Skip to Content.
Sympa Menu

coq-club - A question on positivity conditions

coq-club AT inria.fr

Subject: The Coq mailing list

List archive

A question on positivity conditions


chronological Thread 
  • From: Claudio Sacerdoti Coen <sacerdot AT students.cs.unibo.it>
  • To: coq-club AT pauillac.inria.fr
  • Cc: lpadovan AT cs.unibo.it, schena AT cs.unibo.it, asperti AT cs.unibo.it
  • Subject: A question on positivity conditions
  • Date: Tue, 18 Jan 2000 16:57:09 +0100

 Hi,

 I have read the description of the positivity conditions required
 for an indutive definition, but I can't understand why in Coq the next
 definition is valid:

  Inductive t : Type := k : (List t) -> t.

 where List is defined as usual as

  Inductive List [T:Type] : Type :=
     nil : (List T)
   | cons : T -> (List T) -> (List T).

 Accordingly to the manual, t should appear only strictly positively in
 (List t), but it doesn't.

 Is Coq's implementation different from the theory or have I missed something?

                                             Thanks in advance,
                                                 C.S.C.

-- 
-----------------------------------------
Real Name: Claudio Sacerdoti Coen
Address: via del Colle n.6
         S. Lazzaro di Savena (BO)
         Italy
e-mail:  
sacerdot AT cs.unibo.it
-----------------------------------------





Archive powered by MhonArc 2.6.16.

Top of Page