"Tacit definition", "tacit programming", etc. originated with Ken Iverson, and first appeared in the 1991 paper *Tacit Definition* <http://www.jsoftware.com/papers/TacitDefn.htm>. Before that, phrases like +/, +@*/ were called "operator expressions" and were commonly used at IPSA in the 1980s. The final piece that made tacit definition work was forks. This is described in more detail in the 1991 paper.
A particular form of forks is (f0 , f1 , ... , fn) or (f0 ; f1 ; ... ; fn), where the odd-positioned verbs is , or ; or ,. or ... See *J Introduction and Dictionary*, Sample topic 15 <http://www.jsoftware.com/help/dictionary/samp15.htm>. On Sun, Dec 8, 2019 at 8:24 AM 'Jim Russell' via Programming < [email protected]> wrote: > Never having seen a name for Ken and Roger's official dictionary's > expression format: > X (exp1;...;expn) y > I am going to use the subject phrase, at least until someone corrects me. > > Thanks to Jimmy Gauvin (and Brian), I now understand that this cryptic > format used throughout the dictionary is a (series) of tridents with a > middle ";" verb; hence each verb fork seeing the x and y arguments as a > result of the "normal" (as yet to be fully understood by me) evaluation of > verb trains. (Close enough?) > > This was a real epiphany to me. (Ok, I'm slow!) > Now I wonder which came first? > > My first recall of 'tacit expressions' was while at STSC; it was mentioned > as something the folks at IPSharp were doing. (Don't know if it was > implemented or not in their commercial APL System.) > > Haven't seen it addressed in any of Roger's excellent history of J > articles; is any one willing to describe, or do a (perhaps Reflexive-ish) > article re the history of "expositional tacit? > > > ---------------------------------------------------------------------- > For information about J forums see http://www.jsoftware.com/forums.htm > ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
