Richard, this is great thanks for posting. If you'd like you could write a separate post announcing the tutorial itself, maybe saying a few words how you came about writing it, what it covers, whom it intended for, - whatever you feel suitable. About Game of Life, we'd like to mention a related Demonstration, not sure if you have seen it, - 2D Cellular Automaton Animations. We think you and other would be curious to take a look. It covers not only Game of Life, but also a lot of similar in nature Cellular Automata.

The Caption section to 2D Cellular Automaton Animations is, arguably, incomplete. I did not see anything that would indicate it is incorrect. (We try to catch outright errors in the curation process. Granted, some get through.)

glad you find the tutorial useful. i use Courier when i write technical material becuase it is the traditional typeface for code. (http://en.wikipedia.org/wiki/Courier_(typeface)#In_computer_programming) and i like the look.

De gustibus non est disputandum 

i don't want  to belabor the point (after all. it's not my demo) but i would suggest that the description be changed from "depending only on the number of black cells surrounding a given cell." to "depending on the color of a given cell and the number of black cells surrounding it". it would also be useful to state in the description that the Demo can be used for outer-totalistic cellular automata (it's good to use the proper notation when possible for educational purposes).

btw - here's a pretty interesting video for 1D CA freaks:
Minecraft - Elementary Cellular Automata Visualizer (Complete Wolfram Code) 

note - i am not one of those. i mean, i am a geek - and proud of it - but don't i find 1D CA's all that interesting, possibly because SW has already studied them ad nauseum, and  because i think that dimensionality is all-important (that probably comes from my having been a random- and restricted-walk specialist and dimensionality is all-important in random walk properties and also because i live in a multi-dimensional world (though i'm not sure exactly how many dimensions) and i think (am i wrong?) that the 2D GoL shows ALL the interesting properties of 1D CA's (e.g. being Turing complete)  and additionally, clearly demonsrtates emergent phenomena (e.g. the glider pattern) which has great relevance to foundational (fundamental) issues in physics (on the last page of the tutorial (p.43), i refer to material by Hawking and by Israeli and Goldenfeld which is TOTALLY (and i don't mean just outer-totally LOL) relevant to people thinking about things such as computational irreducibility and the limitations of computability (and hence of physics and perhaps all of science)),   

unfortuntely, as noted in Wolfram code, there are 2^512 = 1.341 x 10^154 possible two-dimensional CA's ( for the simplest CA's with  only two states  and only touching cells being considered as belonging to a neighborhood (btw - does anyone know how this number is affected by the use of the von Neumann neighborhood rather than the Moore neighborhood?).

aside: maybe we can re-write Arthur C. Clark's "The Nine Billion Names of God" short story using geeks calculating ALL of the possible 2D CA's. that way, the story ending will be more upbeat since they'll never complete their task during the 'natural' lifetime of the universe.

the Wolfram Code description seems strange to me (it says that " the Wofram code does not specify the size (nor shape) of the neighborhood, nor the number of states" but that's wrong - s does specify the number of states. 

r = number of rules
s = number of states
n = neighborhood size (radius of the neighborhood)  
r = s^(s^((2 n + 1)^d))

and my question remains: what neighborhood (Moore or von Neumann or something else) does n = 1 in 2 dimensions correspond to - don't BOTH have an n value of 1 (only nearest-neighbor cells are in the neighborhood in each case) even though certainly, they must have a different number of possible rules since one has 4 neighbors whie the other has 8 neighbors.