Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : Ascend Communications-News Only!!! (ASND) -- Ignore unavailable to you. Want to Upgrade?

To: w2j2 who wrote (412)	10/29/1997 2:35:00 PM
From: Maverick	Respond to of 1629

Interview w/ Eric benhamou Erudite Eric 3Com CEO Eric Benahamou speaks out on competitors and technology. 10/27/97 For all of his success as CEO of 3Com Corp., Eric Benhamou has to contend with the fact that archrival Cisco Systems Inc. is running away with the game, amassing acquisition after acquistion and running up revenue. In part to keep pace with Cisco, Benhamou has architected one of the largest mergers in the history of our industry, marrying 3Com to U.S. Robotics. Network World Editor in Chief John Gallant caught with Benhamou at Networld + Interop in Atlanta to discuss the changing face of competition. Q. Eric, besides you're traditional competitors - the Bays, Ciscos and Cabletrons - the acquisition of U.S. Robotics pits you against new competitors and companies like Lucent and Nortel are coming at you from the telecom side. Who do you focus on as the competition these days? A. We've stopped focusing on any one primary competitor and have begun to look at it by market segment. And in every broad market segment we know there are a few people we should worry about more than most. In the enterprise, we worry about Cisco, as well as Bay and Cabletron. There is almost no deal that we win without setting out to beat Cisco. In remote access, it's Ascend more often than Cisco. We hardly see anyone else there. In the NIC space the only competitor we worry about is Intel. All of the others have fallen by the wayside. In modems, the only real competitor is Rockwell. The other modem vendors are basically distribution channel partners for Rockwell.

To: w2j2 who wrote (412)	10/29/1997 2:38:00 PM
From: Maverick	Respond to of 1629

Interview w/ Eirc, Part II On a longer-term strategic basis, we worry primarily about Cisco as it attempts to expand beyond its native space, and we worry about the telecom oriented vendors who may want to expand into the networking space. Q. Some folks say that Cisco, teamed up with Microsoft and Intel in the so called Wintelco triad, controls the industry. Do you believe that? A. In one word, no. They have a strong presence in some markets but are absent from many others that I think are just as strategic over time. Cisco basically attained its position on the strength of being completely focused on the internetworking backbone opportunity using general purpose routing technology. And this is what more lately got them into the carrier opportunity. There's a lot more to networking than that. In fact, there are four major markets in our industry. There is the enterprise market, the carrier/ISP market, the market for small businesses and the consumer market. Over time, these last two markets will be absolutely formidable, as large as the enterprise markets is ten years from now. And what is special about our industry is that all of these markets can be served with the same technological building blocks because you have one bit stream that goes from one side to the other. And it has to be a common set of protocols and technological approaches to be able to let the consumer connect one second to a large corporation, another second to an ISP, another time to a small business, and it's still using the same access method, the same protocol, the same technology, the same system, same network. Q. Cisco's CEO John Chambers says the diversity of your product line - stretching from the consumer all the way up into the carrier market - makes it difficult to deliver any kind of meaningful integration. A. Well this is a strength, not a weakness. We have cultivated that diversity because this is where the market is going. The market is broadening. It is true that it is an ambitious endeavor, and this is not a strategy that makes sense for a lot of people. It makes sense for us because we've been able to grow in every market year after year. We've gained share in

To: w2j2 who wrote (412)	10/29/1997 4:40:00 PM
From: Maverick	Respond to of 1629

Interview w/ Eric, Part III the key segments in every market, and we have developed in a fairly harmonious way a set of channels that give us access into all of these markets. We have a strong enterprise direct channel. Largely through U.S. Robotics we've acquired a strong carrier/ISP channel. We've always had a strong base of channel partners that feed into the small and medium sized enterprise. And now we also have a strong retail channel. In fact, of all the network vendors, we're the only one that has a strong retail channel that can reach into the consumer markets. In addition to this, we have a small number of strategic initiatives that cut across the entire company. One which really brings together the entire vision of the company is what we call Next Generation Networks. We strongly believe we're in the midst of a fundamental change in how networks are built. A key difference between future networks and today's networks is connectivity on an end-to-end basis. And for us the end of the connectivity chain is literally where the user is; at the computer, at the desktop, at the server, and sometimes the user may not even use a classical computer. The user may be an application, it could be a connected organizer. It could be a set top. So for us, connectivity ends there. Another key difference is that future networks have to be dramatically more intelligent in terms of being aware of who the user is, what the application does, what the policy is that they should use to move the bits around. It is fair to say that the products the industry has produced to date have basically done a good job at moving bits faster and cheaper from one generation to the next. But they have not really fundamentally changed what they do. What is changing now is that we really have to look in the bit stream, and can no longer just solve all the problems by going faster and adding bandwidth. We have to intelligently treat and process this bit stream based upon some pre-defined policies. In our view, a lot of this policy intelligence will have to reside in the client access point. If you rely on the core network to make all of the policy decisions, this will result in a completely unmanageable, exorbitantly expensive situation.

