Read a good dance lately?
dancenotation.org
dancenotation.org
Seeing this, I can't avoid to think in formal languages. Maybe, even it's possible to define, formally, a particular dancing style as a grammar that doesn't admit movements from other styles.
There are many layers of complexity here, perhaps not all too different in nature from the layers of complexity when dealing with computers and software. In dancing, the goal is usually to start somewhere in the middle and make your way up towards more abstract ways of thinking about it and also down towards more fine technical detail.
[1]: https://www.worlddancesport.org/WDSF/Academy/Technique_Books
For sign language: https://en.wikipedia.org/wiki/SignWriting
Those systems, designed for different purposes, lack essential elements necessary for martial arts movements, like weight distribution, facing, weapons, footing, and opponents. You would probably have better luck examining the code for mocapped animations, particularly for 3-D fighting games and movies.
But you'll have to come up with your own abstraction. There are just too many degrees of freedom in a human body to have a one-size-fits-all kinetography system that is also easy to read and write. That's probably why some martial art systems have so many names for positions and movements. The periodic table for atomic martial art movements is huge, especially now that MMA has spurred so much cross-disciplinary study.
I think mocap has a lot of potential (though I’m biased here[1]).
For example, there have been some interesting experiments capturing performances by sportspeople of different levels and comparing them. These might be looking for patterns in what the better performers do differently, which might then give some ideas for how to train less experienced performers.
Some elite sports teams and athletes also use mocap technology to analyse their form and look for details that are difficult to see with human eyes in real time.
The challenge is that even state-of-the-art mocap technology is still a very long way from modelling the full complexity of the human musculoskeletal system. Remember that a lot of what you see in games, and particularly in big budget movies, was originally based on mocap data but then manually refined by skilled people for a considerable period of time. The mocap is certainly a useful starting point and it’s getting better all the time, but there’s a long way to go before it will get the fine details right.
That said, even with today’s systems, the amount of data you get from a decent mocap set-up means there is a lot you can investigate, as long as you know what you want to look for. As you say, there are a lot of degrees of freedom in a human body, and coming up with the right abstractions to capture the important details is a challenge in itself.
[1] I help to run www.ballroomgenie.com, which is a training aid featuring animations of ballroom moves using mocap-driven CGI instead of traditional live video footage. This lets us show various useful effects that live footage can’t, typically because of physics or because in our little niche market no-one can afford movie-level camera set-ups and editing to record the same action from many different perspectives simultaneously. On the other hand, even working with high calibre performers and studio as we do, it still takes a lot of work to get acceptable quality, and we’re still a long way from being able to reproduce truly lifelike footage. This is partly because we don’t have a Hollywood budget and a whole team of people to get each animation just right, but it’s also because as far as I know the technology to model a human body accurately enough that things like contact and pressure are spot on simply doesn’t exist yet.
I dance and teach (Cuban) salsa, and I've been wanting to write down a "repertoire" of moves that I know and teach, to give it some structure. Here the system I'm inclined to use:
The space of patterns/figures you can do in salsa is really not very big. You can break each figure into a sequence of positions at which to be on beats 1, 3, 5, and 7 (I guess you can choose higher or lower granularity, but I think this level of granularity is just right for most people). Then, in my mind, a figure is merely a directed path through a graph where each node is a "known" position (~20?). Assuming you know how to get from one arbitrary position to another one (safe assumption for anyone in an intermediate level), then you can just read and write any figure you want.
I think this kind of structure would allow students to learn much much faster (and is basically the method I used to improve my own skill very quickly, except I used a spreadsheet instead of graph). You'll realize that most of the new stuff you learn is simply:
- composite paths made out of other paths you already knew.
- existing paths where you take detours and do something else in the middle. Those detours always start at a node, so now you can use the detour in any other path that goes through that node.
- new ways of getting from one node to another. Similarly, now you can apply this new knowledge to any move that utilizes that node, not just the move that you're learning at the time.
Where I also think this shines is in developing the dancer's (esp. leader's) creativity, which I think is one of the hardest part for most people (technical stuff is easy, creative stuff is hard!). Having such paths laid out makes it really easy for someone to mix and match, and to create potentially endless combinations by simply choosing a different path forward once you get into a known shared position.
And for fun, I'm sure you could just draw out a random path and see if you can make it work into something that looks reasonably good :)
While hand drops etc. may be more New York style, crossed hands (which arm on top) happen in lots of turns.
I couldn't find notations for partner dances, however designing notations for contacts shouldn't be difficult. Notations for lead and follow more so.
It depends how far you want to go down the rabbit hole.
For example, in ballroom, there is also a well-established way that formal technique is documented in books. For each step within a figure, it specifies in considerable detail things like which part(s) of the foot you’ll step on (heel then toe, for example), which way that foot will be pointing (roughly speaking using compass points), how the upper body will rotate relative to the lower body, and what kind of rise and fall you have through the legs and feet.
There is a whole system of teaching and examinations based on this technique, and consequently most teachers really will teach the same basic technique to a beginner learning the same figure in their classes. Of course, this is very useful for a social style where people might dance with someone else from another school the next day!
However, even in the standard technique books with all their detailed notation, that systematic approach doesn’t tell the whole story. There are often notes on how to dance some figures that explain details not captured effectively by the formal notation. Then, once you go beyond the basic technique and standard syllabus figures that almost everyone learns, as with many activities you find that performers and teachers develop their own styles and their own personal ways of explaining more advanced technical points.
At that stage, even though the actions you’re talking about are usually consistent with the traditional technique, there is still a lot of scope for interpretation. So, I doubt you could ever have a systematic notation that captures all the subtleties of the actions that set apart a world champion from a quarter-finalist. It would take more than a lifetime of study, and by then the art would have evolved anyway.
http://articles.latimes.com/1995-06-02/entertainment/ca-8554...