Novell's Smart Global Network and the Future of Computing
Articles and Tips: article
Chief Technology Officer
01 Dec 1995
This DevNote is adapted from a keynote address Sheldon Laube delivered at Novell's BrainShare-Australia conference, in July of 1995. Laube describes several turning points in the history of computing and then discusses the concept of universal access. Novell's Smart Global Network is outlined, with emphasis on NetWare Directory Services, NetWare Connect Services, and Novell Embedded Systems Technology.
We're standing at one of the great precipices of computing. What I mean by that is it's time for a fundamental change in how people perceive the value of computers and what they do.
The first occurred with the invention of mainframe computers. When mainframes first came out, their entire value proposition that generated and launched the computer industry was increasing the effectiveness of business processing: accounting systems, inventory systems, booking systems for airline seats, a wide range of business processes were automated with the advent of the mainframe computer.
The second was the creation of the PC, which once again fundamentally changed the value proposition. The idea of the PC was very different. What the PC was fundamentally about was personal productivity. Spreadsheets, word processing, and personal databases were all about a different value proposition than the original one and that was the value proposition of making individuals more productive.
Today, we stand on the edge of another big change and that literally is the fundamental change of connections being the value of computing in the new network world.
What's really going to be important, as we move to the future, is the ability to make connections, to interconnect, and give people the ability to connect to each other, to connect to the information they need, to connect businesses to businesses. This kind of idea is transforming our world.
The first way to think about that is at the business level. Businesses around the world are driving to interconnect themselves to others; to connect themselves up with their partners, with their vendors, with their customers. The whole idea of the World Wide Web is this idea of being able to make interconnections, to let people talk to each other, to get to the information they want. All over the world today, there is this enormous desire to allow electronic connectivity.
The Value of Connectivity
When I was at Price/Waterhouse, where I have been for the past ten years, we started a project called "Client Connectivity." It was a simple idea and the premise was that we would connect the 200 largest clients of Price/Waterhouse electronically to PW offices all over the world. Our idea was really simple. We were going to set up E-mail connectivity and allow people to exchange messages. But as soon as we started that process, we found that people wanted more and more. They wanted access to our data systems, access to our databases.
We had clients who wanted access to computers that we had when we were doing development work, while our staff wanted access to computers on client sites. Over and over we found that we had a greater and greater need to create the interconnections from business to business.
If you think about re-engineering and what's going on around the world in the world's business, there is this idea of just-in-time inventory systems when you deliver the parts when they are needed. It is clear that just-in-time inventory systems don't work unless you have direct data connection so that you know what the manufacturing organization is doing and when they need the parts.
The value proposition of what people will be using technology for in the next decade will start to revolve around the idea of allowing them to interconnect and that idea of interconnection is not only company to company, but it is all the way down to the individual.
I think that the easiest way to help you understand that is that I'll just tell you about my own mother-in-law. My mother-in-law is about 68 years old. Four years ago she decided that she wanted to write the family history. Our family had immigrated from Poland in the 1800s and she wanted to write the book about the family history.
So we bought her a computer and we started her off with an 8088 and she started typing away in WordPerfect. We then got her, two years ago, a 386/33 to work on. This year, after a lot of complaining about how slow that was, she now is running on a 90 MHz Pentium with a gigabyte drive, 16 meg of RAM, and a 17-inch monitor which is really a sight to behold.
But she created this wonderful book. It's 400 pages, single spaced, 60 pages of photographs, which was published. It was an incredible accomplishment and to her, what her personal computer was, was a faster way to type.
This summer I created cybermonster. Now, what I did was I gave this 68-year-old woman a copy of America OnLine. Now this is a really scary proposition, because now, she gets up in the middle of the night to send E-mail messages to "God knows who." She does reference work, she looks up things about countries for her books. She has now found these role playing places in America OnLine where she is pretending to be a 20-year-old woman.
Now, I don't know about you, but this really scares me. But the real point of the story is that in her mind, the value of her computer has been fundamentally transformed. What was an effective way to type is now her portal on the world. It is her ability to connect with people, with information, with all sorts of strange people in these role playing places and an extraordinary wealth of information.
