Scientific data formats such as NetCDF have made great strides
in areas like interoperability, scalability, and data compression.
By comparison, methods for representing the uncertainty inherent
in the values stored in scientific data sets are less robust. A group
organized by researchers from the National Research Council of Italy's
Institute for Atmospheric Pollution Research is trying to address this
issue by creating a set of conventions for the representation of
uncertainty values associated with data stored in NetCDF files.
A discussion paper describing ideas for storing uncertainty values
is available, and an electronic mail list has been created to allow
broad community discussion of the proposal.