NOTICE: This version of the NSF Unidata web site (archive.unidata.ucar.edu) is no longer being updated.
Current content can be found at unidata.ucar.edu.
To learn about what's going on, see About the Archive Site.
NOTICE: This version of the NSF Unidata web site (archive.unidata.ucar.edu) is no longer being updated.
Current content can be found at unidata.ucar.edu.
To learn about what's going on, see About the Archive Site.
NOTE: The galeon mailing list is no longer active. The list archives are made available for historical reasons.
Andrea & all-in an attempt to address currently open items let me come back to the issue raised by Andrea Aime / Steven Keens:
Q: should we allow, in WCS 1.2, that null values are identified not only by a value or a list of values, but additionally by intervals?
Examples:
myIntNullValues = { -65534, [-1 ... +1] }
myFloatNullValues = { 0, [2.71 ... 3.14], [0.999 ... 1.001] }
Background:
- Currently (WCS 1.1.2) allows zero or more null values in
wcsDescribeCoverage.xsd:
0..* ows:CodeType - in WCS 1.0.0 null values were described by (values.xsd):0|1 valueEnumType ("Enumeration of TBD values which consists of a sequence of intervals and/or single values.") I think I remember we gave up at some time when there was no light at the end of the complexity tunnel, such as in dealing with open/closed boundaries (which are present, but not defined 1.0.0).
Arguments:pro: was possible in WCS 1.0.0, we shouldn' reduce functionality nor backwards compatbility (and 1.0.0 is important for a significant user group) pro: especially when dealing with floating point numbers it is unstable to use specific values and use "==". Intervals allow for a numerically appropriate handling. con: adds complexity to a server implementation, and potentially to a client which attempts to seriously deal with null values. (Of course it is up to a service to make use of null intervals.)
con: some semantic details seemingly waiting for clarification. Q to WCS implementors (in particular: GALEON): 1) do you see a need for intervals in your community? 2) would you see null intervals as a severe implementation burden? thanks, Peter -- Dr. Peter Baumann - Professor of Computer Science, Jacobs University Bremen www.faculty.iu-bremen.de/pbaumann, mail: p.baumann@xxxxxxxxxxxxxxxxxxxx tel: +49-421-200-3178, fax: +49-421-200-493178 - Executive Director, rasdaman GmbH Bremen (HRB 147737) www.rasdaman.com, mail: baumann@xxxxxxxxxxxx tel: 0800-rasdaman, fax: 0800-rasdafax, mobile: +49-173-5837882 "A brilliant idea is a job halfdone."
galeon archives: