> Posting by Claude:

> "I didn't say that each glyph would need a separate method for
> solving it. I suggested that you might stop looking for numerical links
> between glyphs.
> There's a difference."

> This means that the numbers in each glyph follow the same rule, applied
> seven times, just changing a parameter or, probably, two parameters (after
> all, we are looking for a username, and a password, two (seven-letter?)
> words).

Keep in mind that the rule within a glyph could be applied differently depending on the letter in the glyph (A,R,R,I,V,A,L or L,C,C,T,G,L,W).

> Ok: then there should be a general rule, which is as follows

> F(A,B,C) = # of place C in the glyph whose parameters are A and B.

> Problem: find the rule, then find A and B :)

I'm not sure what your function F would give us -- one number? Two numbers? One or two letters?

> Subproblem: find the way numbers in glyph are counted: hours on a clock?
> Simple 1,2,3 and so on rule? Angles (0, 45, 90)? Coordinates (place the
> eight numbers in a 3*3 (another tic-tac-toe) grid, with the center empty)?

> I'm trying some patterns, still with no results...

> Any ideas?

Given Chelsea's comments in her two most recent messages, I wonder if we are supposed to determine four numbers that are missing from the glyph faces? She said in one message that the clocks were numbered strangely, and in another message that "two become two more" on the clocks.