Ah, the buzzword of the 90's. What does it really mean? While the answer to this question will depend a little on your viewpoint, the truth is Open Systems are both more and less than you might expect from reading the trade press.

Open Systems are not new. They are the logical (if only partially achievable) end product of standards efforts begun years or even decades ago. Why are they the logical end product? Well, standards are an attempt to define something clearly enough so diverse people can understand and take advantage of the standard in widely varying environments. The original standards in data processing were mostly physical and electrical standards such as the RS232 electrical standard for terminal (serial) connections. After that have come software standards such as FORTRAN 77, ANSI COBOL, the Pick SMA committee, and others. The problem is these software examples are consensus standards and there isn't a good consensus yet about what makes a "system" "open".

Lets try to create a working definition of "Open Systems". The first problem in defining the term comes when you realize all the companies promoting Open Systems today are talking about systems that are in fact proprietary! You heard me -- look around. UNIX, Oracle, Pick, Windows and others being widely touted as "Open", however they are owned by someone that wants to protect them. This is the definition of proprietary!

If this is true, then what do these people mean when they say Open? Well, today the street definition seems to mean "runs on a lot of different hardware in a similar way." To distill it a little further, most vendors mean your applications and data can be moved to a wide variety of other hardware at minimal cost. Note, minimal cost is not usually taken to mean zero cost -- there is almost always a degree of conversion involved. As a side note, by this pseudo definition, Pick was Open before the term was coined. Just because this definition is much narrower than you may have thought doesn't mean it's all bad -- in fact its a significant improvement over the old days. Those of you as white-haired as I remember the days when changing to another model of computer from the same manufacturer was a real chore. Manufacturers had nearly literal locks on their customer bases because the energy threshold to move your application to hardware from a different manufacturer was nearly impossible to surmount.

So, while the term "Open" is currently still pretty fuzzy, it is clearly mostly good for you and mostly bad for proprietary vendors. This would make for a slow closure rate on a better product definition if it were up to the hardware guys! Fortunately, most of the progress in Open Systems today is on the software front as the physical standards such as RS232, X25 and Ethernet have existed for some time now. The software people such as the MultiValue community, Oracle, and Microsoft have a different vested interest than the hardware people. While they still would like you to use their products over all others, they don't care about which hardware platform you use. Their competitive formulas favor openness as a strategic goal not for your sake, but because it improves their market penetration. Reasons aside, its still a win for the end users.

Lets look at where the term "Open" is going. Open is coming to mean not so much application portability as data portability. Email gateways, the X400 document interchange standard and EDI were the leading edges of these kinds of openness (past tense only in terms of understanding -- lots of implementation pain still being experienced!). True data base interchange is the back edge of the wave. By interchange, I don't mean the old batch types of interchange. I mean real time, interactive access to multiple data bases from one query. The old "TWA wide band" (if you remember that term, you're a computer historian) batch transfers have given way to SQL queries polling multiple data bases on multiple on-line systems and network file server software using PC data bases. A local example is the newest releases of Advanced revelation that can collect and update data not just from a Pick data base, but from five of six other PC databases simultaneously. Similar data interchange functionality is planned or available from many of the Pick like vendors.

This is the dawn of a new age. A real paradigm shift. Once in full force, this new model of interchange will usher in the real age of "Open Systems." Open in the sense that an application won't need to own and control the physical representation of data to make use of it. It won't even need to know who owns the data. "Open Applications" is a better term as the tie to one piece of hardware or one operating system will be gone forever.

How will these changes affect you and your shop? In truth, many of us will still be running our old Basic applications as our business mainstays. Most companies have too much invested in the software and it is too critical to the company to just dump and replace it all. What will change is the introduction of a new spirit of cooperation. I believe that four or five years from now many MIS managers will be running shops with traditional Pick applications running side-by-side with a wide variety of new applications from different sources running in new and different environments.

We already can see the start of these kinds of cooperative changes in the Pick world as DOS, SQL and Windows gateways are coming on line. You will be freed from having to choose one system for all you data processing needs and will be able to pick (no pun intended) the best resource to solve the current problem. In this environment, applications will become more sophisticated and easier to bring up.

Of course life may not be all peaches and cream, while the competition for best all-round system and with it the stigma of Pick as an operating system may be gone, gone also will be the single point of control and responsibility for data integrity. Administration of these mixed systems will become increasingly difficult. When you don't know who owns and controls a data base, you don't know how current and correct the data is. Life will be different but perhaps not carefree.