Slashdot Mirror
Play GNU Chess On Your Scanner
leighklotz writes "Debian developer and Internet Mail Archive founder Jeff Breidenbach of PARC has made GlyphChess, a chess-playing
copier using Python, GNU Chess and DataGlyphs attached to
the bottom of the pieces. DataGlyphs are cool 2D barcodes made out of / and \ marks for ones and
zeros that use the coding from CDs for error coding. If you don't happen to have a Xerox machine at home, it also works with SANE..."
And the "Most pointless thing ever" award goes to...
How do you undo a move, tear up the last page of paper?
Trolling is a art,
Do you get to send the winner a photocopy of your ass when you lose?
I have often regretted my speech, never my silence.
-Xenocrates
The Ubiquitous Reed-Solomon Codes
by Barry A. Cipra
Reprinted from SIAM News, Volume 26-1, January 1993
In this so-called Age of Information, no one need be reminded of the importance not only of speed but also of accuracy in the storage, retrieval, and transmission of data. It's more than a question of "Garbage In, Garbage Out." Machines do make errors, and their non-man-made mistakes can turn otherwise flawless programming into worthless, even dangerous, trash. Just as architects design buildings that will remain standing even through an earthquake, their computer counterparts have come up with sophisticated techniques capable of counteracting the digital manifestations of Murphy's Law.
What many might be unaware of, though, is the significance, in all this modern technology, of a five-page paper that appeared in 1960 in the Journal of the Society for Industrial and Applied Mathematics. The paper, "Polynomial Codes over Certain Finite Fields," by Irving S. Reed and Gustave Solomon, then staff members at MIT's Lincoln Laboratory, introduced ideas that form the core of current error-correcting techniques for everything from computer hard disk drives to CD players. Reed-Solomon codes (plus a lot of engineering wizardry, of course) made possible the stunning pictures of the outer planets sent back by Voyager II. They make it possible to scratch a compact disc and still enjoy the music. And in the not-too-distant future, they will enable the profitmongers of cable television to squeeze more than 500 channels into their systems, making a vast wasteland vaster yet.
"When you talk about CD players and digital audio tape and now digital television, and various other digital imaging systems that are coming--all of those need Reed-Solomon [codes] as an integral part of the system," says Robert McEliece, a coding theorist in the electrical engineering department at Caltech.
Why? Because digital information, virtually by definition, consists of strings of "bits"--0s and 1s--and a physical device, no matter how capably manufactured, may occasionally confuse the two. Voyager II, for example, was transmitting data at incredibly low power--barely a whisper--over tens of millions of miles. Disk drives pack data so densely that a read/write head can (almost) be excused if it can't tell where one bit stops and the next one (or zero) begins. Careful engineering can reduce the error rate to what may sound like a negligible level--the industry standard for hard disk drives is 1 in 10 billion--but given the volume of information processing done these days, that "negligible" level is an invitation to daily disaster. Error-correcting codes are a kind of safety net--mathematical insurance against the vagaries of an imperfect material world.
The key to error correction is redundancy. Indeed, the simplest error-correcting code is simply to repeat everything several times. If, for example, you anticipate no more than one error to occur in transmission, then repeating each bit three times and using "majority vote" at the receiving end will guarantee that the message is heard correctly (e.g., 111 000 011 111 will be correctly heard as 1011). In general, n errors can be compensated for by repeating things 2n + 1 times.
But that kind of brute-force error correction would defeat the purpose of high-speed, high-density information processing. One would prefer an approach that adds only a few extra bits to a given message. Of course, as Mick Jagger reminds us, you can't always get what you want--but if you try, sometimes, you just might find you get what you need. The success of Reed-Solomon codes bears that out.
In 1960, the theory of error-correcting codes was only about a decade old. The basic theory of reliable digital communication had been set forth by Claude Shannon in the late 1940s. At the same time, Richard Hamming introduced an elegant approach to single-error correction and double-error detection. Through the 1950s, a number of researchers began experimenting with a variety of error
I don't need no instructions to know how to rock!!!!
Amazingly enough, it's possible to play chess using these strange "pieces" and a "board." Although the idea is novel i suppose.
- tristan
Heck, I thought that they programmed one of their advanced copiers to play chess. Not just simply using it as a hohum input device. I agree it is a cool way to test their glyphs, but not very interesting beyond that. The thought of programming your scanner/copier in python scratches that nerdy itch much nicer.
Clearly, they're running their web server from the copier, too. Paper jam!
A new and innovative way to get my arse whipped by a computer. As if losing umpteen times on the standard chessboard wasn't enough.
It's all a ploy to give him an unfair advantage over GNU Chess!
Obliteracy: Words with explosions
I think it would be much more productive and fun to scan your butt and draw a face on it. On a much more serious note, chess already can take upwards of 2 hours to play, especially if you're playing one of those super careful people. Playing it like this would have to be a weekend event. It is an interesting technical feat though, but really provides no advancement for current technology. I sat here for 5 minutes trying to think how it might help further a current idea or help people with disabilities, but it doesn't -- at all. The purpopse of new technology should be to do things quicker/cheaper. When technology becomes more advanced and provides less functionality, we might as well bundle it with windows.
