It's purpose is to make machine friendly data easier to understand by humans that are looking at it. Specifically multidiff helps in viewing the differences within a large set of objects by doing diffs between relevant objects and displaying them in a sensible manner. This kind of visualization is handy when looking for patterns and structure in proprietary protocols or weird file formats. The obvious use-cases are reverse engineering and binary data analysis. At the core of multidiff is the python difflib library and multidiff wraps it in data providing mechanisms and visualization code. The visualization is the most important part of the project and everything else is just utilities to make it easier to feed data for the visualizer. At this time the tool can do basic format parsing such as hex decoding, hexdumping, and handling data as utf8 strings, as well as read from files, stdin, and sockets. Any preprocessing such as cropping, indenting, decompression, etc. will have be done by the user before the objects are provided to multidiff.