Last week she called me up and said, "Shel, America OnLine is getting hooked up to the Internet. They have this new Web browser and I need to download it. Is it going to be OK? Is it going to destroy anything?" I said, "No--except the family sanity after you start surfing around."
But the real point of this is that it fundamentally changes the idea of what computing is about. That it is not about productivity, it's not about typing, but the next set of values is about making the connections from my mother-in-law to a wide world of strange and interesting things on the network and from businesses to businesses, to their clients, to their suppliers, to their vendors, and to their purchasers. A transformation as fundamental as the other two that we have gone through in the computing industry.
That's the first piece of the vision of what computing is going to look like in the connected world. The idea of network centric, in which the network is used to allow connections to be made.
The second idea is in order for that to really work, we would like access to those computer resources wherever they are. I'm sure all of you understand how valuable it is to be able to dial in and connect to your computer resources back at your office, back at your own organizations, wherever you are. It is absolutely fundamental to the value proposition of the ability to make connections that you can make those connections, wherever you may be. What people want is pervasive access to computer resources wherever they are.
At Price/Waterhouse, we used to have a phrase about our staff and that was "A good office is an empty office." What we meant by that is we did our best work when our people were out in the field actually working with our clients. When that happens for many of you who work with sales people, same thing. If you walk in their offices and all your sales people are sitting in their office then you say to yourself, "Well, we're not doing a very good job. They should be out there in the field."
In order for networks and computer resources in this whole world we're trying to create to be useful to people, it has to be accessible wherever they are, in their office, at a client site, in a hotel room, even sitting here in this audience. The value proposition of making connections goes up exponentially when you can make it pervasive and allow people access to that, wherever they may be.
Now, I carry around in the US a product called the Motorola Envoy. It is a little one-pound personal digital assistant. It has an antenna on it and it is hooked up to one of the new digital data networks that's being deployed around the United States.
That means I can carry around this little one pound device, sit anywhere in the United States, get my E-mail messages, respond to my E-mail messages, check my calendar, and look up information. All in a one pound device I can carry with me, without any wires, without any connections to a modem, without worrying about whether there is an analog line nearby.
Owning that device has fundamentally transformed my own conception of what computing is. I am no longer tied to my office to have access to my computer resources. I can now stay in touch wherever I might happen to go.
For those of you who think that this is kind of a strange phenomenon and it's going to be very narrow and limited, let me just ask people in the audience: How many of you have cellular phones? Well, just think about that. When cellular phones came out 5 or 6 years ago, the same things were said that we are now saying about wireless data and that is this: "Well, it is really for a very narrow population." But as all of you clearly know, the value proposition of having the convenience of having that phone with you, the ability to be in touch has changed all of your lives.
As data networks become available, the ability to connect up your information systems with the same ease that you have with your cellular phone, will transform how people use computers. So, the second key idea of pervasive computing means that it is accessible everywhere, to networks, to wires, to wireless networks, to infrared, to a whole wide range of communication devices. That wherever you are, you will be able to be connected.
The third key idea of pervasive computing is what I call 'heterogeneous.' That is simply the idea that the world of the future will contain a wide range of new and mysteriously strange computing devices. What I mean by that is that we have an incredibly narrow view of what computers are. A clue about that is our view that computers are changing every 5 or 10 years.
You know, when I started in the computer industry back in the 1960s, what computers were, were these enormous boxes that filled up a large room and that was a conception in my mind of a computer. Now, the vast majority of people who work for me have never seen a computer like that. They have never typed a punchcard themselves on an IBM 826 keypunch, and do not know what the phrase "Do not fold, spindle, or mutilate" refers to.
Now, we have an equally narrow view of computers. They have a screen, they have a keyboard, they have some sort of processing device, and you know, they look like a PC or Macintosh. But, the reality is that computers today surround us everywhere we look. Most new cars being made around the world have computers embedded in them to control timing and engine ignition.
