July Meeting of DC XML Users Group

Dave Sullivan, of Zonar, spoke about the limits imposed by existing DTDs and schemas upon XML’s ability to encapsulate data and allow for its optimal use by existing data systems. Using his work for realtors as an example, Sullivan made clear the almost unlimited number of qualifiers you could have for any element. He gave as an example: a house, a house with a ceiling fan, a house with a ceiling fan by a specific manufacturer.

Sullivan emphasized the importance of establishing conventions concerning definitions and adhering to them.

One thing he said that caught Technoflak’s ear was that to design proper systems, “you have to know the industry, you have to know what you are talking about.”

After the break, Tom Passin, author of the Explorer’s Guide to the Semantic Web, gave an overview of the ideas presented in his book.

He described four views of the semantic web: the machine readable data view, the intelligent agent view, the distributed data base view, and the servant of humanity view (Tim Berners Lee’s view). Passin summarized the ”servant of humanity” view as ”machine consumption of web data + more semantics = good things for people.”

He described the nature of the web as ”open, huge, heterogeneous, evolutionary, and not necessarily benign.”

Passin described Friend of a Friend networks at length and how they might work with intelligent agents to retrieve information. At this point a member of the audience said, ”This is an example of why you need to know more about serialization than you think.” To this Technoflak would only add, precisely so. The ability of intelligent agents to crawl the web and retrieve all manner of information is a perfect example of why software development is too important to leave to software developers.

Passin talked about Resources Descriptive Framework, which consists of a thing, its trait or aspect, and its value. It is like the grammatical construction of subject, verb, predicate. Passin observed that many view Resources Descriptive Framework as the basis for the semantic web. Passin said that, ”real logic people regard the semantic web as ridiculous”, and he described them as ”pretty intolerant in discussion lists.”