I'm still digesting the your suggestions but have three initial comments. a) You often refer to "explicit definitions" that aren't "explicit" in J terminology. My take is that you are referring to a definition as being explicit as soon as it is assigned to a name, whereas in J terminology a definition is explicit where the arguments are referred to explicitly.
For example you refer to this as explicit definition: Mean =. <. 0.5 + (+/ % #) This is actually a tacit definition that doesn't do what you want it to. What I think you wanted was: Mean =. [: <. 0.5 + (+/ % #) NB. The brackets are not necessary An equivalent explicit definition would be: Mean=. 3 : '<. 0.5 + (+/y) % #y' b) I think it would be sensible to separate the ideas for introducing new display symbols for J primitives from ideas about changing how to parse them. c) I'd be worried that using symbols that aren't currently included in the Unicode alphabet may create similar problems to those that APL suffered from (and still does to a lesser extent) before the more widespread acceptance of Unicode. > -----Original Message----- > From: [email protected] [mailto:general- > [email protected]] On Behalf Of Don Watson > Sent: Thursday, 12 March 2009 16:16 > To: General forum > Subject: [Jgeneral] J with APL > > BlankRecently I suggested in this forum that it would not be difficult > to > blend the strengths of APL in with the strengths of J by means of some > straightforward additions to J. It was suggested to me very reasonably > that > I give some examples. What I have done is to write a paper on the > subject > that includes the necessary examples. > > The reason the addition would be simple is that no change is made to > input > or storage; the only change is to the output system, thus not so much > changing the system as changing the user's perception of the system. > > In addition, one additional adverb is suggested to enable explicit > definitions to closer parallel mathematical formulae. > > It is my intention to post a finished product on the Wiki, but before I > do > that I need constructive and/or destructive criticism - and to find out > how > to post it! I would be very grateful if some members of this forum > would be > kind enough to read my paper and comment to me. At the moment the paper > is > at: > > > http://bcompanion.com/Compromisepdf.pdf > > Thanks, > > Don Watson > > ---------------------------------------------------------------------- > For information about J forums see http://www.jsoftware.com/forums.htm ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
