Home » Uncategorized » Mr. Network

Mr. Network

The next decade will be the era of network-centric communications. Since networks are Novell’s turf, by rights, it should be have been a second Microsoft, except that it shot itself in the foot with ruinous acquisitions and missteps-reloaded and fired on the other limb. All this has changed, says CEO Bob Frankenberg, who has refocused the company around network connections.

The emergence of the relatively cheap personal computer changed corporate computing by ultimately shifting the center of gravity from centralized mainframes to decentralized desktops. The next gravitational shift from islands of PCs to network-centered computing is under way. Both local area networks (LANs) and wide area networks (WANs) connected by routers, hubs, and various bridges have formed the central nervous system of distributed, enterprisewide computing. As companies increasingly attempt to unlock critical corporate knowledge, powerful networks are needed to move it between groups and deliver it to individuals who can use it most effectively. As the distinction between inter-networking and intra-networking dissolves, reliable and flexible operating systems for supporting multiple networks will be in great demand. IBM’s Louis tit, Gerstner told London‘s Financial Times recently, “The implications of network-centric computing will transform every business organization in the world.” This may have had something to do with IBM’s decision to spurn Apple and acquire Lotus, creator of a popular networking software called Notes, instead.

But the biggest networking software provider isn’t Lotus or Microsoft. It’s $1.6 billion Novell, based in Provo, UT. The second-largest software company enjoys a distinct advantage: Its networking software dominates 65 percent of the global market for network operating systems the way Microsoft Windows dominates desktop computers. It has 50 million users worldwide and is used by 88 percent of the Fortune 1000. If Lotus Notes, appealing to the high end-user, is the Cadillac of groupware, Novell is the ubiquitous, utilitarian Chevy. But Novell also has a distinct disadvantage: It is relatively unknown compared with rival Microsoft, which now is setting its sights on the networking software market with the introduction this summer of Exchange. Novell also is recovering from a few self-inflicted wounds, such as the acquisition and divestiture of WordPerfect, widely reckoned to be the costliest, in financial and organizational terms, in high-tech history.

Novell Chairman, President, and CEO Robert J. Frankenberg, 48, may have one of the toughest jobs in the high-tech world today. The former Hewlett-Packard vice president of personal information products-who became Novell CEO in 1994 and promptly pared the number of businesses to two from 19-must stem short-term quarterly losses while fending off Microsoft, which trails Novell but is growing rapidly. “Corporate customers don’t see a clear and compelling strategy from Novell that makes them feel warm and fuzzy,” says Patricia Seybold, president of The Patricia Seybold Group, a Boston-based software consultancy. “They also are being seduced by Microsoft’s key server strategy.” Customers who have servers on a network likely will want to use the server to run database applications, as well as manage the network.

Frankenberg says the next version of Novell’s NetWare, clue out shortly, will address the cost of managing and administering these networks. He insists the market will not force customers into either/or choices. Novell’s strategy assumes a multiplatform world, where customers can use, say, Lotus Notes or Microsoft NT with Novell NetWare. “The jugular issue here is, ‘Do I need a $3,000 PC on my desk or will a dumb terminal-the so-called Internet PC-connected to the Internet plug me into all the information. I will need?” says Frank Ruotolo, president of The Future Group, a strategy technology consulting firm. “It’s the last thing Microsoft would want, because it would have you buy big gear with all its powerful software.”

But network management isn’t limited to workgroup or even enterprise links. Frankenberg sees the need for ubiquitous linkages connecting enterprises, work-groups, and supply-chain groups to the Internet to form a single network encompassing smaller networks. The CEO says Novell aims to link up the vast number of intelligent devices embedded in various domestic and commercial products into a “smart global network.” Paul Saffo, president of the Institute for the Future in Menlo Park, CA, predicts that most artifacts soon will have some form of “intelligence.” These “smartifacts,” as he calls them, are everywhere. For every microprocessor in a PC, 10 are installed in equipment or appliances. Over the next five years, that ratio will move to 50-to-1. Frankenberg, an avid fly fisherman (favorite lure: woolly bugger) and aspiring gourmet chef who makes a mean osso bucco, envisions a world in which cars will transmit diagnostic maintenance data to service mechanics and performance data to manufacturers. Vending machines will signal bottlers to refill units automatically. Utility companies will read electricity meters from central locations and adjust power flows accordingly. These links also could be tied in to mobile telephones and faxes. Thus, through Novell Embedded Systems Technology (NEST), Novell is working with Utilicorp to turn power lines into LANs. The company also is partnering with telecoms such as AT&T for its NetWare Connect Service (NCS). Novell estimates a single global network would enable a billion connections. No single company could handle all this. So Novell is striking deals with Deutsche Telekom, France Telecom, Telstra, Unisource, and Japan‘s NTT to deliver NCS worldwide.

