I wrote Data::Serializer several years ago to standardize the interface and add some functionality on top of serialization in a consistent and easy to use way. What this means is that the credit for the hard core work of serialization belongs with those authors, but it makes me happy to see that the abstraction layer I've created for the process has been so useful to so many people. I'd love to hear what any of you out there are using the module for. I've personally found having a very convenient method for storing very complex data structures makes certain kinds of IPC very easy to put together - process A serializes data, writes it to file.tmp, when done renames to file.done, process B looks for file.done, renames to file.working does it's magic and unlinks the original file. Of course there are many different ways one can do such a task - this is just one I've found handy for some tasks.