I haven't tried that out @Anuhya-Y, but I would assume it would support. The video shows permissions being managed for query service as well.

You can go with that @Luca_Lattarini. Just set them as Identity if you intend to use them to stitch with another dataset.

https://experienceleague.adobe.com/docs/platform-learn/tutorials/admin/configure-attribute-based-access-control.html -> this is same as what I posted out. So finally we have this feature available now.

@Luca_Lattarini Non-people IDs are used for storing identifiers that require namespaces but are not connected to a person cluster. For example, a product SKU, data related to products, organizations, or stores.

@Luca_Lattarini This is tricky and will depend from case to case and is heavily dependent on the data that is flowing in. Always do data profiling before you ingest and gain better confidence.  

@Luca_Lattarini you cannot "enable profile" on the dataset if the schema is not profile enabled. 

@Luca_Lattarini  accountId should be cross-device id - do not set that as non-people id. Use non-people id if you need to use it for lookup table like Product (productId).   As far as I know, non-people id will not be used for stitching.   Regards, Chetanya

Also, there are other data element types other than Path - not sure if you have tried that.

@vikash4 I am wondering how will you populate the data element if it's not part of the data layer that gets ingested from websdk/mobilesdk/server API? To ingest the data into data collection, you will need the xdm defined and then you will forward it to non-adobe. Please let me know if I am missing ...

@Luca_Lattarini,   I would look at the data, its quality, and which source is sending it to decide the right merge policy to pick. The intention should be to get the right/best data to build the 360-view of profile for the activation. I would not look for activation/segmentation to decide which merg...