Spatial navigation, which is already being used in some software products, will let you go where the information is by enabling you to interact with a visual model of a real or make-believe world. You can think of such a model as a map—an illustrated, three-dimensional table of contents. Spatial navigation will be particularly important for interacting with televisions and small, portable PCs, which are unlikely to have conventional keyboards. To do some banking, you might go to a drawing of a main street, then point, using a mouse or a remote control or even your finger, at the drawing of a bank. You will point to a courthouse to find out which cases are being heard by which judges or what the backlog is. You will point to the ferry terminal to learn the schedule and whether the boats are running on time. If you are considering visiting a hotel, you will be able to find out when rooms are available and look at a floor plan, and if the hotel has a video camera connected to the highway, you might be able to look at its lobby and restaurant and see how crowded it is at the moment.
You’ll be able to jump into the map so you can navigate down a street or through the rooms of a building. You’ll be able to zoom in and out and pan around to different locations very easily. Let’s say you want to buy a lawn mower. If the screen shows the inside of a house, you might move out the back door, where you might see landmarks, including a garage. A click on the garage will take you inside it, where you might see tools, including a lawn mower. A click on the lawn mower will take you to categories of relevant information, including advertisements, reviews, user manuals, and sales showrooms in cyberspace. It will be simple to do some quick comparison shopping, taking advantage of any amount of information you want. When you click on the picture of the garage and seem to move inside it, behind-the-scenes information relating to the objects “inside” the garage will be fed to your screen from servers spread over thousands of miles on the highway.
When you point at an object on the screen to bring up information about the object, you are employing a form of “hyperlinking.” Hyperlinks let users leap from informational place to informational place instantly, just as spaceships in science fiction jump from geographic place to geographic place through “hyperspace.” Hyperlinks on the information highway will let you find answers to your questions when they occur to you and you’re interested. Let’s say you’re watching the news and you see someone you don’t recognize walking with the British prime minister. You want to know who she is. Using your television’s remote control, you will point at the person. That action will bring up a biography and a list of other news accounts in which she figured recently. Point at something on the list, and you will be able to read it or watch it, jumping any number of times from topic to topic and gathering video, audio, and text information from all over the world.
Spatial navigation can also be used for touring. If you want to see reproductions of the artwork in a museum or gallery, you’ll be able to “walk” through a visual representation, navigating among the works much as if you were physically there. For details about a painting or sculpture, you would use a hyperlink. No crowds, no rush, and you could ask anything without worrying about seeming uninformed. You would bump into interesting things, just as you do in a real gallery. Navigating through a virtual gallery won’t be like walking through a real art gallery, but it will be a rewarding approximation—just as watching a ballet or basketball game on television can be entertaining even though you’re not in the theater or stadium.
If other people are visiting the same “museum,” you will be able to choose to see them and interact with them or not, as you please. Your visits needn’t be solitary experiences. Some locations will be used purely for cyberspace socialization; in others no one will be visible. Some will force you to appear to some degree as you are; others won’t. The way you look to other users will depend on your choices and the rules of the particular location.
If you are using spatial navigation, the place you’re moving around in won’t have to be real. You’ll be able to set up imaginary places and return to them whenever you want. In your own museum, you’ll be able to move walls, add imaginary galleries, and rearrange the art. You might want all still lifes to be displayed together, even if one is a fragment of a Pompeian fresco that hangs in a gallery of ancient Roman art and one is a Cubist Picasso from a twentieth-century gallery. You will be able to play curator and gather images of your favorite artworks from around the world to “hang” in a gallery of your own. Suppose you want to include a warmly remembered painting of a man asleep being nuzzled by a lion, but you can’t recall either the artist or where you saw it. The information highway won’t make you go looking for the information. You’ll be able to describe what you want by posing a query. The query will start your computer or other information appliance sifting through a reservoir of information to deliver those pieces that match your request.
You will even be able to give friends tours, whether they are sitting next to you or watching from the other side of the world. “Here, between the Raphael and the Modigliani,” you might say, “is a favorite finger painting I did when I was three years old.”
The last type of navigational aid, and in many ways the most useful of all, is an agent. This is a filter that has taken on a personality and seems to show initiative. An agent’s job is to assist you. In the Information Age, that means the agent is there to help you find information.
To understand the ways an agent can help with a variety of tasks, consider how it could improve today’s PC interface. The present state of the art in user interface is the graphical user interface, such as Apple’s Macintosh and Microsoft Windows, which depicts information and relationships on the screen instead of just describing them in text. Graphical interfaces also allow the user to point to and move objects including pictures—around on the screen.
But the graphical user interface isn’t easy enough for future systems. We’ve put so many options on the screen that programs or features that are not used regularly have become daunting. The features are great and fast for people familiar with the software, but for the average user not enough guidance comes from the machine for him or her to feel comfortable. Agents will remedy that.
Agents will know how to help you partly because the computer will remember your past activities. It will be able to find patterns of use that will help it work more effectively with you. Through the magic of software, information appliances connected to the highway will appear to learn from your interactions and will make suggestions to you. I call this “softer software.”
Software allows hardware to perform a number of functions, but once the program is written, it stays the same. Softer software will appear to get smarter as you use it. It will learn about your requirements in pretty much the same way a human assistant does and, like a human assistant, will become more helpful as it learns about you and your work. The first day a new assistant is on the job, you can’t simply ask him to format a document like another memo you wrote a few weeks ago. You can’t say, “Send a copy to everybody who should know about this.” But over the course of months and years, the assistant becomes more valuable as he picks up on what is typical routine and how you like things done.
The computer today is like a first-day assistant. It needs explicit first-day instructions all the time. And it remains a first-day assistant forever. It will never make one iota of adjustment as a response to its experience with you. We’re working to perfect softer software. No one should be stuck with an assistant, in this case software, that doesn’t learn from experience.