Frankenberg reckons SGNs will accelerate the value-chain process, but knows that smart marketing will be key to the success of Novell’s technology. Soft-spoken and affable, the Chippewa Falls, WI-born executive, who grew up in Sacramento, recognizes that whereas Novell may have share of market today, Microsoft has share of mind. To America in the 1990s, after all, Bill Gates’ image is what Chairman Mao’s was to China during the Cultural Revolution: It’s become an icon. As Seybold puts it, “Novell may have a leading technology here, but Microsoft has been known to buy what it needs to get into the game.” Frankenberg acknowledges the threat, but thinks the folks from

Provo have a two-year lead over the folks from Redmond, WA. One advantage not yet equaled by Microsoft is a directory technology that helps identify computer users and determines which corporate databases they can access without needing multiple passwords. Alluding to the backlash against Microsoft’s perceived dominance in the industry, Frankenberg adds, “We’re not trying to own anything.”

“The company has taken some hard knocks in the last year or so,” says Ruotolo, “but Frankenberg has cut its losses and restored flexibility. He may not have Gerstner’s deep pockets or Gates’ marketing muscle, but he’s making the most of what he does have.”


Networking software appears to be the next battleground in computing, with Microsoft, Lotus, IBM, Novell, and others all on a collision course to develop the standard. What is Novell’s advantage?

First, some 50 million people use our networks every day-more than the number of people on the Internet. Plus, our technologies address the Internet’s deficiencies: lack of security, spotty performance, complexity of use, and administration complications.

In addition, while we have formed strong partnerships with companies such as AT&T, Microsoft has made a number of enemies, particularly among telephony concerns that will play a vital role in the next stage of computing.

Finally, our employees understand networking better than anybody; in fact, some of them pioneered the idea of personal computer networking.

Some observe that Novell could have been the next Microsoft, but that it doesn’t market itself well and isn’t a well-managed company.

There’s significant truth in that statement. Novell marketed effectively to the early adopters in networking, the network managers and people in the second tier of an MIS organization. However, we needed to reach the people who are making networking decisions for the firm, not just the department. We’ve made substantial progress there in the last year.

Of course, it’s far easier to market now that Novell is in only two businesses-networking and network services-instead of the 19 when I arrived in 1994. In the beginning, we targeted virtually anybody who bought software. When you’re Microsoft’s size, you can do that. When you’re Novell’s size, you can’t.

The other key piece is that for Novell and WordPerfect, each round of improvement originally was an improvement in the technology, not in focusing on users’ needs. Today, customer needs drive the technology, not the other way around. Understanding those unmet needs and imaginatively addressing them is the core of the software business. That shift is now under way at Novell.

What strategies, if any, have you brought from your alma mater, Hewlett-Packard-and have they worked?

I participated in the development of H-P’s current business planning process and brought that with me to Novell. That transplanted quite well, though we made some improvements and adaptations to Novell’s environment and culture.

I also stressed the idea of management training. In the past at Novell, people signed up for management courses and then didn’t show up because they got involved in a crisis. Training doesn’t give you everything you need to be a manager, but a little knowledge never hurts. In addition, we expanded the breadth of that knowledge. Before I got here, only the basics were covered, while other things, such as helping people learn how to communicate, were missed.

Finally, as at Hewlett-Packard, we tied our employees’ rewards  bonus and stock options-to accomplishing breakthrough objectives.


The merger between Novell and WordPerfect didn’t go smoothly. Were you surprised by the clash of cultures in the two companies?

No. People assumed that because the two companies were headquartered in the same valley, the cultures must be the same, and, therefore, would not present a problem. That’s not true. There were significant differences: WordPerfect was a private company, Novell a public one. WordPerfect interacted more casually with the press than Novell did. WordPerfect did almost everything in-house, from duplicating software to making the cafeteria food to owning its own security force, while Novell outsources everything that someone else can do better.

In any case, there were clear opportunities for improvement, and that was initially difficult for the WordPerfect people to accept.

Did you know what you were getting into when you came on board?

Yes. I had had extensive discussions with virtually everyone who worked for then-CEO Ray Noorda, and they told me about the problems they knew. I didn’t find a single case of anybody painting a rosy picture when it wasn’t deserved. When I walked in the door, I had a clear idea of the things that weren’t working well in Novell. I knew less about the situation at WordPerfect, partly because the merger wasn’t complete at that point.

NetWare sales rose 21 percent last year to 825,000 servers. But Windows NT server sales nearly tripled. The growth pattern in the interim looks like they’re coming on strongest Is that because you were distracted by merger missteps?

There was some distraction there, and it wasn’t all due to merger missteps. When I got here, every major product was in trouble, in WordPerfect and Novell. NetWare 4 had enormous reliability problems, and WordPerfect, the Windows word processor, was also unstable. We needed to triage aspects of those products, and that took some time.


What happens now?

By focusing on networking, we now can deliver a much better value proposition than Microsoft does. Networking involves connecting diverse devices, such as Microsoft products, Apple products, UNIX products, and legacy mainframe systems. This allows people to choose what systems they use, while the choices they’ve already made still work. We now have three new clients-the software that lives on the desktop machine and provides the connectivity to the network services-one for NT, one for Windows 95, and one for Macintosh.

