Issue Summary

When Importing using the Schema Import Type - the datalocator for an "Object" Attribute will be represented by the Target Object Type Name. If I have several Attributes on the Object Type I am importing, all pointing to the same Object Type - once imported, all will be populated by the SAME Object, because there is only one Datalocator available.

Steps to Reproduce

1. Configure a Schema "Schema A" with two Object Types: OTA and OTB
2. Create a few objects in OTA, with Names like: O1, O2, O3
3. On OTB, navigate to the Attributes Tab and add 3 Object Attributes, all referencing OTA:
   Ref1
   Ref2
   Ref3
4. Create an Object in OTB with different references OTA objects:
   Ref1 : O1
   Ref2 : O2
   Ref3 : O3
5. In another Schema / instance, configure an Insight Object schema import Type to import from "Schema A". you may use a predefined structure / configuration or set the structure and mapping manually.
6. Note the OTB Object Type Mapping for the referenced Attributes - they are all using the exact same Datalocator: OTA ... and NOT the source Attribute's name...
7. Synchronize the import 

Expected Results

The Imported OTB object will have the correct referenced OTA Objects (imported):
Ref1 : O1
Ref2 : O2
Ref3 : O3

Actual Results

Ref1 : O1
Ref2 : O1
Ref3 : O1

...

Workaround

Use the Insight>Schema Export to export your Schema into a Zip file, then use Insight>Schema Import to import your schema in the target instance - this method will maintain the references as expected.
Another option:
Use the Object Schema Import type to generate the initial Structure (Object types and Attributes), then, add / modify Attributes as required, if these were not created / were not created similarly.

Export a source Object Type to CSV ("User Friendly") and import it into the Target's corresponding Object Type. Map the Reference attributes with Label = ${<datalocator>} so the target objects to reference will be identified by their Label and not by their Key.