According to the article, the edges are dulled. Also they're going pretty slow in these videos and swinging at ranges where they can barely reach each other. It looks to me like a collaborative sort of sparring. If they were trying to hurt each other, or maybe even "win" at all, I assume it would be much faster and more brutal.
At time 1:56 in the the video, the older guy "lets rip" a little bit and speeds up. It gets a murmur and a grin from the crowd, giving the impression that they are aware of what it can be.
Definitely has a bit of a Capoeira "trust me bro this would work great in a real fight" vibe in all the videos. Would be interesting to see more full-speed sparring, and also see how it would evolve with protective gear and stand-in weapons that let them really go at each other.
I wonder how similar / different this to the Filipino martial art known as Arnis, which is also bladed weapon + stick based?
Perhaps an interesting cultural example of convergence or parallelism (depending on how you think about the shared influence of Spanish colonialism w/respect to the development of these martial arts in Columbia and the Phillipines).
They want the tradition to continue, but they should also record everything: get experts and record them from multiple angles, record demonstrations, etc... With image/video => 3d, and AI to recreate the system of fighting, it would be 'saved' for the future.
There seems (to me) that there is too significant of a disconnect between a cultural practice handed down through generations and an AI-trained bot that can tell you about it.
If AI (or a book) can tell me something that has been passed down through generations, it was still passed down through generations. The problem with books is that they often don't/can't give you enough detail to do the thing. AI doesn't have that problem.
Hell, a few books and a bunch of videos don't have that problem. If people forgot about ("regular") fencing tomorrow, and 1000 years from now came across a cache of 1000 hours of fencing matches and a handful of books, I would have faith in their ability to recreate fencing accurately and then innovate on it.
Martial arts are not solely collections of techniques. They also contain important cultural elements, historical context, social relationships, moral standards, etc. Even at the practical level, a significant portion of what separates different styles of martial arts is not just their techniques but also their mentality.
All of that would be extremely difficult to capture via AI.
The older and more experienced I get, the more I've come to believe that the non-technical aspects of the martial arts are essential and neglecting them can produce negative consequences.
What advantage would an AI have over video, especially from a couple different angles? Trying to apply AI to this problem seems like a near-terminal case of have-hammer-see-nails syndrome.
Sounds interesting, and was only able to find one video showing it in a quick search on YouTube:
https://www.youtube.com/watch?v=7_YoaP0cHpQ
i got this one recommended in the sidebar https://www.youtube.com/watch?v=Vo6axEMVRcY
What are the rules? And how do they not end up hurting each other?
According to the article, the edges are dulled. Also they're going pretty slow in these videos and swinging at ranges where they can barely reach each other. It looks to me like a collaborative sort of sparring. If they were trying to hurt each other, or maybe even "win" at all, I assume it would be much faster and more brutal.
At time 1:56 in the the video, the older guy "lets rip" a little bit and speeds up. It gets a murmur and a grin from the crowd, giving the impression that they are aware of what it can be.
https://youtu.be/7_YoaP0cHpQ?t=116
Definitely has a bit of a Capoeira "trust me bro this would work great in a real fight" vibe in all the videos. Would be interesting to see more full-speed sparring, and also see how it would evolve with protective gear and stand-in weapons that let them really go at each other.
You might like the messer sparring video someone linked further downthread. Similar flavor, without the sticks.
holy shit that's terrifying
[dead]
So this is what is meant by hacker news
I wonder how similar / different this to the Filipino martial art known as Arnis, which is also bladed weapon + stick based?
Perhaps an interesting cultural example of convergence or parallelism (depending on how you think about the shared influence of Spanish colonialism w/respect to the development of these martial arts in Columbia and the Phillipines).
Very cool and hopefully they keep the tradition alive.
In Historical European Martial Arts, we have messer fighting
Casual sparring: https://www.youtube.com/watch?v=PY-y_Y7IUfY
More cinematic: https://www.youtube.com/watch?v=HwHNzL9-zpg
They want the tradition to continue, but they should also record everything: get experts and record them from multiple angles, record demonstrations, etc... With image/video => 3d, and AI to recreate the system of fighting, it would be 'saved' for the future.
There seems (to me) that there is too significant of a disconnect between a cultural practice handed down through generations and an AI-trained bot that can tell you about it.
Where?
If AI (or a book) can tell me something that has been passed down through generations, it was still passed down through generations. The problem with books is that they often don't/can't give you enough detail to do the thing. AI doesn't have that problem.
Hell, a few books and a bunch of videos don't have that problem. If people forgot about ("regular") fencing tomorrow, and 1000 years from now came across a cache of 1000 hours of fencing matches and a handful of books, I would have faith in their ability to recreate fencing accurately and then innovate on it.
Martial arts are not solely collections of techniques. They also contain important cultural elements, historical context, social relationships, moral standards, etc. Even at the practical level, a significant portion of what separates different styles of martial arts is not just their techniques but also their mentality.
All of that would be extremely difficult to capture via AI.
The older and more experienced I get, the more I've come to believe that the non-technical aspects of the martial arts are essential and neglecting them can produce negative consequences.
What advantage would an AI have over video, especially from a couple different angles? Trying to apply AI to this problem seems like a near-terminal case of have-hammer-see-nails syndrome.
Reminds me of Haitian Machete Fencing:
https://www.youtube.com/watch?v=7p_NUEn7F_g
Just when I thought I'd seen everything, here comes organized fighting with actual machetes -- oh, and without protective gear.
Looks pretty similar to escrima.
Well, the video linked above is labelled "Esgrima de Machete y Garrote.".