Each file is divided into two parts. The first part, which contains descriptions of the objects in the file, is represented entirely in ASCII so that it can be easily viewed and even edited using a text editor. One can, for example, add a comment annotating an image by simply editing that part of the file. The second part of the file holds information that must be stored compactly, such as image pixel values. There data is represented efficiently with a binary encoding that is independent of machine byte order.
Vista's data file format can be readily extended. You can record custom attributes with each of the images stored in your data files yet still manipulate those images using the existing Vista software. You can also define custom types of objects (e.g., an object describing a set of model parameters). These custom objects can be included in data files alongside standard ones such as images and edge vectors.