 10, some of the initial solid state components became available to hobbyists. I experimented with transistors, tunnel diodes, SCRs and Triacs to create everything from light controllers to radio controlled boats and cars.

I remember the first "computer" training hardware I saw. The Vancouver school board paid over $40,000 for a cart about 4' long by 3' wide and 5' high with just enough logic on it to create an 8 bit binary ring counter. There was only one for the whole school district, and we had it for a whole 2 days.

During grade 12, I got an opportunity to visit the education faculty of UBC several times and work from the Remote Job Entry (RJE) station there with the IBM mainframe - one of the first IBM 360s if I remember correctly.

Outside of electronics, I fixed cars, played with trains, built all manner of mechanical contraptions, experimented with hydrology and chemistry, and all in all had a wonderful time.

Today I sit in front of a P200 with 128Megs of RAM, an ATI All-in-Wonder video capture/TV card, 6 Gigabytes of local disk, a SVHS VCR with an 8mm Sony handycam Hi-8 connected to it. My home LAN has a P200 Pro Linux server with 14 Gigs of storage and 2 tape drives. Even my watch is a Timex data-link. I'm connected to the Internet 24 hours a day, have 40+ channels of TV at my fingertips, CDs, radio, video tapes and 4 phone lines.

Putting Technology in Perspective

Technology is a means to an end. Technology is the means to make us more productive so that we don't have to spend all of our time scrabbling just to live. It started with the domestication of draft animals so we could cultivate more land. It continued with the steam engine and the cotton gin and the printing press. The pace of the technological evolution has previously been slow enough that changes took generations, and so were fairly well integrated into the daily life of the individual. Today however, technology is advancing so fast that most people simply can't keep up. In the blind fear of being left behind we leap from bleeding edge to bleeding edge, careless of the potential for disaster.

The question is, "What is it getting us?". I hope this article will bring some perspective to the question.

A personal bit of history of computers in business

I'll focus on computers in the business world since I've been dealing with this for over 20 years now.

In 1977 I was working for Dataline Systems Inc., a timesharing computer services company with a stable of Mainframes (DEC 10s and a 20) that we sold time on. Prior to this I worked for Bell & Howell creating forms based order entry/inventory systems. One of the products Dataline sold was an inventory system that was designed to track goods not only in distributed warehouses, but actually in the boxcars and trucks while it was in transit. The system used printing terminals for data entry and took weeks to learn because of the arcane commands used to conserve the precious computer resources of the day. The system allowed a national food company to become one of the premier distribution companies in Canada because of the national infrastructure of Dataline. This was a good thing.

At the same time, we in the Vancouver office of Dataline had a secretary/receptionist who typed all our reports and letters on an IBM selectric. While I was there, we actually created a facility to do basic word processing and print to a daisywheel printer from the mainframe - really just a primitive text editing system, but a portent of things to come. As a salesman, I spent most of my time in front of customers since I wasn't expected to do my own correspondence and filing. About the same time, IBM and Wang and Xerox were fighting for the dedicated computer based word processing market. Large legal and accounting firms were changing their steno pools over to Dictaphone with centralized minicomputer based word processing systems. The printers used were really glorified typewriters - faster, with changeable type faces, but not much different from the IBM selectrics that were the latest thing in electric typewriters. In fact, one of the last modifications done to the selectric was to give it an interface to the computer. I even saw a strange device that was meant to sit over the keys and pressed each with a solenoid driven plunger.

We still had the executives of the companies devoting their time to creative thought and passing the drudge-work to a pool of lower paid workers. At that time, the computer was worth far more than the annual salaries of the people who programmed and used it. A basic word processor cost on the order of $40,000 to $100,000 while those who used them were getting maybe $10,000 to $15,000 a year.

