Should we add another field for types? It is important that readers of the dictionary are not left with questions as to what may fill any given sumti place. I'm thinking something like the following, in the case of, say, spaji:
types:
x1 : event
x2 : observer
(This is just an example of what the types could be.)
What we need to do is assemble an exhaustive list of "types" (if we want to call it that, which we might not, I don't know) and agree on names for them. Then we can list the "types" for each gismu place in their respective files.
Should we add another field for types? It is important that readers of the dictionary are not left with questions as to what may fill any given sumti place. I'm thinking something like the following, in the case of, say, spaji:
(This is just an example of what the types could be.)
What we need to do is assemble an exhaustive list of "types" (if we want to call it that, which we might not, I don't know) and agree on names for them. Then we can list the "types" for each gismu place in their respective files.