Conjunctions are named in "boldface erect" on the Vocabulary page. (Adverbs, boldface italic). Have a look
http://www.jsoftware.com/docs/help701/dictionary/vocabul.htm There are about 25 conjunctions, and all except Rank " are excluded from Simple J. You have to do without Tie ` --useful for continued fractions-- and Power ^: which provides While ... Do ... . Often you really need Explicit : and who wants to do without Under &. ? But Simple J is a good place to begin: use trains and rank but no other conjunction. On 11/14/2011 8:54 PM, Linda Alvord wrote: > Here's how to find conjunctions: > > Name Class. Class of (boxed) name: > > _2 invalid > _1 unused > 0 noun > 1 adverb > 2 conjunction > 3 verb > > a=:2 3 5 > by=:' '&;@,.@[ ,. ] > over=:({. ; }.)@":@, > > 4!:0<'a' > 0 > > 4!:0<'over' > 3 > > 4!:0<'by' > 3 > > f=: @ > 4!:0<'f' > 2 > > g=:": > 4!:0<'g' > 3 > > h=:& > 4!:0<'h' > 2 > > Eliminate the conjunctions @ and& to write simple J . > Also, the functions by and over are not simple J verbs. > > > -----Or > original Message----- > From: [email protected] > [mailto:[email protected]] On Behalf Of Kip Murray > Sent: Monday, November 14, 2011 3:58 PM > To: Programming forum > Subject: Re: [Jprogramming] table challenge in simple J > > My definition of Simple J is more restricted and potentially makes the > challenge greater: Simple J uses trains and rank but no other conjunction. > > On 11/14/2011 12:59 PM, Dan Bron wrote: >> How about a more difficult challenge? It's related. No embargo period on > this one. >> >> Write a 13 : '' for "Simple J". In particular, write an adverb which, > given an anonymous tacit verb as an argument, derives a functionally > identical verb (in terms of I/O), but all instances of f@:g are replaced > with [: f g . Obviously f and g here are arbitrary (could be named or > anonymous, primitive or compound, vanilla or parenthesized), and there can > be more than one use of @: in the target, and you could have uses within > uses. >> >> It would be quite interesting to me if this were presented as a tree > transformation using L: S: {:: and friends against the atomic representation > of the target. That's not a requirement though - you could always do string > transformations on the linear rep, rearrangements on the boxed rep, etc. >> >> If this was in any way generalized, in sense that I could reuse it for > different but similar transformations, that would be more interesting still. >> >> -Dan >> >> -----Original Message----- >> From: [email protected] > [mailto:[email protected]] On Behalf Of Raul Miller >> Sent: Monday, November 14, 2011 1:25 PM >> To: Programming forum >> Subject: Re: [Jprogramming] table challenge in simple J >> >> You are correct. >> >> Nevertheless, that kind of restriction probably belongs in the subject > line. >> > ---------------------------------------------------------------------- > For information about J forums see http://www.jsoftware.com/forums.htm > > ---------------------------------------------------------------------- > For information about J forums see http://www.jsoftware.com/forums.htm ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
