The basic idea is to store all the data items in the Name space, when access and operations with a data item can be represented as names. In addition, for managing names, the Name space can be organized by using a directory which is analogical to directories in file systems.
Directories
A directory is seen as any other data structure in the Name space. Analogical to file systems, each directory excluding the root has a root entry named by "." and a link to upper directory named by "..". The separator in the names between the levels of the directories is "/" and the root of the whole directory tree is named as "/". Thus the distinguished path can be defined by collecting all the subdirectories, for instance "/lev1/lev2/".
Note that operations do not create automatically new subdirectories for result data structures. However, you can access data structures for inputs in different directories by using absolut or relative paths.
The directory structure of the NDA is not the same as the file systems has. In the NDA all the data items are stored only in the memory.
Frames
Fields can be stored at the directory or frame level. At the directory level they can be seen as any other entries, but at the frame level fields are seen only through frame entries. This has an effect on naming, because the same names for fields are available at a directory level if they have been hidden behind frames.
Items inside frames are always named as <framename>"."<itemname> and, thus, they cannot be seen directly at a directory level. The separator between the name of the frame and its items is ".". If a frame includes other frames, which can further include data items, fields can be accessed via the whole name, for instance, "frm1.frm2.item1".
The scope of the data in the NDA is defined through names when access to data items is defined by unambiguous names. Thus, in one subdirectory the same name can appear only once as well as inside a frame.