NJ Local Music Scene
We've /.ed PARC. They must still be using an Alto to host the site.
-twb
"PC Load Letter"? WTF does that mean?
DataGlyph techology is patented by the Xerox corporation. The DataGlyph toolkit is a binary only library that you must license to include with your "product". Despite the use of Python and GNU Chess in this example, I doubt very much that DataGlyphs are going to be of much use to the open source community.
It seems from reading the article that the copier does NOT play chess. The copier inputs the positions into the computer, which then plays chess.
While you can buy pressure sensitive boards to attach to the computer, these are pretty expensive. It's a lot easier to play chess on a real board instead of the screen. Notice that when grandmasters play computers there is a person who runs the computer and moves the pieces. So overall, this is a pretty cool hack if you happen to be a serious chess player who also has a sufficient copier already sitting around.
We're setting up a PC. Some hdd conflict, won't boot. CD drive broken, doesn't work. Floppy drive okay but not a single bootable floppy around. Let's see what it provides more, maybe some network boot... I look through BIOS options. Oh well, SCSI. What do we have attached to SCSI? A scanner?! Hey, come on, get a pencil and write some startup code on that sheet of paper, maybe we'll succeed booting it from the scanner! ;)
45 5F E1 04 22 CA 29 C4 93 3F 95 05 2B 79 2A B2
I can run Duke Nukem on a Cue Cat.
I don't need no instructions to know how to rock!!!!
Here you go.
No, let's play global thermonuclear war.
Beauty is in the eye of the beerholder.
So all those glyphs from the Egyptians was really some sort of primitive multiple-player shoot-em up game?
This space for rent.
Finally, after more than 30 years and being spun off of Xerox, PARC finally comes up with a product involving copiers. And it's absolutely useless.
In a scale from 0 to Geek, he get's a 100.
... like Microsoft BOB? Clippy? DOS 4.0? Windows Millenium? IIS? (Well, maybe they weren't so talented after all).
...would be if a controllable electromagnet was attached to the scan head. Then, the scanner could actually move the pieces around too. You'd have to do some 'move blocking pieces into a holding area temporarily' stuff, but that would make the project even more fun, no?
Does it hurt to hear them lying? Was this the only world you had?
Some more details -- the software would constantly grab images of the board and process them in realtime. It should be able to use the redundancy to correct for errors and also to know when a move is done (since you'll move your hand away from the board for at least a dozen frames or so, even if you play fast). The board is a nice regular rectangle, and pieces are black and white circles -- even at an odd angle, it should be easy to determine the full board position.
I feel confident I could do it, but it would take me tens of hours of coding/testing, and I don't have the time, but I bet someone would love to do this for a senior project and opensource the code... please... :)
Well, first of all, he started the project over christmas since he was laid up with a broken leg, and not actually expected to do any useful work.
Second of all, his manager seemed to think so.
Thirdly, most research looks "useless" at first glance. PARC is a research place. If people like you were running it, I guess ethernet and GUIs would never have been invented.
Now if only we could figure out how to profit on any of those cool useless inventions.
blah blah blah
You have all these (admittedly talented) programmers spending all this time on worthless projects
In a way, of course, he's right. By it's very nature, the Linux community is somewhat fractuous, and not nearly as efficient as it could be. Old news, even if the problem isn't solved. (what, everyone doesn't use KDE? oh)
But it doesn't apply here.
This is a case of brilliantly applied science, and if you don't think that it is useful, consider a few of the implications. This is EXACTLY the type of technology that is going to be highly useful in the coming years, because they are coming up with ways to bridge the visual gap, and in developing the best ways to represent data visually in a form that a computer can read, and that will handle error-correction gracefully.
That is the kind of applied research that needs to be done before all of the cool, non-'worthless' stuff can happen. Before you get your car that can navigate on its own, you need a Xerox chess set -- that's how this stuff works.
For people who want to play with error correcting codes, you can get the source code for a Reed-Solomon decoder/encoder I wrote here:
https://sourceforge.net/projects/rscode/
I guess ethernet and GUIs would never have been invented.
Bah who needs 'em. I've got a CLI(and it even runs at 1280x1024 in framebuffer mode) and ethernet is overrrated. What does it do that i couldn't do with some hi-tech cans and string. Don't even have to deal with shitty construction cable breaking that way.
I don't get why people get pissy about cool hacks like this being "useless". Art is also useless. Sitting around watching The Bachelor is useless. Getting a PhD in math is useless. People do this kind of thing because it's fun to them. What's the big deal?
blah blah blah
if / meant 1 and \ meant zero then a glyph that is /\ would be 10 right? So what if the glyph was upside down so that it looked like \/, 01? It looks like the chess scanner program tries to match the pattern with a reference so even if it is upside down it should work, but how is this "glyph" system used to encode information if the slash order become backwards when turned upside down?
By quite a bit, actually. Look here down around 1989.
Also, on the same history page, in the Mid 80's section, you'll see an entry for an expert system named Pride developed at PARC. Pride helped Xerox design their first line of desktop copiers, and is quite famous within the company.
I worked for (long lamented) Xerox AI systems from 1986-88, and consulted for them off and on through 1994, which is how I know about this.
To a Lisp hacker, XML is S-expressions in drag.