Seadragon technology is cool, but to any InfoVis connoisseur, the concept or design of Pivot isn’t really ground-breaking. Two major problems need to be resolved before InfoVis systems can become “killer apps”. First, what benefits does a zoomable, filterable, pivotable, info-graphical interface provide in comparison to text-based interfaces? “The whole is greater than the sum of the parts” is certainly a good formulation, but we need more concrete scenarios and real-world applications. Second, an implicit assumption of most InfoVis systems is that users have some ready-to-be-visualized datasets, and this assumption is building on a less solid ground. Unstructured data is everywhere, and getting them cleaned up and usable is a huge problem. Pivot provides tools for voluntary developers to create collections , which are data in a format understood by Pivot. This approach is reasonable, but may not be enough.