[Prev][Next][Index]
Re: Larch "oversight"
-
To: "Stephen J. Garland" <garland@larch.lcs.mit.edu>
-
Subject: Re: Larch "oversight"
-
From: horning
-
Date: Wed, 06 Dec 95 11:16:30 -0800
-
Cc: davidg@oracorp.com, guttag@larch.lcs.mit.edu, horning
-
Delivery-Date: Wed, 06 Dec 95 11:16:36 -0800
-
In-Reply-To: Message of Wed, 06 Dec 95 11:15:42 -0500 from "Stephen J. Garland" <garland@larch.lcs.mit.edu>
Steve,
Thanks for the analysis.
I think this is a case of one oversight "saving" us from a different
oversight. I cannot recall that we ever really thought about either of
these cases. They are artifacts of the way LSL evolved from a purely
algebraic language. Recall that originally the comparison operator `='
had to be declared like any other operator. Thus there wasn't anything
interesting to say about a sort that wasn't in the signature of any of the
declared operators.
If I were doing it over, I would probably require sort declaration, as in
LP.
Jim H.