Kinetic Waves
March 30, 2009 8:25 AM Subscribe
Yeah, those were really cool. Kinda makes you rethink the use of the word "mechanical" to mean "stiff and unnatural".
posted by DU at 8:32 AM on March 30, 2009
posted by DU at 8:32 AM on March 30, 2009
this is great. thanks, rageagainsttherobots. this should be a nice salve for the rather heated art exchanges frothing around mefi lately. no gratuitous provocation, no pomo posturing. just simple imagination, craft, beauty, and wonder. art that doesn't require theory. or even thinking. aaahhhh . . .
posted by barrett caulk at 9:16 AM on March 30, 2009
posted by barrett caulk at 9:16 AM on March 30, 2009
Awe-inspiring understanding of patterns in nature. Spectacular and beautiful, really.
posted by Skygazer at 9:19 AM on March 30, 2009
posted by Skygazer at 9:19 AM on March 30, 2009
So great.
It's interesting to think about how difficult it is to represent water with complex 3D modeling software and yet this guy does a such good job with only cardboard and string.
posted by scottatdrake at 9:28 AM on March 30, 2009
It's interesting to think about how difficult it is to represent water with complex 3D modeling software and yet this guy does a such good job with only cardboard and string.
posted by scottatdrake at 9:28 AM on March 30, 2009
It's interesting to think about how difficult it is to represent water with complex 3D modeling software and yet this guy does a such good job with only cardboard and string.
So waaay back in Scientific American there was an article on "analog computers". Probably by Martin Gardner (reprinted here I think), showing how string and some weights can often do almost instantly what it takes a computer much longer to do. Anyone who has ever had to write a convex hull algorithm for a computational geometry class knows that it's not easy (not that hard either, but hard enough). Considering you can do it with some nails and a rubber band in a few seconds it makes the software solution downright inelegant.
posted by GuyZero at 9:50 AM on March 30, 2009
So waaay back in Scientific American there was an article on "analog computers". Probably by Martin Gardner (reprinted here I think), showing how string and some weights can often do almost instantly what it takes a computer much longer to do. Anyone who has ever had to write a convex hull algorithm for a computational geometry class knows that it's not easy (not that hard either, but hard enough). Considering you can do it with some nails and a rubber band in a few seconds it makes the software solution downright inelegant.
posted by GuyZero at 9:50 AM on March 30, 2009
makes the software solution downright inelegantIMO, it makes that particular software solution inelegant. Most software solutions indeed. We'll invent something equally good, we just have to invent some more nice and intuitive geometric building blocks to put in the giant invisible software lego box.
What I mean is that the difference between the nice and fun rubber band solution and the arcane AutoCAD version is that there's so much more implicit in the rubber band. The rubber band is the equations. We need computerized objects that contain and manifest the equations and let us get them out of our hair. Then we'll be all up in drawing boat hulls in the air with electronic fingers. Software expressivity just has to catch up a bit first. Let's talk in 10 years.
posted by krilli at 10:40 AM on March 30, 2009
Can anyone help me with this: Make a better rubber band, and nature will invent a better [$PUNCHLINE]. ?
posted by krilli at 10:42 AM on March 30, 2009
posted by krilli at 10:42 AM on March 30, 2009
Very nice stuff, and what an unpretentious guy, too. Very pleasant.
posted by echo target at 10:43 AM on March 30, 2009
posted by echo target at 10:43 AM on March 30, 2009
These sculptures are wonderful. Given the means and the time, it's easy to see how one could get enraptured making these.
posted by bigmusic at 10:56 AM on March 30, 2009
posted by bigmusic at 10:56 AM on March 30, 2009
I expected to snark at the snobby artist but wow, that was cool. Plus, as someone said above, he seems like a down-to-earth guy.
I'll be thinking about this art all day. Thanks for the link.
posted by Tacodog at 11:01 AM on March 30, 2009
I'll be thinking about this art all day. Thanks for the link.
posted by Tacodog at 11:01 AM on March 30, 2009
Considering you can do it with some nails and a rubber band in a few seconds it makes the software solution downright inelegant.
You can? Given the nails and the rubber band, tell me the equation or coordinates of the hull.
Analog computers are pretty awesome, but they have some major problems of their own.
posted by DU at 11:11 AM on March 30, 2009
You can? Given the nails and the rubber band, tell me the equation or coordinates of the hull.
Analog computers are pretty awesome, but they have some major problems of their own.
posted by DU at 11:11 AM on March 30, 2009
Equations and coordinates are just artifacts of non-analog computing, though. If you calculate all your curves and positions by analog means, then there's no need for anything like a coordinate.
They don't scale up so well in complex systems, of course, and drawing a curve that's physically larger than another takes more work, so the drawbacks of most analog computing put it out of the running for real-world applications.
