Dublin Core Meta-data Initiative vocabularies.

module.

vernacular crop names... The CropDescriptorLoader now seems to be
completely functional(?).

crops, thus allowing complete automation (driven by RDF) of descriptor
loading.  src/test/resources now has updated SPARQL query and sample RDF
data files.

descriptions (following the DarwinCore cue)

to a subtle JSON artifact which introduced escaped quotes ('\"') into
language maps when language sub-maps were converted to strings
(toString), added to an Json object, then the total Json object
converted (again) to a string (toString a second time...).

Resolved the issue by only converting to strings (with toString) once...

Crop updates now work (again) with non-English languages throughout, in
a manner that the web interface now updates crop-specific labels.

One last enhancement being contemplated is to perhaps introduce RDF
triplet statements to encode Crop taxonomy information including foreign
language names.

Note that perhaps there needs to be a robust way of updating the
database for additional non-English language strings after the initial
loading of the database with the original (meta-)data.

Property Categories ("ParameterCategory") from
http://data.bioversityinternational.org.

trait vocabulary, to allow for bulk uploading and initialization of
databases.

Core model split off and stored in a
src/test/resources/CoreTraitOntology.n3 model file, with Musa sample
trait meta-data alongside.

CropDescriptorQuery changed to suit, but changes (to ParameterCategory
handling) not yet fully propagated to CropDescriptorLoader.

and updated Crop Descriptor Loader

now without crashing, and creates suitable database records, but on the
GENESYS side, foreign characters are not yet properly displayed on the
web. Not yet sure where the loss of information occurs...

Still troubleshooting CropDescriptorLoader (and GENESYS server side
code...)

Added I18N support.

uploading.  Made 'scale' subordinate to 'method' in the RDF model
(updated SPARQL query accordingly)

specification of crop, property category, property and method meta-data
loading into GENESYS (where property/method is loaded from the 
data.bioversityinternational.org RDF triple store) appears to be 
essentially functional using the current GENESYS data model.

NOT included are proposed new functionality for GENESYS management of
such meta-data, such as recording of the RDF URI, and the handling of
other languages.

Trait class and into the CropDescriptorLoader (accessing the REST).

interface... Doesn't yet work: need to do something with the concept of
Parameter "category"?

(i.e. indefinite or infinity)

management...

Value attribute, reflecting the assumption that a range of Values for a
given option will generally have the same unit type.

Implemented more of the CropDescriptorLoader JSON object loading (not
yet complete, though - missing 'range'... not sure what to put in
there...)

is inaccessible...

computational constraints in systems using the code, e.g. GENESYS)

traitCatalog Map a non-static data structure, so that it is associated
with a given CropDescriptorQuery taxon & language.

inner superclass, and refactored common elements of the API.  Also added
the Concept URI as global identification. Dumping code simplified.

CropDescriptorQuery seems to be working well now and has a pretty good
clean level of internal refactored design for internal Trait++ model.

data transfer object (of sorts) with embedded inner classes for Methods
and Scales.  First iteration seems to work ok.

from data.bioversityinternational.org