The Future Favors GIS

By Jack Dangermond

Editor's Note: The following is based on part of Jack Dangermond's opening address at the Nineteenth Annual Esri International User Conference in San Diego, California, July 26, 1999.

diagram showing various ways in which geographic knowledge will be important in the 21st CenturyGeographic information is making a difference today and will make a huge difference in the future.

Why is this happening?

Widespread adoption of GIS has value beyond simple efficiency or profitability or even communication.

It's my feeling that the future will favor GIS.

If we look at the major forces of the 21st century--population growth, global economic development, consumption of natural resources, the rapid pace of scientific discovery, and the even more rapid evolution of technology--we discover that geographic information science and GIS technology are fundamental to all of them. As GIS gets easier and better, there will be a national tendency to use it for more and more of our problems. I am now beginning to realize that GIS will be an important (perhaps the most important) technology for creating, applying, and disseminating geographic knowledge.

a diagram of a GIS System architectureAs GIS technology evolves, geographic information will be everywhere, embedded into most information applications and services that society uses. I think we can see most of the dots and how to connect them, but we have only recently begun to understand the implications of what is emerging for GIS.

Those of you working in the GIS field will be involved in the advancement of geographic information science, you will be collecting and managing the geographic information, and you will be advancing GIS technology; you'll have a lot to do with managing the future.

The future will also favor those organizations that successfully manage their information and knowledge; doing so will be even more critical in the future.

The New World of Digital Information

a diagram showing in what areas the future will favor widespread GIS adoptionAs we collect and store more of our information in digital forms and as we provide better information access to everyone in our organizations--perhaps even open access--perhaps public access--our organizations are better prepared to succeed. And we become more accountable.

This kind of free access to information will change us. We are already seeing organizations that adopt these kinds of changes become more effective in their decision making.

While you can read about how big corporations such as IBM and Microsoft have adopted this approach, I can tell you from personal experience how it has been at Esri.

After struggling for years with the problem of managing our own information and trying several different solutions, a few years ago we acquired a large back office information system and implemented it--so I know what it's like to be on the inside of such a project. It was painful. We had to reengineer our work flows a lot and change the way we did things. But the result is that we are concentrating now on what the information is telling us--not just on collecting the information.

People within the organization now know better what's going on. We share information quickly across our network, and we can make decisions much more quickly, effectively, and rationally than before.

How does this personal experience relate to GIS technology?

As GIS becomes more than a bunch of projects or small efforts, as GIS becomes an integral part of organizational and societal information systems, things change radically.

We have the means and the structure for measuring change on any scale, even at the global level.

We have a framework for analyzing the information, for doing geoaccounting, for supporting decision making--on any scale.

We can integrate this information into complex work flows: planning, engineering, and environmental assessment.

GIS provides these fundamental elements that are key to any information system: geographic measurement, geoaccounting, analysis, integrated decision making, support for coordinating work flow--and GIS is also a remarkable visual spatial language. All these capabilities are evolving rapidly. Let's examine some of these more carefully.

a diagram showing how all geographic change will be measuredGIS provides us with a kind of framework for systematic measurement of geography. I think that in the future all geographic change will be measured by various kinds of instruments, and then the measurements will flow into information networks where they will be made available to everyone--just like Esri's big back office system makes information available. I think this flow of information will transform societies just as it transforms organizations. As we share this common knowledge we become more effective.

Similarly, GIS is evolving into a technology that can help people plan, design, engineer, build, and maintain the built environment. As planning, engineering, and operations share this common knowledge, more things will receive fair consideration. Ian McHarg, a landscape architect, always criticized the engineering community for not adequately considering the whole. GIS technology allows us to integrate what we know into the flow of our work so that the whole is automatically considered in whatever we are doing.

This implies very large spatial databases; some people are a little fearful of that. But I'm not; I'm encouraged because these large spatial databases will give us significant advantages.

diagram showing that More Things Will Be ConsideredThese large GIS databases offer us the opportunity to move toward a kind of geographic accounting system or geoaccounting.

As we implemented our back office system we were able to account for all kinds of things more effectively. We began to understand what worked and what didn't work much more quickly. And we could discard the things that didn't work much more quickly.

a diagram showing that things that work will quickly be adopted, thus accelerating changeGIS will provide us with a whole new way of accounting--not just for the economic accounts of an organization, but also for the economy of a country, for its economic development, and for a country's biodiversity, its environmental protection efforts, its culture, and its national health.

But it could also be geoaccounting by community or by state or by region or for just my neighborhood.

I'll be able to see what's going on. That will make me more responsible and help make my government more responsive to me, and help make us all more responsive to one another.

