On Mon, Apr 27, 2009 at 5:10 AM, Oleg Kobchenko <[email protected]> wrote: > We need to be careful about using the term "tacit" appropriately. Does it > only mean the use of "@", and S thus attempts to replace it with " "? > However tacit is much more than that: the whole repertoire of > adverbs and conjunctions, and the forks and hooks--that is tacit. > The order of complexity can hardly be denied. One has to grasp the total > entirety of tacit programming in J. It is not just about one conjunction "@", > which has its proper place and meaning and cannot be just thrown away. > How does S represent the rank difference between "@" and "@:" ? > With "@" being space, how is it going to be related with the other > facets of function composition "&", "&.", "&:" and "&.:" ?
>From his discussion so far, I think it is clear space would have @: semantics (though of course its grammar would be different). I very much agree with you that there are some deep, unresolved questions having to do with the grammar of S -- and they would have to be resolved before an implementation would be possible. Computer grammars are hard. People have been building specialized tools for dealing with grammar since the 1960s, and to this day most computer programmers do not understand how they work. Even people working with them often typically adopt stances based on an incomplete understanding. I will not say J is perfect, but I will say that Ken Iverson had a deeper understanding of computer syntax and grammar issues than a lot of people do. -- Raul ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
