SEACOOS CDL v2.0 December 9, 2004 7 4. Publishing and Sharing Data SEACOOS is dedicated to implement, test, and evaluate the OPeNDAP software solution as a form of data sharing. Each SEACOOS partner makes their data accessible to the Internet through an OPeNDAP-server (Seim, et al., 2002). This provides access to distributed data in two ways: simple file download or client-access.  Both methods have their advantages and disadvantages. Another important aspect of data sharing is cataloging the OPeNDAP-servers and URL locations with SEACOOS and other metadata clearing- houses. Simple file download is available because the OPeNDAP-server resides on an existing web- server. Basic web-service allows file download capability.  A user can download the published netCDF file but must have the appropriate utility and libraries to open and manipulate the stored data. The client-access method is achieved through the use of OPeNDAP. OPeNDAP is a client-server based distributed system for access to oceanographic data over the Internet that supports many underlying file formats one of which is netCDF. OPeNDAP supports use of the netCDF interface for clients. The server supports use of URL notation for accessing netCDF data from remote sites running the OPeNDAP server. OPeNDAP also allows subsetting, searching and seamless data access into higher level tools such as MATLAB. The server, client and library source-code and binarie s are available from http://www.opendap.org. By cataloging the OPeNDAP-server or URL with  regional and national catalogs, it is ensured that the data are found and utilized. By listing an OPeNDAP URL with SEACOOS, the data can be found by the SEACOOS data scout and uploaded on a regular basis to the SEACOOS data display.  Contact SEACOOS by email (data@seacoos.org) to add a new OPeNDAP URL. Furthermore, by filing metadata records through catalog facilities such as the Global Change Master Directory (GCMD) or Federal Geographic Data Committee (FGDC) compliant tools, such as Meta-door (http://carocoops.org/metadoor), ensures that the data are found and utilized through FGDC search tools.  The more ways to search and locate distributed data the better. 5. Data Format Categories In order to represent many different data sources, a canonical set of data models or format categories are used to represent in situ and remote ocean observations. These data may come from in situ instruments located on  many different types of platforms such as buoys, moorings, off-shore towers, tide- and stream- gauging stations, ships, towed-bodies, weather balloons, gliders, drifters, and airplanes.  The data may also come from many remote sensing  instruments (active or passive, optic, acoustic or microwave) located on different platforms like satellites, aircraft, ships, radar stations, and moored frames.   The format categories are based on how each of these platforms move (fixed or moving) and how the data are collected (point or profiler or map). By viewing each measurement or field