Jorge's error-tolerant encoding "HOP"

The HOP column is an error-tolerant encoding that I used to build the consensus file. (The name "HOP" has no particular meaning.)

The encoding is derived from JSA, by condensing some stroke sequences to single letters, and omitting several details that are either too prone to error, or that are almost surely meaningless calligraphic variations. In particular, it ignores

ligatures, such as the difference between Friedman's [T] and [CC], or [S] and [2C];
the horizontal line in "intruding" gallows, such as [HZ], [PZ], [DZ], [FZ];
the left "eye" on the gallows, that distinguishes [H] from [D], and [P] from [F];
the connectivity between the `c' and `i' strokes that distinguishes [A] from [CI];
any additional `i's after three consecutive `i' strokes, in particular the difference between [AM], [AN], [AIM], [CIN], [CIIN], etc.
extension of strokes below the baseline, in particular the difference between [G] and [A].

The HOP encoding is, therefore, highly ambiguous. The purpose of this encoding is to remove the main sources of noise that could confuse the statistical analysis, even at the cost of losing some significant information.

The HOP encoding is basically the Frogguy alphabet, with the following changes:

    Frogguy:  9    a    4    x   c    e    t    e'   s    
    --------  ---- ---- ---- -------- ---- ---- ---- ---- 
    HOP:      a    a    q    e   c    c    c    z    z    
            
    Frogguy:  2    i2   iiiv iiv  iv   ig   cg  
    --------  ---- ---- ---- ---- ---- ---- ----
    HOP:      r    w    m    m    m    k    cj  
            
    Frogguy:  qp   lp   eQPt eLPt
    --------  ---- ---- ---- ----
    HOP:      H    H    cHc  cHc 
            
    Frogguy:  dj   fj   eDJt eFJt
    --------  ---- ---- ---- ----
    HOP:      P    P    cPc  cPc

Here is the approximate inverse mapping from HOP to FSG:

    HOP   FSG  
    ----  ---  
    8     8       
    H     H, D       
    P     P, F       
    a     A, G, CI      
    am    AM, AIN, CIIIL, AN, CM
    ar    AR, CIR
    aw    AIR, CIIR
    c     C        
    cHc   HZ, DZ      
    cPc   PZ, FZ      
    ca    CA, CG, CCI, TI
    cc    T, CC        
    cj    6        
    e     E       
    i     I        
    ig    7        
    iu    L        
    k     K       
    m     M     
    n     N      
    o     O        
    q     4        
    r     R       
    w     IR     
    z     2       
    za    2A, 2G, 2CI, SI
    zc    S, 2C

Note that the inverse mapping is (on purpose) ambiguous.

Last edited on 97-08-16 by stolfi