Sets and lists

[marqh]

RDF statements are distinct and unordered. RDF TTL supports a shortened syntax, where a subject () and predicate () may declare a number of objects as a set, separated by a ,.

<subj> <pred> <obj0>, <obj1>, <obj2> .

I propose the same interpretation is used for attribute: value entries in BALD files

Where an explicit ordered list is required, the ttl syntax is again simple and useful, explicitly a space separated, parenthesis bound collection

<subj> <pred> (<obj0> <obj1> <obj2>) .

[jyucsiro]

@nicholascar mentions Existing convention in json-ld for differentiating lists. Could we use this for netcdf-ld in values.
Unordered: key = “[ ‘a’, ‘b’, ‘c’]”
Ordered: key = “@list: [ ‘a’, ‘b’, ‘c’]”