From: krishna.myneni@ccreweb.org   
      
   For typical implementations of 2! and 2@, the storage of double length   
   numbers in memory do not correspond to the native byte ordering for a   
   double length number. For these implementations, the high order bits of   
   the double number are on the top cell of the stack, which is at a lower   
   memory address. For example,   
      
   2variable x   
      
   5 s>d .s   
      
   		0   
   		5   
     ok   
      
   x 2!   
      
   x 8 dump  \ on a 32-bit little-endian system   
      
   A0AE7C0  :  00  00  00  00  05  00  00  00   ok   
      
   In the Forth 2012 standard, and in prior standards, the representation   
   of double numbers on the stack seems to be allowed to be implementation   
   defined. Thus, on a 32-bit system, the stack order of a 64 bit double   
   number, the standard does not specify whether the high 32 bits of the   
   number are on top of the stack or the low 32 bits. Hence, we have the   
   rationale (A.8.6.1.1140) for the word D>S to abstract the conversion of   
   a double to single length number (instead of using DROP).   
      
   To overcome this issue, we could simply decide that double length   
   numbers on the stack have a specific ordering corresponding to their   
   native memory ordering on the system, but this appears to be contrary to   
   some existing implementations, and it may also break code.   
      
   A better way to do this is to explicitly define D! and D@, so that the   
   native memory storage order is consistent for this type.   
      
   For little-endian Forth systems which place the high order cell of the   
   double number on top of the stack, the definitions would be   
      
   : D! ( d a -- ) >r swap r> 2! ;   
   : D@ ( a -- d ) 2@ swap ;   
      
   === Example ===   
   5 s>d x d!   
     ok   
   x 8 dump   
      
   A0AE7C0  :  05  00  00  00  00  00  00  00   ok   
      
   x d@ .s   
   		0   
   		5   
     ok   
   === End Example ===   
      
   More generally, the problem with the double number word set is that it   
   attempts to use the same word set for two different types:   
      
   1) pairs of cell length numbers   
   2) double length integers   
      
   It would be better to separate the words for these two types, the latter   
   having prefix of "D" and the former having prefix "2".   
      
   --   
   Krishna Myneni   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)