Geographic information and GIS methods and techniques, such as probability, modeling, visualization, and exploration of data, will affect and advance geographic science.

a diagram showing that Systematic Analysis Will Lead to New UnderstandingsI think of geography as a kind of mother of all sciences: geology, oceanography, anthropology, and sociology. Geography is the integrative science. It affects all the others.

As we link our work to these large, integrated databases that are shared, people will explore the data, analyze it, and find new meaning in the patterns they discover. It's almost the reverse of most other kinds of scientific investigation; instead of narrowly focused research we do data mining on the vast spatial data resources, which leads to discovering new patterns and relationships and ultimately to new scientific knowledge.

GIS is also providing us with a framework for modeling. This idea of spatial modeling--that we can work out various kinds of flows, link processes together, examine the past, or project a possible future--is just now evolving.

I think that geographic information is going to be integrated into everything we do. It's just going to show up. In the information society, the use of geographic information will be one of the main mechanisms we use for communicating with one another.

The Geography Report

I like to think of the possibilities of a "geography report"--like the weather report.

a diagram showing how GIS will use DBMSsPerhaps, like the weather channel, there will be a geography channel. All the monitoring data and all the analyses will be reported. Perhaps you'll wake up in the morning and get the latest reports on change--not just in your neighborhood or city or state, but in the whole world. And we'll know what the problems are everywhere.

The geography report will be produced as we connect all the sensors in the instrumented universe and all the information they collect flows through the information network.

That will be one kind of "geographic knowledge everywhere."

I believe that organizations and societies that make all their geographic information available in this way will become more efficient and more successful.

It's a kind of evolution; with adequate information, things that work will be recognized and more widely adopted. Things that don't work will be recognized and discarded. In this environment I think that sound new ideas will flourish, effective new methods will be adopted, and change for the better will accelerate.

That needs to happen if we're going to deal successfully with the problems we face in the future. We're going to have to get more efficient and make decisions more carefully and more wisely.

I think the bottom line is that GIS use will transform our organizations and our societies. It will help us improve our methods and do better science. We'll be more accountable. We'll make better decisions. We'll be a more geographically conscious society, able to consider problems in a holistic way.

Geographic accounting--making decisions wiselyGIS will provide the geographic knowledge for the 21st century, and those working in the GIS field will be the ones who do it. That's why this work is so important; we have to understand that and be conscious of that context in all the work we're doing.

Is this a practical vision? Perhaps. But most of the databases we've been building are focused on specific projects, specific pieces of the globe; they're not the kind of multipurpose data infrastructure I've been talking about. If we try to overlay all the data in all the thousands of existing GIS databases we'll discover that we can't really do it; the data is inconsistent; we don't agree on reality. Even after years of working on the U.S. national hydrological network, there's still disagreement. There are inconsistencies in the way we think. These inconsistencies are going to cause us problems.

I don't know exactly what the answers to these problems are, but I believe that the future will favor investments in spatial databases that are designed and built with multiple uses of the data in mind. Good designs will have only minimal structure focused on a particular application, and they'll have vertically integrated classification systems. The future will also favor spatial databases that are linked to work flow so that the data is automatically maintained.

A Spatial Data Infrastructure

The future will favor those spatial databases with agreed on, documented, standardized data models, databases that provide metadata and that open up and share their data sets.

As a community we need to promote investment in these databases; we need standardized means of data collection without limiting innovation; we need standardized ways to do modeling--and ways to share models; and we need to look at standardized ways to deliver cartography so that people who look at maps get the content quickly.

Of course a GIS requires user focus and good data, good organization, good people, and technology that can support it. But it is also my feeling that we need to stop thinking of GIS as just a project and start thinking of GIS as part of a kind of coordinated information system.

Computers are scalable and networked and more mobileOur spatial data assets are growing. Their value probably exceeds $100 billion. I know that the value of USGS's spatial data assets is estimated to exceed $20 billion, and that's just one organization. And the rate of data collection is increasing: regional data, national data, and global data.

There is growing interest in coordinating our spatial data efforts. Standards committees are working to coordinate them, the FGDC is trying to coordinate them in the United States, and there are similar efforts going on throughout the world.

What would a spatial data infrastructure require? It requires standardized data sets, a network of providers, appropriate administrative mechanisms, and financial support. It requires a framework of policies for dissemination and sharing. A spatial data infrastructure requires sound organization and technology that can support an open, multipurpose, multiparticipant system. And it requires leadership.

Now I don't know how this all comes about in organizational settings, but I do know that we have many fine footprints--organizations that work--that those in the GIS field have already created.