For example, I went to buy my own mother, who is 85, a new microwave oven a couple of weeks ago when I was visiting her in Florida. We went out to an appliance store because we wanted to get her a microwave and I was looking for a simple microwave. You know, you type in "10 seconds", "go", right? Because my mother is not interested in high technology and she just wanted a simple one.
But those just don't exist anymore. There is no such thing as a microwave that you just type in the time and press "go." They have the most enormous range of controls and things that you press, and things that light up and buzz and beep and do all sorts of exotic things. There is even one we found that has a complete LED display and you can cycle through 200 different food items--hamburger, chicken, soup, pork--and then you type in on the keyboard how much it weighs and press "go" and it automatically calculates how long to cook the item for you. So, there is even a computer in there.
What is crystal clear is that in the world of the future, all of these types of computing devices will exist. There will be computers in your telephones, in your cars, in your microwave ovens, in your wall switches, in your toasters, even your washers and dryers have computers that sense and decide how long to dry things for you. Our fundamental notion of what makes up a computer is going to dramatically change and has to change to encompass this heterogeneous sort of world.
What is equally clear about this heterogeneous sort of world is that the value proposition of it explodes when you can connect up these nontraditional devices to the resources available on the network. Just take your car, for example. It is not a very long stretch to imagine putting LED glass into an automobile. When you pull into your garage it blasts up a message to a receiver. The next morning you get a message in GroupWise that says, "It has been 5,000 miles since your last oil change. Would you like to schedule an oil change? Just click here," and it goes out and schedules it for you.
Being able to have that happen, fundamentally changes the value proposition of your car and the value of that car and manufacturers who can do that will add new value to whatever the deals are by being able to hook them up to the network.
Take the microwave oven. If it was hooked up to the network, it wouldn't be limited to 200 food items. It could have access to a million different food items and how to cook them. Now, I was talking about this to someone else and I said, "You know, companies who make appliances might want to use this for diagnostic purposes, so in your dryer you have a computer and you call up and say, "Gee, my dryer doesn't work." They can go through the network and actually look at the dryer and determine what is wrong and perhaps tell you, "Why don't you try taking that sock out of the lint filter?"
But the whole proposition of this is if you can take those new nontraditional devices and link them into the power of the network, you can fundamentally transform people's conception of value and create entire new industries and entire new sets of computing value by being able to link up this whole world of computers with the network. What is equally clear is when you think about those computers of the future that look a whole lot different than the computers we know, like these here on the table, is that most probably they won't be running the same operating system.
One of the things that is very interesting about the computer industry is that there are all sorts of different views of this. One of the views, espoused by some people, is that for all computers to really be good they should all run the same operating system.
Now, it is equally clear to me every morning when I get into my automobile and turn it on, that I am really pleased that my automobile does not run under Windows 3.1. It is just so obvious, if you have your toaster and your microwave, you probably don't want it to run under Windows. You see, it is clear to me, it is not clear to everybody, but it is at least clear to me.
What is really important is that your view of computing in the future takes this into account. That this heterogeneous world will always consist of many different types of computing devices, running different types of operating systems that are linked together through the network. This is not simply true of the future, it is true of the past.
Most of the data systems that we work with, the companies who are our clients, there are almost none of them that have only PCs. Mainframe computers are a reality of life. They will always be part of the computer infrastructure along with Suns, along with HP machines running HP AIX, along with machines running Windows NT. Our vision of heterogeneous computing in this networked world has to encompass and fully support all of those devices, from IBM 3090s to toasters.
So, that is the key idea here. We are going into a new type of networked world. But, one of the things that is equally clear to us at Novell is that no single company can do it all. It is unrealistic to expect that any single company can fill the entire space and the entire needs of computing that will be enabled in this networked world.
To fully exploit what being connected means, will require partnerships of all types. Partnerships between the networking companies like Novell and communication companies like AT&T or PTTs around the world, partnerships between us and the people who make toasters and microwaves, or people who make automobiles.
