truth of '()

[jinx@chamartin.ai.mit.edu (Guillermo J. Rozas)]

    I don't recall that we have any consensus for eventually declaring that
    implementations -must- have distinct () and #F objects.  My understanding
    is that making the boolean significance of () unspecified is to accommodate
    implementations in which they are distinct, where it might be inconvenient
    to arrange for () to be false.  

I agree completely with Alan.  I might encourage implementors and
dialect designers to distinguish () and #f.  I might even distinguish
them myself if I were designing a new dialect, but I would never agree
to forcing them to be distinct.