To: w2j2 who wrote (412)	10/29/1997 4:41:00 PM
From: Maverick	Respond to of 1629

Interview w/ Eric, Part IV The best place to look up policy and to make choices is at the access point. This is the point where the policy request is made, and this is what has to be translated into networking language, to protocols, and signal throughout the rest of the infrastructure. So for us this is where you can most effectively add this intelligence. And this ties to my point about end to end connectivity. Because behind every NIC, behind every analog modem, digital modem, serial modem, and so on, you will have to figure out how to move this bit across a network. Not just push it down through a buffer and send it across. It will figure out all of the various policy decisions that you may have to make. So for us this is a fundamental departure from the way we have built networks in the past. Q. Ok. But why does Wall Street put a significantly higher evaluation on Cisco? Does Wall Street think Cisco has more control over the industry? A. Well they have executed very well, and they have occupied a strong position in the enterprise, which is a highly visible part of the market. You will notice, however, that their rate of growth has substantially declined over the last two or three quarters. This is true at the top line. This is also true at the bottom line. Their financial margins are unlikely to go up. They are trending down, and John [Chambers] has repeatedly signaled that it was going to come down further. You will also notice that the bulk of their business comes from segments of the market whose hay days are behind us. People no longer get excited about the growth rate of the general purpose router market. Our market capitalization is a fraction of Cisco's, and yet we have almost the same revenue base. This is legitimate on the part of investors. They're watching us closely before they commit to a high level of investment in our company because we're attempting to do something that has not been done before

To: w2j2 who wrote (412)	10/29/1997 4:45:00 PM
From: Maverick	Respond to of 1629

Interview w/ Eric, Part V - the largest technology merger in the industry. It is exciting, exhilarating, but it is risky. We made no secret of it, but the pay off could be huge. Q. Let me get your reaction to this quote from Chambers about the merger. ''I think the merger speaks to how rapidly we [Cisco] are pulling away from the rest of the industry. The fact that Eric would do a combination with a company of equal size halfway across the country speaks to how much risk he is willing to take on to stay close to Cisco.'' A. It is definitely one of the factors that went into our thinking. Critical mass is a factor. It was not just aimed at Cisco. It was aimed at a much broader phenomenon: industry consolidation. We'd like to take credit for starting this trend. The first acquisition made in this industry we made and it proved to people that you could make this work. And we also take credit for the fact that we raised the stakes. We did not make just a small technology acquisition. We made a substantial merger. And we can make it work too. Let me give you an inside view of why the risk is highly mitigated. Most large transactions fail as a result of poor and untimely communication to the employees. In our industry, we're not about bricks and mortar. We're about intellectual capital. We're about people. And unless you can harness the people assets properly in a combination, then it's not going to work. We believe that we will make this work because we have understood the challenge extremely well, and because we now live in the Internet time. The Internet, and the intranet, which is the infrastructure by which we run the company, is an essential tool that enables us to bring the company together faster than it was possible before. Here is my philosophy about this. I believe that every employee who's company gets engaged into a large combination will have a time constance of about three to six months during which they'll be somewhat worried about what might happen to them, to their company, to their colleagues. But they will give it a shot, particularly if they have been well served by their employer up to that point, and say 'Alright, at least I'll wait and see what happens.' If enough communication, enough answers reach this employee