In 1981 I started selling microcomputers at the Radio Shack computer center in Vancouver. The store had only been open for a year or so, and a business I had been in had been a customer for a Model 1 to do its accounting. I was impressed with the abilities of this revolutionary machine so much that I left the stereo business I was in and went to sell computers. The system that the stereo store had cost about $10,000 with all the software, and easily did all of the work of a $100,000 mini computer of the day. It had a fully functional A/R, A/P, G/L and Payroll system, all in 48K of RAM and 300K of disk, the maximum the system could handle.

I quickly found that the typical business customer for the systems we sold was a complete neophyte in office automation - in fact was more of a technology hobbyist. They'd purchase a minimal system, even ones with just the little audio tape drive as the only long term storage medium, and expect their poor receptionist to use it to do mailing lists, type letters, and maybe even do some accounting.

What really happened was that the poor girl (sorry, but that's the way it was, most of the receptionists of the time were women, typically fairly young) would be told that she should enter every telephone caller into the mail list program so that literature could be sent out. She would also be asked to type letters using the word processor program in between phone calls. The problem was that it took upwards of 5 minutes to get out of the word processor and into the mail list program. The systems simply were not able to do more than one thing at once. The poor girl would end up making notes about the phone calls and have to put them into the computer after hours when the typing was finished and the phones were quiet. The boss would get upset because the mail list wasn't up to date, and the level of frustration in the office grew. I saw it happen.

About 1983 the Radio Shack Model 16 came out. This was an adaptation of the Model II where an extra couple of boards were installed that turned the system into a multi-user (and multi-tasking) computer running the XENIX operating system from Microsoft (yes, Microsoft - more about that at a later time). This got the microcomputer up almost to the level of the time sharing systems I had been selling in 1977 in terms of real productivity. The systems cost around $15,000 and up, but they rivaled the minicomputers of the day in the $100,000 plus range. It was possible to have 3 different people all working at once on the system. It was also possible to switch back and forth between applications such as Scripsit word processing, Real World accounting, and Profile 16 data base without taking large amounts of time. The productivity of the reception and accounting people went up and the boss started to have a terminal on his desk too.

At the same time the IBM PC came on the market along with the Apple Mac. These systems were less expensive than the Model 16 and so many people purchased them instead. In fact, many people simply purchased several of them to allow more than one thing to be done at a time by different people. The fact that there were now offices with several computers in them engendered the concept of print and file sharing - networking (and a small company called Novel was born to fulfill the need).

Disaster Looms

Up to about this (late '70s to early '80s) point in time the use of computers was largely done by people who had been trained in dealing with the vagaries of reliability and the possibilities of massive loss of information due to mechanical and human error. Mainframes had backup systems including separate power supplies with battery backup, tape drives with the backup tapes stored off premises in secure areas, and physical security of the actual hardware installation.

In the early 1980s a new generation of computer users learned the hard way about the needs for such mundane things as backups and physical security. All that wonderful productivity gained by capturing information once at the source (the phone caller's info going into the mail list database, the sales receipt going into the G/L, etc.) went out the window when the only copy of the information was trashed by spilling coffee on the diskette, using a magnet to hold it on the side of the file cabinet, crashing of one of the days primitive (and very expensive) hard disk drives, or the theft of the computer itself.

The hang-dog look on the face of a previously exuberant business owner coming in the door with the offending diskette or hard disk almost became laughable. The panicked voice on the other end of the phone pleading for quick delivery of a replacement system became commonplace.

Business users quickly learned that they needed backup systems and trained people to make sure things didn't get out of hand. The cost of using computers turned out to be far more than the initial purchase price of them, a fact that large system owners have known for decades. But there was still that pot of gold at the end of the rainbow called efficiency. We strove for the goal of being more efficient than our competitors, reading all of the marketing bumf that the burgeoning computer industry started to push on us.

Of course the other side of the coin was almost as bad. Even if the computer actually functioned and didn't get stolen, it quickly caused as many problems as it cured in many cases. The demand for absolutely perfect correspondence actually dropped the productivity of the typist in many cases. Many non-typists started to hunt-and-peck their way to digital correspondence, further lowering the productivity of the office as a whole.