Still, there's lots of great ones that come in handy. Want to find the center of a square, or any regular polygon with an even number of sides? Just connect the opposite corners with straight lines, and the intersection is your center. This is even cooler (and handier) when you discover that it even works when the objects are in perspective. If you're ever drawing some architecture with vanishing points, this technique is really handy for making sure the window spacing (or whatever) comes out right.
posted by echo target at 11:44 AM on March 30, 2009
They don't scale up so well in complex systems, of course, and drawing a curve that's physically larger than another takes more work, so the drawbacks of most analog computing put it out of the running for real-world applications.
Still, there's lots of great ones that come in handy. Want to find the center of a square, or any regular polygon with an even number of sides? Just connect the opposite corners with straight lines, and the intersection is your center. This is even cooler (and handier) when you discover that it even works when the objects are in perspective. If you're ever drawing some architecture with vanishing points, this technique is really handy for making sure the window spacing (or whatever) comes out right.
posted by echo target at 11:44 AM on March 30, 2009
If you calculate all your curves and positions by analog means, then there's no need for anything like a coordinate.
There is if I want to analyze the curve. For instance, what if I want to figure out what curve is described by a hanging chain and from that derive other properties via pencil and paper? I need that equation.
There's a book in the library here about analog computers written before the days of digital ones. It's full of awesome drawings of wheels and gears and whatnot. I'm afraid to check it out for fear I'll never resurface.
posted by DU at 11:54 AM on March 30, 2009
There is if I want to analyze the curve. For instance, what if I want to figure out what curve is described by a hanging chain and from that derive other properties via pencil and paper? I need that equation.
There's a book in the library here about analog computers written before the days of digital ones. It's full of awesome drawings of wheels and gears and whatnot. I'm afraid to check it out for fear I'll never resurface.
posted by DU at 11:54 AM on March 30, 2009
a photo I took of an analog computer. From Downsview Air Museum in Toronto. I have no clue what it was used for.
posted by GuyZero at 12:04 PM on March 30, 2009
posted by GuyZero at 12:04 PM on March 30, 2009
It amazes me how many cool people like this there are out there. And if you find one, they usually have a few others in their circle.
...uses everything from wood to cardboard...
That leaves what? Particleboard? Masonite?
posted by StickyCarpet at 12:04 PM on March 30, 2009
...uses everything from wood to cardboard...
That leaves what? Particleboard? Masonite?
posted by StickyCarpet at 12:04 PM on March 30, 2009
Yeah, for abstract analysis and measurements it's hard to beat the non-analog means. The two complement each other, though. A hanging chain is a catenary, but as you add weight along the length of it (as in a suspension bridge) it becomes a parabola. You can show this by equations & graphs or you can show it by making a model, but you understand it best if you do both. And it's really pretty amazing what can be done and what levels of precision can be reached with processes that never use a 1 or 0.
posted by echo target at 12:10 PM on March 30, 2009
posted by echo target at 12:10 PM on March 30, 2009
GuyZero, that's awesome. What does it do?
posted by echo target at 12:10 PM on March 30, 2009
posted by echo target at 12:10 PM on March 30, 2009
It puzzles museum visitors. There was no signage or description on it at all. It was WWII-era so it may have been for aircraft design or calculating bomb trajectories or anything.
posted by GuyZero at 12:13 PM on March 30, 2009
posted by GuyZero at 12:13 PM on March 30, 2009
I believe they also made analog computers for firing shells air-to-air, which is pretty impressive. That looks way too big to be on a plane, though, and way too complicated to use in combat. Maybe it's a code-breaker?
posted by echo target at 12:24 PM on March 30, 2009
posted by echo target at 12:24 PM on March 30, 2009
It may have been for route plotting; there were a number of stations that had maps and these elaborate mechanical digitization mechanisms for getting map points into some sort of mechanical computing system.
To bring this back to the OP, one can image that 50 years ago Margolin would have done exactly what he's doing today except the output would be aircraft instead of art.
posted by GuyZero at 1:08 PM on March 30, 2009
To bring this back to the OP, one can image that 50 years ago Margolin would have done exactly what he's doing today except the output would be aircraft instead of art.
posted by GuyZero at 1:08 PM on March 30, 2009
That guy is a genius.
posted by ornate insect at 5:18 PM on March 30, 2009
posted by ornate insect at 5:18 PM on March 30, 2009
one can image that 50 years ago Margolin would have done... aircraft
or perhaps he would have been a an groundsweeper in the the east bay regional parks district, written about native california, founded magazines, and been knocked out by the inherant goodness of the human race.
well, starting 40 years ago, anyway, when one can imagine his son Reuben was born.
posted by ioesf at 5:50 PM on March 30, 2009
or perhaps he would have been a an groundsweeper in the the east bay regional parks district, written about native california, founded magazines, and been knocked out by the inherant goodness of the human race.
well, starting 40 years ago, anyway, when one can imagine his son Reuben was born.
posted by ioesf at 5:50 PM on March 30, 2009
« Older Maurice Jarre | Cockpit, amirite? Newer »
This thread has been archived and is closed to new comments
posted by rageagainsttherobots at 8:26 AM on March 30, 2009