Andrew,
This is a very nice description of the relationship between DIF and ISO. I was a bit surprised not to see any mention of 19115-2, particularly for the Sensor and Platform crosswalk. Seems straightforward to map these to MI_Metadata.acquisitionInformation > MI_AcquisitionInformation > MI_Instrument.identifier > MD_Identifier.code and MI_Metadata.acquisitionInformation > MI_AcquisitionInformation > MI_Platform.identifier > MD_Identifier.code. The authorities in both cases would be the appropriate GCMD Keyword list (sensor or platform). I believe the GCMD Sensor Keywords also have a hierarchy at least two levels deep. The higher level of that hierarchy could provide a value for the MI_Metadata.acquisitionInformation > MI_AcquisitionInformation > MI_Instrument.type. A positive side effect of this approach is that citations can be provided for the instruments after the initial crosswalk from DIF. In the FGDC Remote Sensing case, these citations are already in the standard and can be cross-walked directly. I would do this in addition to the aggregation groups (btw, are those crossReferences or largerWorkCitations?). A potential downside/complication (?) would be the transition from MD_Metadata to MI_Metadata, but we are likely going to do that for all of our metadata anyway because the inclusion of the platform/instrument information is close to ubiquitous (or should be).
Dealing with the GCMD hierarchy is always interesting and important. Breaking it up into separate thesauri could be useful in some cases. I would consider also providing the complete keyword as well with a reference to the overall keyword list as the Thesaurus.
I am also interested in the problem of translating from ISO back to DIF as a mechanism for supporting GCMD as a discovery tool on top of an ISO compliant repository. We would also crosswalk back to FGDC to support Geospatial One Stop and possibly to Dublin Core. These crosswalks would be “discovery-only” tools (rather simple) and would include a link back to the repository for a more complete collection of metadata objects. This is the general strategy that we have been exploring. We will be suggesting extending the CI_OnlineResource.function code list to include “completeMetadata” as part of this approach.
Ted
==== Ted Habermann ===========================
Enterprise Data Systems Group Leader
NOAA, National Geophysical Data Center
V: 303.497.6472 F: 303.497.6513
"If you want to go quickly, go alone.
If you want to go far, go together"
Old Proverb
==== Ted.Habermann@xxxxxxxx ==================
|
BEGIN:VCARD
VERSION:3.0
N:Habermann;Ted;;;
FN:Ted Habermann
ORG:NOAA/NGDC Enterprise Data Systems Group;
EMAIL;type=INTERNET;type=HOME;type=pref:ted.habermann@xxxxxxxx
TEL;type=WORK:3034976472
TEL;type=CELL;type=pref:303-819-1795
item1.ADR;type=WORK;type=pref:;;325 Broadway;Boulder;Colorado;80305;USA
item1.X-ABADR:us
X-ABUID:0BF99EC4-4EBC-4571-ACF3-50665EE59B87\:ABPerson
END:VCARD