To: w2j2 who wrote (412)	10/29/1997 4:46:00 PM
From: Maverick	Respond to of 1629

Interview w/ Eric, Part VI in the span of less than three to six months, if strategy gets solidified, communicated fast enough, if their role in the strategy is made clear fast enough within that sort of mental time window, then you've got it made. Most large combinations, such as the ones that have taken place in say the banking industry, typically have failed because people are left in the dark for a long period of time, and they're given the opportunity to disconnect, to bail out mentally. I'm not going to give employees the opportunity to disconnect. In fact, I think the most critical time is already behind us during which we enlisted all of the people towards a new dream. Not one that is fundamentally different than what they had before, but broader and with new motivation. So as you can tell, I am speaking about management techniques, and how they are changing in the Internet age. We're in this business, and we're leveraging our own tools to make these combinations work better. As an example, day one after the merger was finalized, we integrated our intranet and the U.S. Robotics intranet We integrated our mail systems, we integrated our voice mail systems, we integrated all of the possible means of communication between the two organizations. In fact, before even the merger was sanctioned by the shareholders, we created a special intranet site that was only open to the senior managers of the two companies such that we could do all of the pre-planning that we could. All of the pre-planning that the SEC would let us get away with so that we would hit the ground running. This form of intense, detailed level of communication has never been done before in any merger. We're the first two companies to do it.

To: w2j2 who wrote (412)	10/29/1997 5:03:00 PM
From: Maverick	Respond to of 1629

Interview w/ Eric, Part VII Q. Where do you think Cisco is weakest? Most vulnerable? A. First, their fundamental bread and butter product line is no longer in a high growth segment. In fact, the general purpose router market is at best a flat segment and will probably decline before the end of the decade. That product line is getting subsumed by many other new segments, including layer 3 switches, which I think will provide all of the traffic management and control the enterprise needs. In that space we're clearly ahead of them. From a product perspective, from an intellectual capital perspective and from a marketing perspective. The second weakness is their failure to parlay their strong position in routing into the remote access field. This should not have allowed Ascend to get ahead. They should not have allowed U.S. Robotics to get ahead, and they did. The third weakness is more of a positioning weakness. When they purchased Stratacom it was with a deliberate intention to compete for the network infrastructure of carriers, and to compete against telecom equipment vendors. Cisco wants to have more and more share of the CO. And in that battle they don't compete against us, they compete against Siemens, Lucent, Nortel, Fujitsu, Erickson ... very, very large established companies who have broad, deep relationship with carriers. We on the other hand have chosen to not cross this line. We are forming partnerships instead. The other weakness is in the area of the small business and consumer space where they don't have appropriate channels. Q. Are you concerned about Cisco's partnership with Microsoft? A. I watch it carefully, and yes I would be concerned if I felt it was going to become a sort of brotherhood. I don't think it is a brotherhood. It is in the best interest of both Intel and Microsoft to insure that they don't tie their strategies to any particular network architecture vendor. Q. One area 3Com has, I would say, a significant lead over Cisco is in Gigabit Ethernet. How do you plan to take advantage of that?

To: w2j2 who wrote (412)	10/29/1997 5:07:00 PM
From: Maverick	Respond to of 1629

Interview w/ Eric, Part VIII A. I'm actually surprised at this on their part because I would have expected them to naturally move into gigabits. And I think it's probably not a lack of desire on their part, but probably some execution problems that have arisen over time. We're going to take advantage of being much earlier to market with a product we're extremely proud of, the CoreBuilder 3500, and soon the 9000. Basically we're building a system strategy where that line of products sits at the core and provides a combination of bandwidth expansion and traffic control in data centers and major campus backbones. It can also be used in gigabit POPS for Internet connectivity. It can even be used by service providers who are aggregating high bandwidth traffic and need to switch this traffic inside the POP as you move into the backbone. But the primary application will be in the enterprise market, data center, and campus backbone. We think this product can be sold in two ways. Either as just a great add-on, bandwidth expander, traffic controller product, even in existing Cisco accounts without even causing Cisco customers to throw away their Cisco products. It simply is an add-on to enable us to encroach into network designs which Cisco has had total control over. And secondly, can be sold as part of a comprehensive system where this sits at the top of the bandwidth hierarchy. You complement this with the other CoreBuilder family members, the 5000, the 6000 and the 7000. And then at the next level, you get into the wiring closets, and you add to this mostly SuperStacks but some CoreBuilders as well switching elements. So we can create a very complete enterprise switch fabric. Q. Let me ask you about your announcement with Juniper. How much of that is positioning and how much is reality? You're talking about established Internet providers risking their networks on a start up. A. Well most internet service providers have built their networks on start up products. Ascend was a start up. U.S. Robotics was a start up, and they helped build the ISP industry. For us what Juniper brings is an extraordinary set of talents who really understand very, very high performance routing. We know this is a company that will give Cisco trouble. Companies like

To: w2j2 who wrote (412)	10/29/1997 5:09:00 PM
From: Maverick	Respond to of 1629

Interview w/ Eric, Part IX UUNET are investors in Juniper because they are going to be a large customer. And they're knowledgeable investors, having already heard what Cisco could do for them and what Ascend could do for them. Q. Does the potential merger of WorldCom and MCI trouble you in that these companies tend to choose strategic partners? A. There's a bit of concern in terms of what they might want to do with their network. But I believe this is more good news than bad because I think this creates a telecom service provider that can get the economies of scale required to stimulate the growth of the market. If you look at the various components of the new World Com, we built most of the CompuServe network. We have several tens of thousands ports into the UUNET Network. The MCI Dial Network is probably our fourth largest dial network in the world. MCI itself is offering all of their ATM service with our access builderproduct name. family on premise, so we have very strong relationships with all of the key players there. Q. Let me ask you about two smaller competitors. Bay and Cabletron have both undergone senior management changes. What's your reaction? A. Bay has done a good job creating a better, more competitive product portfolio for the company. They had a strong customer base, and a strong base of resellers, and if you couple this with a better product cycle inevitably they can resume their growth. I mean the market is there. So I wouldn't be surprised if they had a resumption of growth over the next few quarters. However, it takes more than just resuming growth to become a leader. They have lost a lot of their image and relationships along the way. There's enough left to make a viable company that delivers good eturn to shareholders. But I think too much time has passed for them to expect to become the number one in that industry.

To: w2j2 who wrote (412)	10/29/1997 5:12:00 PM
From: Maverick	Respond to of 1629

Interview w/ Eric, Part X Q. What about Cabletron? A. I think Cabletron has perhaps more inertia in its business than anyone else by virtue of having such close direct relationships with a few large Enterprise customers. Therefore, it would take a long time for Cabletron to be wiped out by a business down turn. But by the same token, Cabletron has always suffered from being too narrowly focused, so they miss several technology waves. They denied the benefit or the necessity of making acquisitions for too long. And then when they started making them, it was too little too late, and not well enough executed. So in many ways I think Cabletron has de facto admitted to be more of a niche oriented vendor, strong in that niche, but nevertheless not a broad based vendor that can set the pace of the industry. Q. How about Intel. Why won't Intel take over the NIC market given the control they have over the PC component? A. Because there is a fundamental architectural difference between the processing subsystem and the communication subsystem. It requires very, very different engineering skills, and they're just not known for that. It is remarkable that a company as financially wealthy as Intel has tried repeatedly to buy its way into this market by either buying companies, or engaging in a lavish promotions or by aggressively slashing prices, and still end up losing shares quarter after quarter against us and really not being a player in networking. I mean for every large networking project that goes on in the world, Intel is not invited to the table. They may sell some networking technology as part of the computer, but they're not invited to the table as a networking vendor. So this is a sort of a side effect, or a by-product of the Intel inside campaign. Intel is a great microprocessor company, one of the most phenomenal ompanies in the world that I have immense respect for. But in some dimensions, not in all. I don't believe that Intel has made it's mark in networking yet. Q. Do you expect that they will?

