A few comments (opinions) on the Dog Language guide.
One should bear in mind that the type and intensity of a particular signal may vary with the particular dog. Watching hundreds of dogs being introduced to play groups you can see both the variation and the learning process that takes place, especially with those just learning to be social.
On direct eye contact, never assume. Remember that dogs have co-evolved with humans and many will modify and adjust as they gain experience. Thus, there are many dogs who see a direct stare from a human as only an invitation to lick the person's nose. With any new dogs I meet, I let him tell me how he's going to interpret things like that.
Your own dog interprets your body signals much more than you think. Many think that a dogs obedience is about the fact that we have taught them certain words: "sit" ,"down", "stay" etc. But the words are just a little part of obedience.
I feel you made a very good point often missed elsewhere. Few training books mention this, but it should always be taken into account when training a dog. And don't forget the other side of this coin. How many times have your heard somebody say:
I never said a thing, but somehow my dog knew that I was going to (or wanted him to) do X.
Or,
I never said a thing, but he must know he did wrong because he looks so guilty.
On anthropomorphism, you may have gone just a little too far. There are a number of notions that dogs do express as people do, just not to nearly the same degree. Just because a dog won't work towards a goal that's years away like a person, that doesn't mean they always live in the moment. Hope, for instance, is a future projection and extinction studies on training dogs have shown its high value which means it does exist.
Add to that the very large difference between various dogs (and various people, for that matter).
Ah, how very much we humans want simple guides to a dog's body language. But the reader should remember those guides are just a start, and you need to interpret the intensity and specialization from the particular dog's response to a stimulus. Your section on hackles, for instance. I feel it is generally correct, but there are exceptions. And the same applies to all the items. Not really a criticism of your article, but a qualification for people reading it. I find far too many people seem to insist on simple rules that always apply the same way.
On dominant dogs, nicely stated! At the shelters, we make much use of them in play groups. I liked your Husky/Black-dog example.
Nice write-up on the alpha roll! A slight correction here that relates to a different context, where a dog may physically force another dog on their back as part of a play sequence. Often nothing to do with submission, and I agree with what you stated.
On Sound and dog growling, it's fun to watch new volunteers running the dog play groups. A few dogs vocalize quite a bit and what would be seriously warning sounds from some of the other dogs mean little here. Initially, you can tell the difference from the rest of their body language. Over time, the human ear learns enough discrimination to go by only the sound, keying on very small differences.
On your Touch section, I feel the paw-on-the-sholder is contextual and not always a fight preamble. Often I've seen one dog stiff and hard when throwing a paw over another's shoulder. The other dog then signals back and this signal together with the originator's response may cause them to break apart, start to fight, or often to begin a play session. This last case is more commonly found during a short period when a dog is first learning play socialization with others.
On your pictures, I was pleasantly surprised. I think you did a great job there. Even after all the comments I threw in here, I think you did very well.