BOT2001 Report
Defining and Creating Bots

By Brian Proffitt

Whenever someone would ask what was really a bot and what was an intelligent agent at the BOT2001 conference, invariably someone would answer, "go talk to Pallmann."

"Pallmann" is David Pallmann, the chief technologist for NQL Inc. and the author of Programming Bots, Spiders, & Intelligent Agents in MS Visual C++ (MS Press, 1999). And as the author of a mainstream book about bot technology, a lot of people listen to what he has to say.

Not that writing a book automatically commands a lot of respect. But Pallmann would certainly earn this anyway, given that he is the inventor of the Network Query Language (NQL), the product upon which his company is based.

Pallmann was one of the speakers at this year's BOT2001 seminar, and began his talk with his definition of what bots are.

To Pallmann, a bot the equivalent to an automated program, or "any software that doesn't require constant human intervention."

In contrast, Pallmann's definition of an agent is a bot that serves a specific body, whether it be a person, department, or organization. Agents, he explained, are aware of the the needs and wishes of their masters and are skilled in specialized areas.

Intelligent agents, Pallmann added, as simply agents that are worth having. By being able to automate tasks with a minimum of human intervention, intelligent agents are very distinguishable from traditional software that does require more human interaction.

This automation is a key factor in gaining public acceptance for bot technology.

"We have found that automation is rejected by users unless it is very, very intelligent," Pallmann told his audience.

Public acceptance is still up in the air at this point in bot history, since there are many examples, both good and bad, of bot activity.

On the negative side, bots raise serious privacy issues, can interfere with Web server operations, and, Pallmann reminded his audience, computer viruses are actually bots themselves.

Still, many positive factors favor the development of bots, particularly because they are desperately needed to contend with the huge amount of information and data that exists on the Internet.

Given that we live in world where bots are more necessary than not, Pallmann has a unique perspective on the future of bot development. Using his NQL programming language, developers and non-programmers alike can easily construct autonomous applications.

NQL is similar to SQL, in that it too is designed to seek and acquire data across a given platform. In SQL's case, the platform is a database, while for NQL, the platform is a network--perhaps even the Internet itself.

Pallmann touched on some of the technical aspects of NQL, emphasizing that its modularity was how anyone would be able to use it to create bots without a great deal of programming knowledge. One of the big advantages to using NQL is that fact that it is not tied to one specific platform, which enables it to query data from almost any source.

NQL will have more uses than bot applications, Pallmann predicted. Citing NQL Inc.'s new ContentAnywhere content management application, which uses NQL as a primary tool for acquiring data

Pallmann invited the audience to download NQL 1.0 from the company's Web site, as well as view the ContentAnywhere demo site.