From: robfi680@gmail.com   
      
   Typical process for NaN boxing is to set the high order bits of the   
   value which causes the value to appear to be a NaN at higher precision.   
   I have been thinking about using some of the high order bits of the NaN   
   (eg bits 32 to 51) to indicate the precision of the boxed value. This   
   would allow detection of the use of a lower precision value in   
   arithmetic. Suppose a convert from single to double precision is being   
   done, but the value to be converted is only half precision. If it were   
   indicated by the NaN software might be able to fix the result. I also   
   preserve the sign bit of the number in the NaN box.   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)