What Are The Issues Facing End User Hardware Vendors in Building ISDN Business Cases? by Dennis C. Hayes President, Hayes Microcomputer Products, Inc. ISDN Forum Washington, DC 9 - 10 July 1991 I am very pleased to have this opportunity to speak to all of you. I'm sure many of you were at the workshop in April held in Atlanta or have read some of the articles that came out of that planning session. You'll probably remember my spending a lot of time talking to you about the perception of ISDN and how that perception has changed throughout the last few years. I also talked about how ISDN is made up of two industries: the computer industry and the telecommunications industry. Each industry has its own different culture and its own different concept of time. While the telephone companies have been working at what they thought was breakneck speed to make the biggest change that has ever been made in the telephone network, the computer industry became frustrated and began to wonder why it was taking so long. The peak of expectation for the promise of ISDN became, after a few years, the valley of disappointment. With meetings like this forum and the work being done in the various standards bodies and the real capabilities of ISDN today, we are once again climbing on this chart. I spent much of my time at the COMDEX show in Atlanta and the PC EXPO in New York recently explaining to computer media editors and executives why the computer industry got frustrated and is now blind to the advances being made in ISDN. Our task is to reset expectations so our customers perceive the reality of what is possible within their planning horizons. Let us not create another spike of expectation but rather a realistic view of what can be done. And what can be done now. Recently I had the opportunity to meet with data communications managers and MIS directors of major corporations in the New York area. I used this opportunity to learn of their plans for ISDN. Out of the 20 companies, about seventeen of the companies said that they would be implementing ISDN. One said a definite "no." Two said "hopefully." When I asked them WHEN they plan to implement ISDN, the answer was not in one year or in two years -- it was when all the pieces pull together and service is available everywhere. And, they had no real understanding of how their companies could benefit from ISDN. They were looking for assurance about all the pieces. Each of us in this room represent a part of one of the five pieces that make up ISDN in North America: telecommunications technology, computer technology, service providers, computer distribution and sales -- and most importantly, the customer. I have been asked to address today the issues facing my piece of that puzzle as a manufacturer of CPE/ISDN adapters. And I can only talk about it out of my experience. I am not sure that "issues" is the right word. For me, it has been the decisions facing me in building an ISDN business case. These decisions have been based on the issues facing my customers. I cannot make a good business decision without knowing what my customers need. These decisions have been focused on three major areas: customer needs, products and technology, and stages of ISDN deployment. And, I will be the first to admit that they have been tough decisions to make. The first decision I had to make was: "What business am I in, and given that business, what do my customers need or want from me? In other words, what do my customers need to do their job and to be competitive in the global marketplace?" The market has been telling me that it needs more speed. The speed of the typical modem available to use on a PC has doubled every 18 months. In ten years the speeds have gone from 300 bps to 38,000 bps. In February of this year the CCITT passed the V.32bis standard for modem modulation and just last month we introduced our ULTRA 144 modem that supports this standard. With the 14,400 bps line speed we are reaching the theoretical limit of the analog phone line. We are approaching Shannon's law and we won't be able to go much faster over dial-up lines. To continue to get significant increases in speed we must move onto the first step of the digital ladder of communications. CCITT V.Fast is probably the last modulation standard for modems and the last step for the analog data world. This standard will take a year or more in the standards body before we know exactly what the modulation will be. The speed increase will only be about 33 percent, not the usual doubling or quadrupling that customers are accustomed to. Given this, at Hayes we realized we had three ways to approach the customer need for high speed communications. First, we could approach the situation as a modem company but that solution would be nothing more than a high speed "ISDN modem." Then, if we approached the customer need as a voice company, we could provide customers with a telephone set that allowed data transmission. And third, when we approached this need as a computer communications company, we realized we could offer more. I know that people have seen Hayes more as a modem company but for a long time we have seen ourselves as a computer communications company. As a computer communications company, when we looked at this customer need, we began to realize the full capabilities of ISDN and made the decision to support fully its integrated voice and data capability when designing products. We are not only able to create products that provide the customer a high speed replacement for a modem or a wide area connection between local area networks or a remote node on the LANI but we can also provide the multimedia transmission facility. The computer industry desperately needs to begin exploring the transmission of multimedia and to deliver to this rapidly increasing market over the next few years. It may not be obvious to everyone at this forum, but at Spring Comdex, multimedia software was the main interest for almost all attendees. Customer applications are changing. Data files are getting larger -- pushing the need for higher speed. Applications have more pictures, more images or fax, more occasion where video or voice are included with data. And customers want to use these applications on local area networks. Today's modems can connect a remote workstation into the network but don't have enough speed to make available the full resources of that network. So, I've talked about my first decision which was "what do my customers need." Now that I know what my customers need, the second decision I had to make was, how do I provide the products to meet their needs? In order to