To: w2j2 who wrote (412)	10/29/1997 5:13:00 PM
From: Maverick	Respond to of 1629

Interview w/ Eric, Part XI A. Well if by this you are referring to a possible transition of some of the business to mother board, this doesn't bother us or worry us at all because we are players in this segment of the market. We don't believe it is a large segment. Nor will it become a large segment because it is still a very fast changing field. And there is still a lot of new things that have to go into this communication subsystem frontend. Often times it is a bother, it is a constraint to have everything hard coded in the mother board itself. Nevertheless, there are segments which will be perfectly acceptable. And therefore we will compete in that business, and in fact we are today. If you open a Dell Computer right now you will see that it has 3Com chips on its mother board. You will also see that some other Dell Computers have 3Com NICs in them. So the fact that Intel is there with silicon expertise has not really helped them all that much from a technical standpoint. The factories that they use will always give priority to Pentiums and processors. Paradoxically, this is a semi-conductor company that is a bit like the cobbler who mistreats his children. In other words, we are going to get better, more advanced technology from our foundries than we would if we were a division of Intel. Indeed today they use fairly old, outdated technology that has a higher flaws- per-die than we do. We use smaller geometries. We have higher yields on our fab lines. And all of this gives us the technological advantage. But if we go back to the fundamental trend in the industry, which is end-to-end connectivity, it's very clear to us that a piece of the network that exists in a computer or in an appliance on a connected organizer or a server, is an integral part of the connectivity chain that cuts across the entire network. It has to evolve in synchrony with the rest. When you add new protocols, when you add policy intelligence, when you add encryption, when you add quality of service, it has to be end to end. It has to be incorporated in that sub system. So we believe that the dynamic that drive this piece of the computer are very different from the dynamics that drive the MIPS inside the processors.

To: w2j2 who wrote (412)	10/29/1997 5:16:00 PM
From: Maverick	Respond to of 1629

Eric, Part XII In fact, even the operating system that runs in this computer, Windows, Windows NT, even UNIX, is structured in such a way that it deals with communication as a sub system which sits behind a service interface, and in many cases, this subsystem will be implemented in the form of a separate card. Or at least a separate area inside the computer that has software and silicon all embedded. So for us, we look at this as an integral part of our approach to building next generation networks. Tactically, we're going to fight against Intel and every product, even marketing promotion, and we're going to gain share quarter to quarter as we have. Strategically, we're going to turn this into a way to promote a different approach to building networks. It will help us not just there, but also inside the core itself. Q. Why do you think Wall Street gets so nervous about Intel as a competitor? A. Well, Intel is perhaps the most powerful company on earth, so I couldn't blame Wall Street investors for worrying about what Intel might do to a small company like 3Com. However, every investor or analyst who has spent a half an hour looking under the covers sees this doomsday idea of Intel wiping out 3Com is completely preposterous. Just look at the numbers. The portion of the NIC market in which Intel competes is between 5% and 10%. So it is remarkable that because of this overlap people would worry so much. Q. I want to ask you about the 56K modem market. Is 56K the height of the modem technology? A. It is likely that the analog modem market will end at fifty six kilobits. I think the analog modem is a narrow way of defining the modem market. Our view is that there will be a diversity of new modem technologies over the coming years. Most of them digital based. Some of them land lines, some of them cable, others copper. Others fiber. Others wireless. And we want to parlay our dominance of the modem market into all of these segments. So it is likely that there is going to be very soon a 56K standard, and that most of the world will migrate to it because it's just the best you can get out of existing infrastructure with minimum incremental investments.

To: w2j2 who wrote (412)	10/29/1997 5:20:00 PM
From: Maverick	Read Replies (1) \| Respond to of 1629

Eric, Part XIII What's exciting to us is two things. First, this is a market we stand to lead in a fundamental way. In the process of all of these technology transitions and tough market competition, we've actually gained share, and we've emerged as a strong leader and we own the intellectual property, so this is a good market for us, a bad market for everyone else. But what's more exciting is the other piece, which is people will want to get other forms of modems from us because, just like you have ''NICs are us,'' you also have ''Modems R' Us'', and this is a way for us to turn this into an advantage. Q. One final question. We're hearing more and more about voice from traditional data networking companies. What's 3Com doing here? A. Well I think it's the broader topic of delivering multi-media services on the data infrastructures. I think this is a fundamental shift in the marketplace that we want to ride very aggressively by providing platforms that provide multi media services. Our AccessBuilder platforms have been used very successfully by large corporations to deliver voice, video and data integrated on the same trunks. In addition to this, voice over IP is an area in which we've made major investments in collaboration with Siemens. We will also support voice services on the subscriber side, subscriber side meaning today you buy a modem from us, you buy Palm Pilots from us, you buy data devices from us. At some point, these devices will be able to support voice as well. And, of course, that voice will be digitized, and the voice bits will be sent into IP packets across the data network. Q. Did you say you were going to put voice on a Pilot? A. Did I say that? There's going to be so many things in the Pilot you're going to be blown away.