DataWeek 2013 Conference & Festival

I often get asked how many nodes or edges are in Gravity’s proprietary ontology. It's a common way that people judge the "quality" of an ontology. On the other hand, I’m rarely asked how our edges are weighted, the number of weighted edges, or if they’re weighted at all. As the amount of data, entities and raw facts being generated on a daily basis grows exponentially, we need to begin to take a look at whether dumping terabytes of new facts into our ontologies is the correct way to deal with the situation. Most publicly available ontologies have no concept of weighted edges, yet continue to focus resources on increasing the number of entities and attributes associated with them. I believe we need to turn the problem upside down, and as an industry that is well aware of the potential power of a truly semantic web, begin to take a different approach.