As mentioned in #49, the rules for what constitutes a "known-expression" in the spec haven't been formally specified.
On the one hand, it's kind of obvious (literals are known, constant bindings are known, expressions whose subexpressions are all known are themselves known), but it wouldn't hurt to spell it out explicitly in case there are some non-obvious corner cases.
In particular, some expressions which look like function calls are known, and some are not, while function identifiers themselves should probably be considered known.
As mentioned in #49, the rules for what constitutes a "known-expression" in the spec haven't been formally specified.
On the one hand, it's kind of obvious (literals are known,
constantbindings are known, expressions whose subexpressions are all known are themselves known), but it wouldn't hurt to spell it out explicitly in case there are some non-obvious corner cases.In particular, some expressions which look like function calls are known, and some are not, while function identifiers themselves should probably be considered known.