Why not simply the AFP?

Tobias

On 18/06/2018 12:36, Lawrence Paulson wrote:
I have just formalised most of the HOL Light development of quaternions. It's a 
great advertisement for type classes: showing that quaternions constitute a 
real normed division algebra and an inner product space supersedes most of the 
HOL Light proofs (many files), which are devoted to developing the arithmetic 
and topological properties of quaternions. In the course of this, I also ported 
the HOL Light development of the three-dimensional vector cross product.

So where does this material belong? Arguably not in Analysis, which is already 
too large.

Another question: I did not port the quaternion material relating to the 
constant reflect_along, which is defined as follows:

(* Reflection of a vector about 0 along a line. *)

let reflect_along = new_definition
  `reflect_along v (x:real^N) = x - (&2 * (x dot v) / (v dot v)) % v`;;

Is this concept useful, and if so, where does it belong?

Larry

_______________________________________________
isabelle-dev mailing list
isabelle-...@in.tum.de
https://mailmanbroy.informatik.tu-muenchen.de/mailman/listinfo/isabelle-dev


Attachment: smime.p7s
Description: S/MIME Cryptographic Signature

_______________________________________________
isabelle-dev mailing list
isabelle-...@in.tum.de
https://mailmanbroy.informatik.tu-muenchen.de/mailman/listinfo/isabelle-dev

Reply via email to