Investigating iconic combinators

Various properties of two bracket abstraction algorithms are discussed and some of those properties are proved in "Investigations into Iconic Representations of Combinators". Algorithm (L) is uni-variate and uses yes-no representations, whereas algorithm (M) is multi-variate and uses array representations. It is shown that

  1. both algorithms are structure-preserving;
  2. there is a straightforward connection between [x]P and [x]Q, produced by algorithm (L), if the primal components of Q, except the first, are a permutation of the primal components of P, except the first;
  3. there is a simple connection between the two abstracts produced when algorithm (L) is used repeatedly on the same input term, but the abstractions are performed in a different order in each case;
  4. there is a straightforward connection between the array representations produced by algorithm (M) and the yes-no representations produced when algorithm (L) is used to abstract the same variables individually from the same input term and
  5. using algorithm (M) multi-variate abstraction can be "partitioned".

Since writing this paper about 15 years ago, I have continued to study bracket abstraction algorithms.

Reference

  • Antoni Diller, "Investigations into Iconic Representations of Combinators", in Javier Blanco (editor), Argentine Workshop on Theoretical Computer Science (WAIT2000) Proceedings: Tandil, September 4-9, 2000, [Buenos Aires, Sociedad Argentina de Informática e Investigación Operativa (SADIO), 2000], pp. 52–62; a PDF version of this paper is available on this website.

© Antoni Diller (27 March 2014)