A useful technique for a simple-substitution cipher, particularly those of the difficult puzzle variety (aristocrats), is the "consonant-line" method, which identifies a small group of consonants.  This is described in chapter X (by George Lamb) of Helen Fouche Gaines classic text, Cryptanalysis.  This was the first book that I special-ordered (from a book store); I was 13 at the time. It is now available on Google books.

A useful technique for a simple-substitution cipher, particularly those of the difficult puzzle variety (aristocrats), is the "consonant-line" method, which identifies a small group of consonants.  This is described in chapter X (by George Lamb) of Helen Fouche Gaines classic text, Cryptanalysis.  This was the first book that I special-ordered (from a book store); I was 13 at the time. It is now available on Google books.

It works fine here (Maple 12), in both 2D Input and Maple Input.

It works fine here (Maple 12), in both 2D Input and Maple Input.

Shorter, but less efficiently:

map(`[]`@op, [G]);

Somewhat obtusely:

ListTools:-LengthSplit(map(op,[G]),2);

Shorter, but less efficiently:

map(`[]`@op, [G]);

Somewhat obtusely:

ListTools:-LengthSplit(map(op,[G]),2);

I forgot to include the evalf (I just typed it in, rather than cutting/pasting). 

Note that with the Array approach, you don't need to use evalf because the Array was declared as datatype-float[8] (which is a hardware float).

I forgot to include the evalf (I just typed it in, rather than cutting/pasting). 

Note that with the Array approach, you don't need to use evalf because the Array was declared as datatype-float[8] (which is a hardware float).

Good suggestion. I find it much easier to deal with symbols than floats. Also, it is easier to pass both equations to solve and solve them simultaneously:

e1 := a*x3 + b*y3 + c*z3 - d = 0:
e2 := cos(theta)* x3 + sin(theta)* y3 - 1 = 0;
sol := solve({e1,e2},{x3,y3});

params := [a=0.6907,b=0.3336,c=0.6415,d=2.203]:
solnum := subs(params, sol);

Good suggestion. I find it much easier to deal with symbols than floats. Also, it is easier to pass both equations to solve and solve them simultaneously:

e1 := a*x3 + b*y3 + c*z3 - d = 0:
e2 := cos(theta)* x3 + sin(theta)* y3 - 1 = 0;
sol := solve({e1,e2},{x3,y3});

params := [a=0.6907,b=0.3336,c=0.6415,d=2.203]:
solnum := subs(params, sol);

The file with the .hdb extension stores the help pages; not having it will not affect the operation.

Did you reassign the Maple libname variable so that the directory containing the new library precedes the default directory?  Just type libname in a Maple and see what it is assigned to.

Your first interpretation of "total derivative" is, I think the common one. I suggested using the Jacobian because I didn't see how the total derivative would apply here [there was no mention of the independent variable] and, more to the point, the Wikipedia entry for total derivative states that it is a synonym for the differential map.  I've never heard that usage, but it seemed a better fit...

Your first interpretation of "total derivative" is, I think the common one. I suggested using the Jacobian because I didn't see how the total derivative would apply here [there was no mention of the independent variable] and, more to the point, the Wikipedia entry for total derivative states that it is a synonym for the differential map.  I've never heard that usage, but it seemed a better fit...

Thanks for the counterexample.  It has a limit of 1 at 0 along any straight line through 0, but along the curve y=x^2 the limit is 0.

Thanks for the counterexample.  It has a limit of 1 at 0 along any straight line through 0, but along the curve y=x^2 the limit is 0.