Quick-Start Introduction to PDS Archiving

OBJECTs

An OBJECT (in the PDS sense) is a description of a data structure which contains both logical and physical descriptions of the data. For example, in a PDS file label the OBJECT definition will include information about the type (integer, real, etc.), size in bytes and offset of each field in the record, in addition to defining what that field is logically (a pixel value, a temperature, a time, etc.) and perhaps providing additional information useful for interpretation (the filter used, the epoch of coordinates, and so on).

PDS defines OBJECTs for use in all PDS label and catalog files. These defined OBJECTs cover very broad ranges of data, yet through the use of keywords the generic objects can be made very specific. Using the pre-defined OBJECT structures, rather than formulating new ones for each data set, facilitates the creation of general tools for use across all data sets. Consequently, the use of existing OBJECTs to describe new data is strongly encouraged.

At the bottom of this page is a brief list of the most common objects used in labels. There are additional objects, but these tend to be very mission- or node-specific. A complete list of OBJECTs can be found in Appendix A of the PDS Standards Reference.

Keywords

Each OBJECT will have certain keywords which are required because they describe fundamental properties of the data. For example, some indication of record type and/or size is always required, so that file readers can read the data; time tags indicating when the data were taken are also usually required; the IMAGE object requires the user to indicate whether it is a color or grey-level image, and to indicate how the color planes are interleaved; and so on.

In addition to the required keywords, other keywords may be included to provide as much detailed description of the data as possible. Some keywords are specifically recommended for each object, others may be included as the data preparer sees fit.

Standard Values

Some keywords will have standard values associated with them. For these keywords, acceptable values are listed explicitly in the Data Dictionary. These standard value lists are applied to ensure that when a particular value appears for that keyword the meaning is well understood. It is relatively easy to add new standard values to an existing keyword, however, so data preparers should not hesitate to ask their coordinating node to be prepared to submit new standard values where appropriate.

IMAGE

This is used to describe a single raster image.

HISTOGRAM

This is generally used in conjunction with an IMAGE object to provide histogram information.

TABLE

A regular, flat, columnar data structure, possibly with heterogeneous data types for the columns. The data may be in either ASCII or binary form, provided the record length is always fixed.

SERIES

A special breed of TABLE, appropriate for a sequence of data which is completely regular along one axis (for example, data counts measured every n seconds). The SERIES object provides the start and end values and the increment for the axis value, which then does not appear explicitly in the file.

OBJECTs in Other Contexts

OBJECTs are also used to describe data structures in catalog and volume description files as well, although in these cases there is often no external data file being referenced. For example, here is a DATA_PRODUCER object which identifies the party responsible for producing a particular data set. It appears in one of the volume descriptor files included on a CD-ROM:

     OBJECT           = DATA_PRODUCER
     INSTITUTION_NAME = "GODDARD SPACE FLIGHT CENTER"
     FACILITY_NAME    = "N/A"
     FULL_NAME        = "INTERNATIONAL HALLEY WATCH"
     DISCIPLINE_NAME  = "SMALL BODIES NODE"
     ADDRESS_TEXT     = "CODE 684\n
                        GODDARD SPACE FLIGHT CENTER\n
                        GREENBELT,MD 20771 USA"
     END_OBJECT       = DATA_PRODUCER