What is the architectural use case for passing function names as strings? You could just pass the named function around imo and not lose IDE Code completion and all that.
Right now you're correct, we could just pass the functions. In the future, however, we'd like to extend the language so that you can call `remove` with a regex to make it easier to remove "groups" of functions (so that, every time you add a new function of a particular "group", you don't have go back to every spot where you remove things of that group and add another line).
There are a few other things we need to fix before we get there though, chief among those being memory use