from Paul Erlich:
I think I once posted to the effect that this measure is simply the shortest
distance along the edges of the triangular lattice I have proposed.

But it appears I was wrong about that. For example, your expressibility
measure for 81:80 would be 4*log(3) while for 135:128 it would be 3*log(3) +
log(5). However, in the triangular lattice both intervals have length
3*log(3) + log(5).


Rectangular:
                                                    135/128
                                                       |
                                                       |
                                                       |
                                                       |
                     1/1-------3/2--------9/8--------27/16-------81/64
                                                                   |
                                                                   |
                                                                   |
                                                                   |
                                                                 81/80

Triangular:
                                      

                                                     135/128
                                                       /
                                                      /
                                                     /
                                                    /
                 1/1-------3/2--------9/8--------27/16
                                                    \
                                                     \
                                                      \
                                                       \
                                                      81/80


So it appears I was wrong about the triangular lattice. Yet the triangular
lattice distance measure seems more intuitive to me as a "complexity
measure" when the only consonances are 3:2, 5:4, 6:5, and their inversions.

--------------------------

It would seem that your measure and mine are
very closely related. For example, here is the right half of the lattice of
all points which form intervals with 1/1 whose "expressibility" is below
log(100):


25/16-75/64
  | \   | \
  |  \  |  \
  |   \ |   \
 5/4--15/8--45/32
  | \   | \   | \
  |  \  |  \  |  \
  |   \ |   \ |   \
 1/1---3/2---9/8--27/16-81/64
  | \   | \   | \   | \   |
  |  \  |  \  |  \  |  \  |
  |   \ |   \ |   \ |   \ |
 8/5---6/5---9/5--27/20-81/80
  | \   | \   | \   | \   |
  |  \  |  \  |  \  |  \  |
  |   \ |   \ |   \ |   \ |
32/25-48/25-36/25-27/25-81/50

while here is the lattice with "expressibility limit" 150:

125/64
   | \
   |  \
   |   \
 25/16-75/64
   | \   | \
   |  \  |  \
   |   \ |   \
  5/4--15/8--45/32-135/128
   | \   | \   | \   |
   |  \  |  \  |  \  |
   |   \ |   \ |   \ |
  1/1---3/2---9/8--27/16-81/64
   | \   | \   | \   | \   |
   |  \  |  \  |  \  |  \  |
   |   \ |   \ |   \ |   \ |
  8/5---6/5---9/5--27/20-81/80
   | \   | \   | \   | \   |
   |  \  |  \  |  \  |  \  |
   |   \ |   \ |   \ |   \ |
 32/25-48/25-36/25-27/25-81/50
   | \   | \   | \   | \   |
   |  \  |  \  |  \  |  \  |
   |   \ |   \ |   \ |   \ |
 128/--196/--144/--216/--142/
  125   125   125   125   125

Meanwhile, if we consider the --- connections to have length log(3) and the 

|
|
|

and

\
 \
  \

connections to have length log(5), the points within log(100) units along
the connections from 1/1 are:

25/16-75/64
  | \   | \
  |  \  |  \
  |   \ |   \
 5/4--15/8--45/32
  | \   | \   | \
  |  \  |  \  |  \
  |   \ |   \ |   \
 1/1---3/2---9/8--27/16-81/64
  | \   | \   | \   |
  |  \  |  \  |  \  | 
  |   \ |   \ |   \ |   
 8/5---6/5---9/5--27/20
  | \   | \   | \   | 
  |  \  |  \  |  \  | 
  |   \ |   \ |   \ |   
32/25-48/25-36/25-27/25
                                  
while the points within log(150) units are:

125/64
   | \
   |  \
   |   \
 25/16-75/64
   | \   | \
   |  \  |  \
   |   \ |   \
  5/4--15/8--45/32-135/128
   | \   | \   | \   | \
   |  \  |  \  |  \  |  \
   |   \ |   \ |   \ |   \
  1/1---3/2---9/8--27/16-81/64
   | \   | \   | \   | \   |
   |  \  |  \  |  \  |  \  |
   |   \ |   \ |   \ |   \ |
  8/5---6/5---9/5--27/20-81/80
   | \   | \   | \   |
   |  \  |  \  |  \  |  
   |   \ |   \ |   \ |   
 32/25-48/25-36/25-27/25
   | \   | \   | \   |
   |  \  |  \  |  \  |
   |   \ |   \ |   \ |
 128/--192/--144/--216/
  125   125   125   125

So my intuition was only a little bit off when I thought they'd be the same!