To truly exploit the world of computing and the network world we are trying to build, will require a rich set of partnerships to exploit the opportunities that exist and it is really a very simple idea. There are too many creative people out there and believe me, I know. Every place I look, every place I go, I meet incredibly creative, talented people who don't work for Novell and that is the way it should be because this audience is filled with them. Because you are close to customers and you have ideas about how to creatively solve problems.
Novell's Product Strategy
The architecture of Novell's product strategy follows from its view of the fundamental principle of networking. Thus, the product strategy is divided into three distinct categories:
Systems and services that build the infrastructure within the smart, global network
Open programmer interfaces, enabling the network to integrate heterogeneous applications, systems, and platforms from all types of developers
Products that enable universal access to the global network.
Each of these product categories is briefly described below, with greater detail available in separate Novell white papers.
Figure 1: Novell's Smart Global Network.
Smart Network Services
The NetWare network operating system is at the core of Novell's strategy for building a smart global network. The NetWare OS is a high-performance system optimized for running network services. NetWare is the fastest, most reliable, most scalable, and best supported networking environment available today, and is the perfect platform around which smart network services can be built.
The first of these smart network services is NetWare Directory Services, or NDS. NDS is the nerve center of a NetWare network. NDS makes the network smart by knowing who you are, learning what resources you typically need, and linking you to whatever information or service you desire. For the user, NDS is like a dynamic Yellow Pages, providing an organized view of everything accessible through the network. For the network manager, NDS is the service that keeps track of, secures, and administers all objects on the network -- users, groups, devices, resources, servers, etc.
For the large enterprise, NDS reduces the cost of owning and operating a network. For the small business or independent workgroup, NDS is the technology that will support a seamless connection to the larger global network. With over ten million users, NDS is quickly emerging as the de facto standard for networks worldwide.
One of the reasons that NDS reduces the cost of operating a network is because it is integrated with other essential network services. Consider Novell's GroupWise network-based messaging product, for example. GroupWise is the industry's fastest growing groupware solution, providing unparalleled e-mail, calendar scheduling, and task management services.
Leveraging NDS for its master address book of network users and resources, GroupWise delivers a "universal in-box" for all types of personal and professional communications, including e-mail, appointments, faxes, and voice messages. The next generation of GroupWise will deliver conferencing, multimedia content, Internet connectivity, and workflow support.
Other smart networking services are available alongside NDS and GroupWise. File and print, security, transaction processing, licensing, management, remote access, host connectivity, database, routing, and telephony are just a few of the core networking services found within the NetWare family. Others, including distributed objects and electronic commerce services, will be available soon. Together, these services now comprise the backbone of 2.5 million workgroup and enterprise NetWare networks.
Linking these 2.5 million separate networks into one global network community is the role of NetWare Connect Services (NCS). NCS provides plug and play networking for organizations, and will be available starting in late 1995.
Through a number of partnerships with select telecommunications providers, such as AT&T, NCS will enable NetWare sites to connect their NetWare LANs securely to any other NetWare LANs, virtually anywhere in the world. Once connected, these sites operate together as one network, able to leverage the smart NetWare services for their cross-business communication and commerce.
Internet connections will also be an integrated part of the service, effectively eliminating the distinction between "the net" and local networks by seamlessly allowing users to access data and functionality resident anywhere in the networked world. For example, the NetWare Directory will map frequently accessed sites on the World Wide Web, so that Internet Web sites can become indistinguishable from other network resources managed and accessed through NDS. NCS will ultimately include worldwide dial-up and mobile access for individual users, completing the vision of a true global LAN accessible from any place, at anytime.
And for NetWare networks to get directly on the Internet, Novell will shortly make available a robust NetWare Internet Server, integrated with NetWare services for the most manageable Internet connection and Web publishing system available.
Universal Network Interface
Today, developing software that takes full advantage of the network is difficult. Developing software that takes advantage of a heterogeneous network is almost impossible. Net2000, Novell's revolutionary new set of APIs, will remove these obstacles by providing developers with a simple, universal programming interface to the diverse global network.
