Re: Re: Re: VMs: Cardan grille

["jan" <hurychj@xxxxxxxxx>]

Hello Jacques, 
  
  
=======  You wrote:  

>Then:
>
>(1) you do not need a grille. What you do is take, say,
>    20 numbers, like this: 7, 3, 2, 4, 5...
>
>    Which give you the number of nulls you write: 7 nulls,
>    one true letter, 3 nulls, one true letter, 2 nulls, and
>    so on. Look: no grille!

Exactly, my point also. I am not claiming the grille was really used, but the
similar algorithm could have been. 
>
>(2) you cannot use a grille to decrypt your text, because,
>    as you said, you wrote a draft first (using a grille)
>    then copied it with the lines sagging right. Result:
>    apply your original grille onto that and what do you
>    get? Gibberish, because the positions of the "true" 
>    letters no longer match the positions of the holes 
>    of the grille.

Well, you have to rewrite it again, keeping the spaces and you have to  know the  total number
of characters per  line (counting the  spaces or  not,  it depends). As for as  the line length,
we do  not know it ( it may have even vary line by  line, and  not necessarily as it is in the VM)
So one cannot use the grille on rewrite either. Another reason the grille is not needed.

While coding  using the grille  would be  cumbersome, such algorithm can still be used,
providing it is written the way you have shown above, and the  total number of letters
 per line is specified (including  or not including the spaces - they do not have their true function any more,
but it will make the difference in solving it). The described matrix (not a grill anymore :-) 
can be then filled in, and nulls will be simply crossed out. Mind you, the solving will be a pain in neck, but
what if it was meant to be?

Jan 

______________________________________________________________________
To unsubscribe, send mail to majordomo@xxxxxxxxxxx with a body saying:
unsubscribe vms-list