Roger wrote: > a. I disagree that the functions in questions should be niladic.
I agree. Though there are other ways to interpret the phrase "should-be-niladic" (which don't involve admitting zero-argument verbs into the language). > it is better to be strict now than to be > sorry about being lax later. Fair enough. Can you sketch an example (or two) where permitting an arbitrary argument to such a function would cause us to be sorry later? And, out of historical interest, what prompted the change when it occured? -Dan ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
