EQV? note

[Jonathan A Rees <JAR@MC.LCS.MIT.EDU>]

Let me repeat for about the third time my request that EQV? have
machine-independent semantics.  I believe that this is a logical
consequence of the intended "spirit" of EQV? which (I thought) is that
it be a *clean* fine-grained equality predicate.  I feel a little
insulted that people haven't even taken this suggestion seriously enough
to argue against it.  Judging form the number of "I" answers to the
survey's EQV?  questions, most people seem to disagree with this,
although they won't say why.  I'll assume people have just forgotten
that I suggested this, but I would like some reaction.