Dataset name

A dataset name, often abbreviated DSN, identifies a dataset on the system. It is made of segments called qualifiers, separated by full stops, with each qualifier up to eight characters and the whole name up to forty-four.

A typical name looks like a short path read left to right from most general to most specific: an owner or project at the front, then something describing the application, then what the dataset actually holds. Names are uppercase, start with a letter, and cannot contain spaces.

The convention is not decorative. The first qualifier, the high-level qualifier, usually determines both who owns the dataset and how the security rules apply to it. Sites define naming standards precisely so that access can be granted to whole families of datasets at once.

When you type a name in ISPF, a quirk catches everyone out early: if you leave it unquoted, your own user ID is added to the front automatically. Putting the name in single quotes uses it exactly as typed. That single rule explains most "dataset not found" confusion in the first week.