Successful organizations, those that have lived a decade or two and those that are mature, have good leadership. They understand and serve their users, they manage and integrate their data well, and they nurture their technical people. They're supported by management, they know how to cooperate, but they own their own destiny; that means not leaving it in the hands of outsiders.

Esri has GIS software for both desktop and server architecturesThis year at the User Conference the Making a Difference award was given to Larry Sugarbaker at the State of Washington. He and his wonderful staff have been able to create that kind of organization. It's part of the infrastructure that makes that State work.

If GIS is to become part of the information infrastructure, this is the kind of organization we'll need.

We need good people. GIS organizations require them. We need people who know about geography, who know about more than just the mechanics of a computer, who know about GIS applications and how to create them and how to support their users. Good people create a culture of sharing and cooperating--a culture based on sharing information, understanding what people want and need, and then helping them. This takes special people.

These people are beginning to be recognized by upper management in the United States. At the Cabinet level they're making the decision that GIS is turning out to be important and somebody had better be in charge of it. We haven't had that kind of acknowledgment in the past. There have been CFOs for money and CIOs for managing the rows and columns of a database; but recently the USGS announced the appointment of a GIO--a geographic information officer. I like that idea a lot.

Many of you reading this are going to fill similar positions in the future.

Systems in Transition

Today's software is being developed using object-component technologyLet's talk about the technology for a few minutes. It's changing, and it's also changing the whole concept of information infrastructure as a result.

We are moving from project and departmental systems to systems that are designed for multiple users, multiple applications, and multiple purposes. The underlying technology is evolving into networks--Intranets or the Internet--with more powerful machines, more powerful software, more powerful communications.

In the 21st century these technological changes will occur even faster, accelerated by the integration of computing, networking, and communications. It's all coming together.

We know all this. We read about it everywhere. There's information overload about it.

But it's worthwhile to pause a moment and see what it means for the future of spatial infrastructure.

Computers are becoming more connected through open networks, TCP/IP, and HTTP networks. The networks include everything from very big machines--servers--to very tiny ones that we literally wear on our bodies, and this will continue to evolve.

The standards to support this--UNIX and Windows and Java and COM--are becoming interoperable so that what we do on one machine can talk with what we do on another machine.

ArcGIS is reusable, embeddable, and extendableGIS is beginning to use databases that are also becoming scaled--from very small personal databases that are inexpensive to very large enterprise databases for work groups.

These databases are becoming more spatially literate in the sense that they can store and manage geographic dimensions and spatial information more generally.

There are two basic computing environments or domains: the desktop-centered and the server-centered worlds, the latter with thick or thin clients, centralized data management, and uniform service throughout the network.

As these architectures continue to evolve, the communications networks are getting much bigger; in the last year the Internet has tripled in capacity, and we're now seeing an order of magnitude increase in network performance in Intranet implementations.

These changes will continue, and I think they will bring these two computing environments closer together. And GIS will be part of it all.

GIS will exist in both domains: on desktops, database servers, application servers, and in browsers as plug-ins. It will be embedded. GIS will be based on a component architecture, components that can make up a complete system or can be deployed independently or interdependently.

Today's computer software is also changing; it's moving away from procedural tools and large chunks of code--to very small components. The component architecture standards are Java and COM. We've adopted these standards at Esri; they're becoming ways to reengineer all of our tools. That has given us the opportunity to embed these tools and craft new kinds of end user solutions.

GIS data and GIS technology are already being embedded in all sorts of things--in the future they will really be everywhere. SAP, for example, is going to embed GIS in about 20,000 back office systems, so GIS will be in most of the Fortune 500 companies. GIS and the data many of you provide will be inside these organizations.

Microsoft is working to put maps inside consumer products. GIS and the data you provide will be inside those products as well.

NIMA is releasing ship navigation products that depend on GIS and GIS data.

Are all of these applications GIS? No, probably not, but they embed geographic knowledge and the data and tools to leverage it, and I believe that in the future we'll see even more embedding and distribution of geographic knowledge in these ways.

And GIS software? Well, we've been busy at Esri. GIS software is getting much easier to use, we've introduced some new and interesting methodologies and tools, we have a richer data model that makes knowledge more accessible, we're doing Internet work, and we have a strong and enriched data management technology.

Our efforts have focused on about a dozen areas of software development: usability, software architecture, development environment, spatial analysis, modeling, cartography, data management objects, database models, metadata standards, interoperability, and dissemination of knowledge on the Internet.

We focused on these areas because our users have told us, year after year, in surveys and in other ways, that these were the things they wanted us to focus on: make it easier, more productive, more fun.

So, for Esri's part of this effort to create the future, we'll continue to follow your guidance.

Working together, I think we have a great future ahead of us!

ArcNews home page

Contact Us | Privacy | Legal | Site Map