determine what was best for our customers we had to make the investment in research and development. Once again, a tough financial decision, but one that was necessary. Hayes began ISDN research and development as early as 1985. Two years later we demonstrated our technology at the International Switching Symposium in Phoenix. We conducted an early placement program where a few of our customers used our technology in various ISDN trials so that we could learn about the user environments. Today we provide training seminars and extensive developer support programs among a list of many activities that we learned were necessary to empower the technology and make it commercially viable. What was critical in this was we had to decide to go well beyond developing a product. So, we put in place what we call the technology empowerment. Because we have a large customer base using Hayes modems we did not have the luxury of beginning our ISDN development as though we had a blank page. We had to provide a platform for compatible deployment and transition from modems to ISDN adapters. Part of our original work was done in developing the working hardware interfaces: the Hayes Standard AT Command Set for ISDN and Hayes ISDNBIOS. These are interfaces that software developers could use to develop customer applications. Hayes had the expertise of data communications and we called on our relationships with companies like Siemens, GPT and Northern Telecom to provide feedback and expertise on voice communications to be sure that we were providing comprehensive voice support. The Hayes Standard AT Command Set for ISDN and Hayes ISDNBIOS are hardware interfaces that will be able to map into the higher level APIs being developed at NIU, at France Telecom with the PCI project, and at Australia Telecom -- just to name a few. Eventually there will be standard APIs to make it easier to have application transportability between systems, but in the meantime, we can still get the job done and transition our customers to ISDN because we have these working hardware interfaces for developers. Last year we made our first commercial ISDN product available to the market. Today we have two hardware products that support ISDN Basic Rate. These products are sold through the same distribution channel as we sell all of our products in North America and are compatible with existing communications software used with our modems. In addition, we have jointly developed a product with Northern Telecom which they now sell in Canada. We also have had over 12 developers who have announced new software applications written to support the Hayes Standard AT Command Set for ISDN and Hayes ISDNBIOS, and are working with many others who will announce products in the future. The developers that have announced products include companies such as Newport Systems who have developed an ISDN LAN Router to interconnect Novell LANs using ISDN, and Marble Associates who have developed an ISDN multimedia application for the NeXT computer. I invite you to visit our hospitality suite to see some of these applications. We see developers as an important part of our ability to service our customers with applications. For example, in order for the multimedia applications to develop, they need a transmission facility, and we must provide a platform for developers to write applications to use the facility. To support the growing momentum of multimedia in the computer industry, the integrated voice and data -- the ability to transmit fax, images, and video -- make ISDN Basic Rate increasingly the first level of transmission that makes sense for multimedia. With an ISDN Basic Rate connection, and the appropriate software running on a local area network, you can operate a PC workstation as a remote node and have all the functions you would have as if you were in your office on the LAN. Your access to shared resources on that network is the same. We have looked at my first two decisions: customer needs and the products I can provide. Now this brings me to my third decision. Just as we the vendors and manufacturers have stages of ISDN deployment, so do customers. What are the stages of deployment for customers and what solutions can we develop to help them through these stages? I believe the first stage of deployment will be a transition stage. We have customers that say they want solutions with increased benefits but without having to fight or have increased headaches. They will not jump into new technology or a new strategy and change their entire operation without gaining confidence in the widespread availability of ISDN service and gaining experience with applications that help them get their job done. We, the computer industry and the telecommunications industry, need to work together to come up with the realistic strategies and solutions for our customers. You see, the perception has been that ISDN would come in and replace everything from modems to local area networks to telephones. That was mis-information for our customers and we must demonstrate that ISDN is a new technology that co-exists with existing technology environments. Customers want to know: Can I use what I currently have and how easy is it to get what I need? Customers want the transition to new technology to be affordable and easy. They want to be able to move to the new technology with confidence. Customers CAN use what they have today: their PC, their local area network, their communications software -- and with our hardware products, they can even use their modem or analog telephone or a regular fax machine. Our hardware interfaces work with existing communications software applications and provide software developers the tools to create new applications. ISDN is going to co-exist with everything else, as I said before. It is going to co-exist with LANs. It is going to co-exist with modems. We must make it clear that the co-existence will occur as one transition strategy. The customer premises and central office modem pooling solution is widely proposed but appears, in our opinion, to use excessive network resources from the local central office switch. When you do it that way, you have to make a call into the switch where you are then connected to a service that does the translation from digital to analog. It then requires another call from the other side out to the other end. That's two different phone calls plus a service charge for the conversion. When we had to consider how we could help that co-existence we made the decision to provide our customers with an X.25 PAD in our modems beginning three years ago. The packet switch network will be an