By Damon Judd

OK, so maybe it's not a fad or a passing fancy. Everyone these days seems to be clamoring to serve up their geospatial data via the Internet. Some of the reasons why are obvious-simple, low-cost access to maps and data, wide distribution across multiple platforms, ease of use, and the ability to share information with a wide and diverse audience of stakeholders without the headaches of installing and learning a complex application. After all, just about everyone can surf the Net, right?

But what does it take behind the scenes to make it possible to view maps, images, and database information via the Web? For one thing, it takes standardized methods to serve up and transmit geospatial data over heterogeneous networks with varying bandwidths and data transmission protocols.

With the current widespread acceptance of standard protocols, languages, and Web services such as HTTP (HyperText Transmission Protocol), SOAP (Simple Object Access Protocol), Java, and XML (eXtensible Markup Language), software vendors now have the ability to turn Web browsers such as Netscape and Internet Explorer into widely available desktop mapping applications. Now a standard Web browser can be used for viewing and querying large amounts of geospatial data using tools that require only a minimal change to the user experience (see Figure 1).

One example of a standards-based solution is the product developed by ESRI for this purpose, the ArcIMS (Internet Map Server) software tool. It was developed using Java, and an extension to XML (ArcXML), and can be deployed using an existing Web server such as Apache, Microsoft IIS, Sun ONE, WebLogic, or OracleAS.

Because many organizations have already made the investment to establish and support a Web server and are likely to have some expertise in developing and implementing websites, the incremental investment to add spatial servers is no longer the obstacle that it was perceived to be just a few years ago. Furthermore, many organizations such as some counties and large cities have been collecting and building geospatial data for their constituents and need to make those datasets accessible to a wide audience. In San Francisco, the SF Prospector application is an example of that need-SF Prospector speeds up and simplifies the process of finding the optimal location for a business in San Francisco by providing public access to a range of valuable information (see Figure 2).

On the other hand, there are still some technical challenges associated with serving up geospatial data over the Internet. Performance of the client application (e.g., the amount of time it takes to refresh the map display) is potentially a big one. Raster data in particular tends to use up lots of storage space and therefore can lead to serious performance degradation, especially when those data need to travel across low bandwidth transmission "pipes." After all, not everyone with Internet access has a T1 or T3 connection. One solution to this issue is the use of image compression schemes such as ECW (ER Mapper Compressed Wavelet format), MrSID, and PNG.

Image compression using wavelet type mathematical transformations applied to the image data, dramatically reduces file size while maintaining optimum image quality. Humans usually relate image quality to optimum visual perception. Various types of image compression wavelet transformations (e.g., MrSID and ECW) have been developed that work in various ways. Specialized "fast" wavelet compression applications have been developed for allowing users to browse large image files over the Internet. The most efficient processes developed compress only the part of the image requested by the user using special ancillary software operated through the client's Web browser.

PNG is an extensible file format for the lossless, portable, well-compressed storage of raster images (see Figure 3). PNG provides a patent-free replacement for GIF and can also replace many common uses of TIFF. Browse to: http://www.w3.org/Graphics/PNG/ for more information.

Security issues present another technical hurdle. By making the database server that stores geospatial data available to the public via the Web server, access to private and confidential data is potentially opened up to hackers or other forms of unauthorized access. Fortunately, many security issues with websites have already been addressed and are essentially no different for Web-based GIS applications than they are for any other Web-enabled database application. The use of encryption, firewalls, multiple levels of user authorization, and other similar techniques are fairly common with many existing Web servers and database servers.

Another newly emerging standard that is likely to significantly influence the deployment of vector map data over the Web is the SVG (Scalable Vector Graphic) format. Some large software vendors including Adobe, Apple, Canon, Corel, Hewlett-Packard, Macromedia, Microsoft, Kodak, and Sun have contributed to the standard specification and are supporting its further development. Browse to: http://www.w3.org/ Graphics/SVG/ for more details.

Why Use SVG?

SVG is an open standard XML format that greatly extends the potential of graphics-driven application development for extranets and intranets. In contrast to other formats, SVG offers:

• a remarkably rich graphics language

• the ability to create static, dynamic, and interactive documents

• wide compatibility with other XML technologies and existing standards

• the ability to be viewed on a wide range of platforms and devices

Like the PDF document format, I predict that SVG will soon emerge as a common approach to translate, transmit, and display vector map data over the Web quickly, simply, and with considerable flexibility.