Helping communities define themselves on-line

With the explosion of the Web it seems that every Tom, Dick and Harry has a Web site, but how does an entity as diverse as a community create an on-line version of itself?

And once having done so, how can the site be easily maintained and altered?

A group from the University of Waterloo is working on that problem, and has come up with the Community Network.

The idea is to create a virtual network which provides a wide array of information about a community from the community’s point of view, as distinct from the interpretation of a search engine. “If you use Yahoo to search for K-W (Kitchener-Waterloo) hospitals, you only get St. Johns’ Ambulance,” said Paulo Alencar, a research associate professor at the University of Waterloo in Waterloo, Ont. Community Networks allow a community to “expose itself according to its own view,” not that of Yahoo, said Alencar, one of the architects of the K-W Community Network.

At a recent talk at the University of Toronto, hosted by Communications and Information Technology Ontario (CITO), Alencar talked about “producing methods and abstractions for designing, building and maintaining complex evolving component-based software systems” which could ultimately be used to help create these community networks.

Creating a framework for a community network is no easy task since the system must be easily extended to match the natural growth and change of a community. “We tried to use simple off-the-shelf software to make the connections between the components in the system,” Alencar said in a later interview. According to the Computer Systems Group at the University, who designed the framework for community networks, the idea was to create “a complete open system architecture based on current COTS (commercial off the shelf) software.”

The framework is defined as a “set of components and rules for collaboration”, thus allowing the “building and reusing [of] software for a specific application domain.” The components included everything from databases and browsers to search engines and hypermaps.

Alencar defined a framework more specifically as “a system which is not the solution for one specific application but a solution for a class of applications.” Ultimately one could use the same framework for many situations with only a few small changes. Thus, communities starting up their own network could use the same basic template to get started.

One interesting aspect to the test site that has been developed ( is the use of hypermaps, which are referred to as “a geo-referenced multimedia data sets.” A map can be viewed as a collection of objects where each object can have several types of associated data stored in a SQL database, according to the U of W group.

One can “use the maps to link and order multimedia components and allow queries to the database,” Alencar said. “The hypermap by itself can be defined as a framework and in the map you have many different objects and each object can be shown in different ways,” Alencar said.

Each object can represent a building, a park or an arena, and all the data associated with it. The object can also be linked to a variety of other sites.

For this type of work, HTML coding is limited since one link is limited to one document. The U of W group tagged the database information with SGML or XML, which is much more versatile. This allows designers to create customized tags to provide specific functions. Unfortunately, many people are using browsers that cannot read XML. Using a COTS component, Janna Systems Inc.’s LivePage, the data is retrieved from the database and then converted to HTML for the browser. Ideally, a browser would be able to read the XML code directly and in the future this may be the case.