The "being nibbled to death by ducks" syndrome took on whole new meaning.   The computer required supplies that were expensive and not readily available (at that time). This lead to instituting some interesting false economies like re-using diskettes and creating "flippies" (floppies with an extra notch cut so you could use both sides). A mail list started on the computer was corrupted for some reason, resulting in having to re-enter all of the items from the last print-out (kept solely for the purpose - what a backup method!) Overtime caused an evaluation of whether a second computer might not be a better bargain, and the whole process started over. Throw in the incompatibilities, software upgrades, hardware failures, poor power, and all of the other problems that the small to medium business community had never had before, and the result was all sorts of chaos. Some businesses simply went bankrupt - I saw it happen.

Aside on reliability

With all of the development pored into the computer systems of today, you would expect that today's systems would be far more reliable than those of 15 years ago. I don't think it is so! I see today's systems crash almost as often as the systems of the early '80s, at least in relation to the time they are in use. My Model 1 accounting system ran for days at a time without crashing. My current desktop system crashes about every second or third day. What is wrong with this picture?

My current hardware is pretty high quality. In fact, it is the same for my desktop as it is for my LAN server. The interesting thing is that the LAN server has been running for over 6 months now without a re-boot. The difference is that my LAN server is running Linux and my desktop is running Windows 95.

The economics of the computer revolution

Back in 1983 I discovered the XENIX operating system when Radio Shack brought out its new Model 16 computer. XENIX was a commercialized version of AT&T's UNIX operating system, produced by a new startup company from Redmond Washington called Microsoft. There were only a couple of problems with it: AT&T got a license fee for each copy sold, and it required far more memory than the typical micro computer of the day could hold, at least 512K if any amount of work was to be done. At that time the typical system maxed out at 64K and most shipped with less than 16K. RAM was also quite expensive; several hundred dollars per K (i.e. thousands per Megabyte where today it is less than a hundred for 16 Megs)

If XENIX, or some other multi-tasking, multi-user system with similarly large RAM requirements had been adopted as the most reasonable small computer operating system the industry would never have taken off the way it did. Instead, the continuation of the single tasking program loaders of the time kept the price of entry down and sold lots of systems. The adoption of business computing probably would not have happened quite so quickly, but then it probably would have had more visible benefits in my opinion.

Today, we have a large segment of the PC market dominated by the WINTEL (Windows and Intel) technologies - Intel processor based hardware with Microsoft operating systems on them. From the suppliers' and consultants' point of view this is wonderful. The hardware keeps getting more powerful, but the software keeps needing the power, so the upgrade spiral is generating more and more sales. The Windows operating environments (95/98 and NT) appear simple on the outside, but have so many "gotchas" that it takes a specially trained consultant to bring them up to any semblance of functionality and keep them there. The training is another profit center for Microsoft, and is a wonderful differentiator in the consultant's marketplace. Windows NT and the Microsoft applications that run on it are so inefficient that the hardware resellers love it because any system that will be called upon to do much more than display the login screen will have to be one of the premium systems, and the only way that the hardware/NT mix will have even reasonable reliability is if the OEM vendor does their own tweaking of the drivers and installation. Everyone sells their premium service and gets the maximum profit out of the poor customer, and the systems still crash far too often to be a joke.

As the cost of the various hardware components has gone down or their capabilities and capacities have grown, the bloat of the operating environments and programs, and the need for highly skilled help seems to remove whatever speed and/or flexibility might have been gained, and the cost of all the upgrades seems to be in direct correlation to the decrease in the cost of the hardware. The US Federal look into the marketing practices of Microsoft seems to point out that there is room for improvement somewhere.

I think it is time to revisit the cost/benefit picture for PCs in business.

[ Home ] [ Contents ] [ Search ]
[ Up ]

Copyright © 1993-2007 Richard C. Pitt - all rights reserved
Updated June 17, 2005