The idea is that naming makes the thing a meaningful whole, not a fragment of text. Many of my early misunderstandings came from imagining that names would execute as though the corresponding text had occurred there, instead.
On 1/23/09, Anssi Seppälä <[email protected]> wrote: > Thank you, > In other words assignment puts extra ( ) around the sentence. > Anssi > > > At 00:09 24.1.2009, Henry Rich wrote: >>You meant >> >>inf2num =: _2e9&>. @: (2e9&<.) >> >>Henry Rich >> >> >>Anssi Seppälä wrote: >> > Thanks to the new handling of _ and __ in J 6.02 NaN error occurs in >> > code that worked fine earlier versions of J. Therefore I need a >> > operation that changes _ and __ to a large positive and negative >> > value to avoid NaN error and leave all small numbers as they are: >> > >> > In J session I developed: >> > _2e9&>. ] 2e9&<. _ 0 __ >> > 2e9 0 _2e9 >> > >> > So this works fine and should be assigned to inf2num >> > >> > inf2num =: _2e9&>. ] 2e9&<. >> > >> > However: >> > inf2num _ 0 __ >> > 2e9 0 __ >> > >> > Can someone please explain why inf2num does behave similar and not >> > handle __ as the equal command does in session ? >> > >> > Anssi >> > >> > ---------------------------------------------------------------------- >> > 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 > ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