Microsoft, on the other hand, connects its own products and begrudgingly connects you to existing mainframes. In addition, Microsoft has been ignoring about 80 percent of the market for networking, which comprises all the other intelligent devices other than PCs, such as automobiles, appliances, and personal digital assistants.

I’m interested in connecting these devices, and using networking to control them, to find out information about them, and to extract data from them. For instance, a car has at least eight microprocessors in it and sufficient memory that we can do some interesting things. We can connect it over a wireless network so you can get services, such as weather and traffic reports, guided tours, directions, diagnostics, and service reminders. We’re working with two manufacturers to put this capability into cars. They plan to have them available in the 1998 model.

There are also mapping systems that use CD-ROMs. However, that isn’t our highest priority market for what we call Novell Embedded Systems Technology (NEST), which is software that embeds networking into intelligent devices. Right now, we’re looking to plug together office equipment such as fax machines, scanners, and printers, so they become a service on the network. This would allow you to walk up to a fax machine-which would have an infrared link instead of cables            and just say, “I want to fax this,” and the machine works it out and sends it off. Twenty-four of these products are shipping now, and 200 more should ship this year.

Embedding also plays a role in the work we’re doing with Utilicorp to turn power lines into local area networks. This would mean you could plug a printer with NEST on it and a PC with networking on it into a wall socket, and they’d be connected without cables.

We also will provide connectivity over the television cable to your house, so there will be competition for your data connection. Utilities, several long distance carriers, the local regional Bell operating company or PTT, cable operators, and mobile operators all will compete to connect you to the network.

Will this change telephony, as well?

Absolutely. You’ll be able to connect with this data network the same way you do on the voice network to anyone anywhere in any of the service areas that subscribe to the data network. Like the Internet, it’s universal; unlike the Internet, it has guaranteed performance and security.

What’s to prevent somebody else from connecting people, too?

Other folks will try. Part of our edge is being first; another part is that the technology is very difficult to develop. Can our competitors catch up? Sure, but our best estimates show that our closest competitor in that arena with a potentially similar capability is Microsoft, and it’s at least two years off.

We have this technology now, and it’s being adopted by all the key players, including the telephony companies. Plus, half the UNIX world will be using our core technology.

Do you think people will use Novell technology because they feel antipathy toward Microsoft and its power?

Perhaps. People want to see a balance, or as is sometimes said, “The enemy of my enemy is my friend.” We have a different strategy than Microsoft does. We’re not trying to own it all, but rather to have great capabilities that do part of this whole thing extremely well.

When will Novell’s next new product be released?

The next release of NetWare, code-named Green River, will be this summer. It will include a licensing capability, because the cost of management and administering networks is enormous about 75 percent to 80 percent. Only 3 percent is in our software. So we’re aggressively going after the process of installing, configuring, updating, and licensing software. This means you would be able to buy a copy of a product, put it on one server, and deliver it to all users on the network, saving the software manufacturer a lot of headaches and the customer the money from paying someone to put it on every personal computer.

This year, we also will release a Group-Wise product, XTD, which provides over the network services such as public e-mail; calendar; scheduling; collaboration; conferencing capabilities; full work flow on image or text documents; and a secure connection to the Internet.


Some IT management gurus are convinced that the future of work will be groups of people in different networks all sharing information. Can Novell’s products fulfill that prediction?

Before I answer, let me explain replication. This process replicates portions of a database-documents, transactions, ledgers, etc.-so you have local access to it. And when changes are made, the replication can be repeated at appropriate intervals to assure that within a certain range, it’s up to date. Lotus Notes currently has this feature; NetWare does not.

We are bypassing the replication process by enabling collaboration on the creation of those documents on a Web site, which means the collaborations can then be represented, updated, and shared back without moving a copy of the document to all the parties interested in it. Our document collaboration capabilities will ship this year. They will use the network and not require replication to near the degree that’s out there now, though we will have replicating facilities for when you go mobile.

Is an interpreted language such as Java or a device that doesn’t require an expensive computer a likely threat to your business?

No. Actually, the more different devices there are, the happier we are, because we can connect them. Java and the so-called network terminal or network PC represent new opportunities for us. Running Java on top of our NetWare servers makes them a better competitor to NT, since you can have Java on a Macintosh and another one on OS/2, UNIX, and Windows. It turns out our servers are the fastest environment for a Java-like interpreter, because we don’t have multilevels of software.


What do you regard as the chief threat to your plans?

To make all this work, the partnerships we’ve established must take hold and thrive. Without the AT&T infrastructure, the network connect service wouldn’t work. Without original equipment manufacturers, there wouldn’t be intelligent devices that connect to the network.

I shouldn’t say anything about threats without mentioning Microsoft. Microsoft’s ability to stall the market-something it learned well from its mentor, IBM-is a significant part of its threat. But on the delivery and execution front, it presents a little less of a threat.

About J.P. Donlon

J.P. Donlon
J.P. Donlon is Editor Emeritus of Chief Executive magazine.