This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.
another answer from Russ Rew: > I can't help but wonder why arrays of numbers are allowed, but not strings -- it > seems that strings are > the more obvious usage for attribute values. A property of netCDF data is that a file created by the interface in one language, for example Java, should be readable by the API of any other language, for example Fortran-77. The latter language doesn't actually support variable-length strings, so it's difficult to write an API for Fortran-77 that could deal with attributes that were arrays of variable-length strings. Many of the netCDF's limitations derive from the requirement that Fortran-77 programs must be able to read all netCDF datasets. --Russ