I deal a lot with images, and I hear things about storing the binary data in a field in the database.

Does anyone actually do this? What are the advantages over simply storing it's path location as the field, and then just reading it off the disk?