The supermarket industry's call for more open systems has certainly been heard by manufacturers, but in some cases the changes have been more surface than substance. Information systems executives have had to learn, sometimes the hard way, that what's called an open system may be closed to certain types of hardware, software or both.
hardware costs, as products become more of a commodity, and the ability to more easily migrate hardware to a higher level when new versions are introduced.
Following are some leading IS executives' thoughts on open systems.
SN: Has the supermarket industry realized the full benefits of open systems? What more needs to be done?
HOMA:I don't think the benefits of open systems have been realized. It's not so much an open systems issue as a standards issue. The Association for Retail Technology Standards group should take the initiative to publish standards.
Integration is more of a challenge than it's ever been. At the store level, retailers have been selecting best-of-breed applications. That solves a particular requirement, but with very little thought of how that system will be integrated with everything else.
SMITH: The industry hasn't yet fully realized the benefits of open systems. Even so, there has been remarkable progress in the past five years, when there was a lot of proprietary technology being used. Whether you are talking about the front end, or the scales, or any other activity in the store, a lot of that has opened up -- and that's the key to the remarkable progress.
Integration is the problem. That's where we really have to [make progress] through the use of standards; to allow these applications to talk to one another more rapidly and more easily.
We are moving in the right direction. My message to manufacturers and vendors of software and hardware is to continue to move forward in that area. I think we need that as an industry -- to be able to interconnect.
NICHOLSON: One thing to remember is that "open systems" is a term that has so many definitions, it's not even funny. As you go from booth to booth at MarkeTechnics, everybody's system is "open." Then as you explore it, open means that it uses this database tool, or it means that the software would run on somebody else's hardware, but it's proprietary software and you can't make any changes to it. So what open means to people is very different.
SN: If that's the case, how do you measure the "openness" of a system?
DRURY: Open systems will have a number of definitions. For me, open systems means I can take any tool I like that I'm familiar with, and use it against a database or on a network, if it's generic enough or compliant enough with open standards.
NICHOLSON: To me, open systems means that the hardware is off-the-shelf, and the software will run on any off-the-shelf hardware. Very, very few packages, especially when you get into a front-end system that involves both hardware and software, are really totally open. Very few vendors are able to be open across the board.
HOMA: I don't think there is value in open systems for open-systems sake. The sobering fact is that the cost of maintaining these systems is much higher than people realized. Open systems give you a much cheaper hardware platform, but that's only a small part of the total cost. A lot of the tools that have been available on a more proprietary system are not available, so you end up paying either through support, by developing your own tools or through poor service levels. The value of open systems is in the value of standards and finding a way to integrate the data across multiple systems.
SN: What other benefits do you see from the continuing move toward more open systems and integration?
SMITH: I think the technology holds the potential for faster equipment speed, greater data storage capability and lower cost.
DRURY: Using technology, such as frame-relay systems, that is more open and more Transmission Control Protocol/Internet Protocol compliant, enables a company to establish private, virtual networks linking it and other companies together. Open architecture is really what's driving the industry, having manufacturers, suppliers, brokers and retailers talk to each other in very simple ways. With proprietary technology, it was very difficult to have a private virtual network when everybody had a different system.
Use of the Internet and intranets will allow new kinds of information to be exchanged. For example, if a retailer wants to provide a supplier with point-of-sale information for a pay-on-scan system, the cost of doing that over traditional value-added networks is prohibitive because they're sending an enormous amount of data every night from point A to point B. The Internet and intranets are much more efficient and much less expensive. If you use the Internet, which is not what we're doing, you literally can do it for free. There are whole new ways of doing business that wouldn't be possible using old technologies.
NICHOLSON: More open systems would bring two very specific advantages. One is cost. When hardware is truly open, it becomes a commodity, and is priced as such, as opposed to a monopoly where you have to get it from one vendor.
For example, we have a front-end system where the disk drive has to come from the maker of the hardware. We can't go and get a PC off-the-shelf disk drive and put it in there. That disk drive costs us roughly ten times as much as an off-the-shelf disk drive would cost because there's only one source for it, and production is limited.
The second thing is migration. You buy a piece of hardware today and it'll be obsolete in six months -- if you're lucky. As new innovations come along, if that's a piece of hardware where you literally pulled components off the shelf and put it together, you can migrate yours in the same manner, just by pulling out the old disk drive and putting a new generation disk drive in.
