my experience with k (versus j), at the turn of the millennium, was that it was very good at dealing with 'missing values' possibly due to its list (Scheme) heritage.
k shares the array heritage in the APL family, but utilizes ragged arrays very well - which j struggles with. A DBMS must somehow (they are in reality) deal with those 'missing values'. If the j community can create that capability integrally, then j can become as successful as k has become in becoming q. my experience with k (versus j), at the turn of the millennium, was that it was very good at dealing with 'missing values' possibly due to its list (Scheme) heritage. k shares the array heritage in the APL family, but utilizes ragged arrays very well - which j struggles with. A DBMS must somehow (they are in reality) deal with those 'missing values'. If the j community can create that capability integrally, then j can become as successful as k has become in becoming q. ~greg krsnadas.org ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