easy, affordable connection for the analog world to the digital world. As you can see, an X.25 PAD in our modem that supports X.32 or Dial X.25 provides the most cost effective solution for our customers and capitalizes on the fact that all RBOC's are connecting their ISDN networks to the public packet switched networks. The packet switch networks are adding a feature to their networks that will let them do out calling on the far end. The modem on the other end can answer and connect with the packet switch network. You don't have to put new modem pools in. You use the existing packet switch modems that are already there. The telephone company won't have to make a capital investment in addition to modems, other than the ones that were already there on the packet networks. And you only made one phone call. It's less trouble and less expensive. And the packet switching is an inherently lower overhead technology for delivering the data across the country or around the world. But, as I said at the beginning, as a computer communications company that has developed CPE/ISDN adapters, Hayes is only a part of the puzzle. We need to have support from other parts to make it easy for our customers to make the decision to buy ISDN. We must work together to break down the barriers and make the purchase and installation easy. We cannot expect our customers to be willing to "blast" their way through the barriers that the industry has allowed to exist. And, as much as we are making progress on tearing down the barriers, I am afraid that we are building new barriers. New barriers that we will have to face in the next stage of a customer's deployment. After we have transitioned our customers to ISDN, they will then begin global deployment. The pattern we've seen is that they buy two or so ISDN adapters for evaluation. They get an application working, and then they want to buy anywhere from 10 to 30 and try them out in a department. When that's successful, they're ready to start planning for hundreds of installations and deploying them not only across a facility but also across an organization which is national or global in scope. As many of you know, products like modems which are regulated by PTTs around the world, have different requirements in different countries. As a result of those different requirements, three years ago we had to change the way we designed our modems. With this new design methodology, we have been able to get ULTRA 96 approved in over 30 countries in a year's time, including the most difficult tests in France and Germany. And it has been our most expensive modem development ever. I am afraid that we are building the same barrier in the ISDN world that we had in the analog world. Different standards for different countries. While many of the protocol standards will be consistent, there are still significant differences in many countries that would be barriers to ready deployment of equipment. What is happening with ISDN is that there is no overall standard for the full function of ISDN. If we say we are global industries, shouldn't we be acting and thinking globally? How can we expect our customers to buy into this technology if we're allowing service providers in certain countries to create trade barriers and deployment barriers that don't do anybody any good. The promise of global compatibility and connectivity is falling on its face from the customers point of view. The success of ISDN is going to be driven by the success of the computer applications we provide our customers. Most of our customers are international, doing business in various countries. About half of the computer industry business is outside of North America. The only hope is for the user community to insist on global standards not only for protocols but also for functionality so that they can expect to implement consistent applications on comparable equipment wherever they do business. Without comprehensive global services standards, my job as a manufacturer of CPE/ISDN adapters is more difficult. The equipment will be more expensive. Delays will occur in the availability of products in different countries. More importantly, look at what it costs our customers when we let the environment diverge in different countries. Differences tend to hold back acceptance and deployment of new applications and the whole process of commercialization get throttled. Working on ISDN 1 addresses part of our problem. And I applaud the telecommunications industry for this work. But, we need to get into ISDN 2 and 3. ISDN 1 defines a set of functions and a common way for implementing those. But it does not cover what's normally expected in the area of supplementary services. Many of our customers have told us they need supplementary services. So it's important to address a more inclusive national ISDN definition -- And then to address it globally as a full-featured ISDN definition that is truly standardized between switches and service providers on a global basis. And, that's not being done. We're letting all of this happen without dealing with that issue. We, as a group in North America, are not working with our counterparts in Europe and Asia to be sure there are complementary standards. We're not driving the activities of the CCITT to get global standardization of ISDN features on the agenda of every service provider in the world. My dream has always been to be able to walk in any of my offices or any of the hotel rooms where I might stay and plug in my laptop computer that's equipped for ISDN and be able to get full service connection into my home base. The way it's going that is not going to be possible. That capability hasn't been addressed. Until we can integrate our standards like ISDN 1 with European and Asian standards -- where ten years from now the markets are going to be much larger than in North America -- we cannot say that all of the pieces have fallen into place. North America is only one piece of ISDN deployment. Yes we have made much progress and ISDN is a real tool that businesses can use to create advantages for their operations, but we must not stop here. Today, I challenge you. How can we complete our job of making ISDN successful if we don't provide our customers with the best, most convenient and affordable solutions wherever they need it? We can't persuade our customers to buy ISDN, but our solutions for creating applications that give our customers a competitive advantage will close the sale. All of us here have an important role to play, both at this conference and when we go home, and the most important job that we all have is to serve the ISDN customer. ###