Open pietercolpaert opened 3 years ago
The usecase for this would be that certain tree:values
can be bigger than the actual data itself. If you fragment a dataset by administrative boundaries for example, you might want to give each boundary its own dereferenceable identifier instead to save on bandwidth.
Those values can then be cached for as long as the cache headers say they can be cached, that's not really a TREE concern.
@hdelva proposed using qualified properties for easier processing of really complex tree:values
https://biblio.ugent.be/publication/8664563
Not sure whether we should add this to the specification as it brings challenges regarding how long a processing of a literal can be cached for example.