From: nowhere@never.at   
      
   James Harris asked:   
      
   ...   
   >> Of course we have to write a working decoder before and then encode it   
   >> either by an 2:1 algo, that's what Terje and Kerr use, or what I   
   >> use is encode only the bytes out of range to be modified by SMC.   
   >> So my shortest decoder was 54 byte, but made it 59 for lesser SMC needs.   
      
   > Do you mean the smallest base64 decoders in the above were about 54   
   > bytes long, stored in textual form about 108 bytes long, and being   
   > decoded 2 to 1?   
      
   no, my SMC variant was 54 byte code, stored as 54 byte text.   
   This needs to be 'partial' modified by SUB and XOR.   
      
   >> Kerr wrote the yet shortest with only 44 byte and I adopted it for SMC   
   >> it came down to total 180 (+29 for base64 encoded Hello world test code).   
      
   > You mentioned before that you had got down to 209 but it sounds from the   
   > above as though you had reduced it to 180 - with the other 29 being test   
   > data. I would call that a 180-byte solution!   
      
   Yes, but because we use the same test string we count total size.   
   __   
   wolfgang   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)