Net2000 is comprised of interfaces accessible through the C and C++ programming languages, along with OLE-based components compatible with all popular high-level programming tools, including Visual Basic, Delphi, and PowerBuilder. Through compatibility with developers' favorite tool sets, Net2000 will make building solutions for a global network as easy as building standalone desktop solutions today. Commercial, corporate, consulting, and content developers will find the process of developing Net2000-enabled solutions refreshingly straightforward.
And perhaps best of all, the Net2000 interface establishes a client-server standard which allows diversity both in client-based application solutions that access networks and in server-based network services comprising the network itself. Net2000 will preserve heterogeneity, allowing customers to select best-of-breed solutions from a broad variety of computing vendors.
The Net2000 interfaces will be consistent across all computing platforms, and will be made available on all popular desktops and application servers, on NetWare itself, and on select intelligent devices as well.
Among the more remarkable implications of global networking is that work can become an activity, not a place. To achieve this, the network must be as mobile as the user, and must be made transparent and easy, whether the user is seasoned or casual, in the office or at home.
Novell's family of Universal Access products will together provide individuals, businesses, and intelligent devices with connections to the global network at anytime, from any place. Fundamentally, Novell believes there are six overall methods by which people will access networks in the future:
Network client software that allows users to login to the network and share files and printers, from the office or on the road
Information browsers, such as Internet browsers, to search and view information residing in the global network
Groupware clients, enabling users to communicateusing e-mail, collaborate on projects, and take advantage of sophisticated document management and workflow services
Network-enabled office application suites, bringing the power of the network to traditionally stand alone word processingand spreadsheet users, and increasingly integrated with groupware,
A new world of network-enabled smart devicesthat allow people who don't use PCs to leverage the smart network
An unending range of third party network applicationsthat will be written to Novell's Net2000 programming interface
Novell Embedded Systems Technology
Of course, a key part of universal access is allowing nontraditional devices to access network resources. Our next initiative for Novell and Novell Embedded Systems Technology is really a very simple idea and that is: Can we just give manufacturers of all of these nontraditional devices a simple set of routines, that they can burn into an EEPROM, which allowed them with whatever operating system they want to design for this new nontraditional device with full access to network resources.
It is a very interesting idea, because it is clear that you would like to have these nontraditional devices make use of all of these network resources.
Let me give you an example. I carry around my Motorola Envoy. Once again, it's a little one pound device and it doesn't have a keyboard. Occasionally, I have a desire to send a letter from it. I might actually want to send you a FAX and I want it to come out with the letterhead and my signature on the bottom. Now, one approach to solving that problem and making that available is take all DOS, and all Windows, and all WordPerfect, and squeezing them together and try to get them into this very tiny device.
Now, there is an enormous amount of evidence that approach doesn't work. You can't take all of that stuff and make it work in a one pound device that doesn't have a keyboard and has a very small screen.
Our vision of solving this problem is fundamentally different. It says, "Why don't we allow this one pound device to access the whole set of resources that live on the network?" So, the case might be that if I want to do my letter, maybe what I do is I simply enter the text on this one pound device and then pick out a layout from a little list of layouts. Then send that text up to the network where it is laid out, spell checked, formatted, imaged into a document, given to a mail transport agent, which then faxes it out and delivers it to the right person. The intelligence itself lives on the network rather than in the device.
I give you another example. Take our car, the automobile now and most of the automobiles that you buy today are keeping diagnostic information. Maybe what the automobile manufacturers would like to do is to enable you to get, on a weekly basis, a graph of your miles per gallon. Lots of people do that manually, keep track of that information. Well, once again one approach is to take all of DOS and all of Windows and all of the charting package and squeeze them onto the car.
Most auto manufacturers are not interested in turning their automobiles into general purpose PCs. I don't know why, but that's what they told us. Our idea is to say to them OK, why don't you embed 100K of EEPROM that costs $100 or $50 (or probably $10 at their quantities) and an infrared transmitter and now what you can do is have your automobile computer link up to the network and simply spit up the data with some instructions about what's done and connect to the network and using network based charting package, which actually rolls the data out, does the graph, talks to a mail transport and then delivers it to your desktop.
