When I search for “Aidin;Y^(2a)” using derinet.searchlexemes(), I get “Aidin;S” as a result, despite the existence of both variants. This means (among other issues) that the derivation Aida;Y → Aidin;Y_^(2a) present in 1.6 is impossible to remove, should someone attempt doing that.
When I search for “Aidin;Y^(2a)” using derinet.searchlexemes(), I get “Aidin;S” as a result, despite the existence of both variants. This means (among other issues) that the derivation Aida;Y → Aidin;Y_^(2a) present in 1.6 is impossible to remove, should someone attempt doing that.