Writing on Standing Water

A reader writes "Engadget is reporting on Japanese scientists who have found a way to 'write' characters on the surface of water using waves. This looks very cool - but the time required to change character seems very high (15-30 seconds). From the article: 'Liquid-based displays are nothing new -- in a vertical orientation, at least -- but apparently it's a lot more difficult to coax a standing pool of water into forming recognizable shapes and characters.'"

Neato!

[blugu64]

Looks pretty cool, what are some of the real world applications of this?

I thought they meant the opposite

[oskard]

For example, writing a word in water and having the perimeter of the pool recognize the waves, and convert it to digital text. Microscopically, that could actually have a use with a liquid enclosed touchscreen.

John Keats

[j1m+5n0w]

I believe that phrase likely originated with the poet John Keats [wikipedia.org]:

He died on February 23, 1821 and was buried in the Protestant Cemetery, Rome. His last request was followed, and thus he was buried under a tomb stone reading, "Here lies one whose name was writ in water."

Re:outer space

[peacefinder]

A freind of mine once wrote a message with a much lower refresh rate... he planted trees. Somewhere in northeastern Washington, aliens (or pilots) may be startled to see a certain naughty word beginning with "f" spelled out over a couple acres.

Hmm. Actually, I ought to get the coordinates and check google maps...

Re:Processing time?

[gardyloo]

Part of the trouble is that these things are NOT stationary. There's only once in a long time that all of the waves produced by those actuators end up forming the characters they want them to, AND all the rest of the surface is smooth. And that can only be done with some sophisticated feedback as to what waves are present. I haven't read their paper, but I suspect they either sense some inductances at the edge of the tank, or do some fancy laser-scanning of the surface. I can easily see incorporating the continually-changing conditions into the calculations as taking a long time. And cylindrical Bessel functions are not so easily precomputed if you need 50 of them at a particular time. I'd think the easiest way to do that is to set up 50 analog circuits with the appropriate parameters and continuously feed in the water heights along the edge of the tank.

      For applications... I can't answer this in full, since part of my research is sort of related. But for detecting things buried in the seafloor, ripples on the seafloor do some amazing things to signals. Having a reliable way to set up such ripples in the laboratory is very useful.

CAT Scan in Reverse

[Mignon]

I think this is somewhat related to CAT scans, but applied in reverse. My crude understanding of a CAT scan is that sensors are placed in a ring shape around the object to be scanned and a series of "slices" are scanned. For each "slice", this gives a set of axial views through the object (the 'A' in CAT). Bessel functions are the mathematical tool that let you convert the axial data - which is a kind of sum across a diameter - into data at each point in the circular cross section.

It seems they've reversed this process and solved for the axial data given the point-by-point data - e.g. the rasterized character.

By the way, CAT scans and Bessel functions are one of the examples of "abstract" math that later turns out to have practical application.

Re:John Keats

[j1m+5n0w]

Nice to see some people still know their poetry. :-)

Actually, I don't. I just happen to have recently been reading the (science fiction) Hyperion books [iblist.com] by Dan Simmons (not to be confused with an epic poem by Keats of the same name), in which Keats (or, actually, a reconstruction of Keats meant to be similar to the original historical Keats) plays a significant role.

Re:Interesting, but why?

[Amouth]

you could also use the idea on more substances like say.. epoxy or rasien. to create shapes in a certian form quicker than some one could carve one.. i am thinking for master molds for boats and stuff.. where it takes months to a year to make the master.. if you could put in a plotted file and get a close shape form this thing.. even if it took it a week to make the shape it would be worth every bit of $

Rangoli on Water?

[140Mandak262Jamuna]

In India women do holiday decorations using colored powders. Patterns like this: [twilightbridge.com] or this [bremen.de] Google image search on Rangoli [google.com]

Sometimes they do it on a large piece of blotting paper and carefully place it on a large bowl of water. The paper soaks the water and sinks, leaving behind patterns floating on the surface of water.

No need to understand Bessel functions and Legendre Polynomials.

Additional pics, and this was done in 2004

[riffer]

Here's a link to a PDF document [mes.co.jp] from Akiken's website that gives some more details about the project. Unfortunately I can't read Japanese, so maybe someo kind soul can provide a translation? However, the PDF does have some additional pics, taken in sequence. I'm not sure but it seems like the oscillation affect takes a few moments to build (the 15 to 30 seconds I'm guessing) and then momentarily generate the character.

The nature of water being fluid means it would be unlikely the character would remain for long, unless the agitation level was very high. Which may well be the case... Water can seem pretty solid if it's under enough pressure and/or moving fast enough. Also, based on the PDF and the Akiken website, looks like this was something they developed in 2004. But presumably it wasn't published at the time...