The fundamental idea of NEST is to enable people with big applications to access intelligence which lives on the network.
The next key idea that this networked world starts to deliver is that the applications we think of today will evolve from being PC centric to distributed. Let me just tell you what that means. We once again have these narrow views of the world. Most applications that we need to know about run on a PC and occasionally they may make use of disk space that lives on the network. The next expansion of that are client/server applications where a piece of your application runs on your PC and a piece of your application runs up on the network.
But the world we're trying to build is far richer than that and not only can you have your application run on one computer on the network like the client/server application, but you can allow applications to run across a wide range of computers that live on the network and build an entirely new class of application.
If you want to get a real life example of what a true network distributed application is really like, just think about the World Wide Web. How many of you have played around on the Internet and have surfed the Web? Perfect. Well, just think about it for a second, you go to launch on Mosaic or Netscape and you browse around for 1/2 hour or 45 minutes. In that 45 minute period, most probably, you have accessed anywhere between 10 and 15, 20 different Web pages, then probably were hosting on anywhere from 5 to 10 different manufacturers' computers. Loading, probably, 3 or 4 different operating systems, from probably six countries around the world.
The value proposition of what makes the Web such an extraordinary application is its inherent distribution across the network and its ability to live on a wide range of machines all over the world. That's a true example of what a truly distributed application looks like and we believe that the world will be filled with these sorts of applications.
Part of what we are doing is building the infrastructure to make that happen. Our Tuxedo technology is exactly the type of technology that enables people to build these distributed applications. Tuxedo is basically an engine that allows you to describe an application that lives in multiple back-end computers.
So, you can say to Tuxedo, "I'd like to have this application and purchase a banking application. The part of the application that handles deposits runs on Windows NT and runs on this set of computers. The part that handles withdrawals was written as an NLM, and the part that handles transfers was written as a UNIX application." Tuxedo allows you to start that application up, to actually decide when a transaction comes in from any number of workstations, which of the background services it should send it to and automatically load balances and synchronizes the overlying databases that make that happen.
So with Tuxedo, if you have your deposit system running, as I said, on Windows NT, you can tell Tuxedo what you'd really like to do is have this deposit system running on four machines running Windows NT and Tuxedo automatically decides which of those machines to send it to based on load balancing algorithms. Or if you decide you need to move the application from one NT machine to another and graphically use the interface, you can grab that application and move it from one machine to another.
Tuxedo automatically shuts it down, starts it up on the other machine, and directs the right transactions to it. It is a fundamental technology to enable people today to start building these distributed applications in a heterogeneous environment.
It is clear to us that this Tuxedo, NCS, Network Directory Service, NetWare Embedded Systems Technology, and even our applications are forming the foundations upon which we build an entirely new world of computing. It is a world of network computing. It is a world which has no limits in terms of where you are or your ability to get to it. It is a world with no limits of what type of computers you can use to access it. It is a world of no limit of what type of servers that you run in it.
Our commitment is to help build the infrastructure to allow this world to occur, to support the widest range of computer devices to live within this infrastructure, to build the foundation tools such as directory services, security services, authentication services, as well as file and print and faxing and a wide range of new services to make this world real. Then all of you can start to build applications that exploit.
It starts with our foundation of millions of NetWare servers and tens of millions of people connected up to the network today and extends that to Bob's goal of a billion nodes of a network which include toasters, microwave ovens, automobiles, radio, cellular telephones, all built together in a networked world.
We believe that the opportunities to deliver value to people, to change the way that they live and they do their job will explode in a networked world. Our commitment is to work in partnership with all of you in determining how we exploit it. How we build the applications, how we deliver value and how we move forward to change the face of the world in the new network environment. Thank you very much.
* Originally published in Novell AppNotes
The origin of this information may be internal or external to Novell. While Novell makes all reasonable efforts to verify this information, Novell does not make explicit or implied claims to its validity.