1770 Mechanical Chess Player Inspired Babbage

dipfan writes "A new book tells the extraordinary true story of a clock-work chess-playing "machine" named The Turk that wowed Europe and the US in the 18th and 19th century, beating Benjamin Franklin and Napoleon, among others. Although it turned out to be a cleverly designed trick, the device is credited with inspiring Charles Babbage (the father of the computer), who played and lost to the automaton in 1820, with the idea that a mechanical engine could be programed to perform tasks... and the rest is computing history, right up to IBM's Deep Blue. There's an article by the author at Wired, and the preface and first chapter of the book The Mechanical Turk available online."

gain computers, lose clockwork

From the first link .. "Kempelen's contraption was, of course, a hoax. It would have been impossible to build a genuine mechanical chess player using 18th-century clockwork technology."

What is sad to me, is that with the progression of 20th-century computers, and digital watches where even an analouge-faced watch is controlled by quartz crystal and battery, it seems as though the *art* of clockwork has been forgotton....

Just knowing it's possible (even when it isn't)

We once had a customer ask for a software feature that looked virtually impossible to implement, but the customer claimed that our competitor's product had the feature and that they would buy our product if we added this feature to it. So we figured it couldn't be that hard then, and we managed to add the feature with a couple days effort.

Of course it later turned out that the competing product did not have this feature and in fact nobody had ever done it before.

G.

Modern comp required to beat human in chess?

You gotta be kidding! My old 486 always beats me, and that damned thing is generally slower than a dead rock!

A picture of the machine:

http://web.media.mit.edu/~wsack/CAA/chess-machine. html

Ebook heads-up

Here's a free ebook [chesscentral.com] on Maelzel's Chess Player, written by Edgar Allan Poe. It looks pretty good [chesscentral.com].

any sufficiently short midget...

what's that expression? any sufficiently short midget is indistinguishable from magic!

Deep Blue is not the End

The end-all be-all of chess was not embodied in any creation by IBM, that's for sure. Computer-chess history did not end with Deep Blue, and is still alive and well on the ICC and freechess. The software that is being developed right now is A LOT better than anything the Deep Blue team ever came up with, and I have a feeling that if IBM hadn't pulled the plug on Deep Blue it would have probably lost its next match. But don't take my word for it, already Chess software is approaching the strength of Deep Blue by using hardware 1/100th as powerful. I'm sure that in 5-10 years the best machines will regularly beat the world champions on normal PCs.

If you ever get across to London...

...you must go and see the working model of Babbage's difference engine #2 at the Science Museum [sciencemuseum.org.uk]. It was completed in 1991 by the staff using Babbage's drawings and worked first time.

Steam Man

When Professor Campion unveiled Boilerplate in 1893, the concept of a mechanical man was not a new one. Edward S. Ellis, in 1865, wrote about a prodigy that constructed a non-sentient automaton called the Steam Man. At the time, it was considered to be nothing more than an elaborate novelty item, like Boilerplate. Stories of its feats were relegated to the tabloids and "Edisonades." In the account entitled Steam Man of the Prairies (the first of several such publications), Johnny Brainerd, a teenage dwarf, invented "a man that shall go by steam." Here is how it was described: This is a later, cruder version [bigredhair.com]

The Turk is no different than chess programs today

How is The Turk different than modern chess programs today?

Even the best chess programs (Big Blue, etc.) today require the input of humans. They are given instructions, and apply those instructions in a "brute force" fashion to all data in its parameters. The vast majority of the calculations that a computer is asked to make is pure bullshit.

Human intelligence will always have the distinct advantage of eliminating a lot of worthless calculations.

Some extra pictures to illustrate the story

An excellent story, but a little bereft of graphics. Here are some extra pictures to flush out the idea of the device:

• A photo of reconstruction of the device [cowderoy.com]
  
• An exhibition photo [cowderoy.com]
  
• Sketch of the device with open doors [thocp.net]
  

Good read...

Very interesting article... however I find it unfortunate that we don't know how he pulled the hoax off. Based on what I know about automata, it may be very possible to build a chess playing machine. However doing this a hundred+ years ago? I doubt it mostly due to the fact that creating the gears and other mechanisms needed required an amazing amount of time, skill and perfection. In fact this is why I heard Babbage's machine didn't work and the project fell through. I believe someone recently (if someone can find a article for this) built babbage's machine using the old blue-prints and it worked. Another thing is, if this is a hoax I wonder who was the playing the chess. The article definitly points out that the machine was very good at what it did. They only mention one case of it being beaten (along with the napoleon incident), which would mean whoever was playing was damn good. If someone was that good, why would they hide behind the guise of a machine and not reap the benefits of being one of the best chess players in the world? Oh well, definitly a good read though.
Oh one more thing, the duck? They mention that it could take food out of a hand... how the hell did it do this? The last time I checked, motion sensors, digital cameras and such hadn't been invented yet. How the hell did the thing see where it was going, and have the ability to interact with a specific location?

Re:Good read...

Even if you're REALLY good at something, if you play against someone whom you consider or assume to be considerably inferior to you, you will tend to unconsiously dumb down your strategy, and then even if you're a grandmaster, anyone of relatively decent skill will be able to beat you.

The people playing the Turk weren't really playing to win. They were playing to see if this machine could play the game. They were too amazed by its ability to play AT ALL to bother much with trying to beat it. They might even intentionally make stupid "I wonder if he'll catch this" mistakes which ultimately sacrifice the game for them, no matter HOW good they might be.

Probably the only time it got beat was the one time that someone actually paid attention to the game itself, rather than the opposing player.

-Restil

For the lazy

For those of you too lazy to read the entire article:

"The chess player had indeed been controlled by a concealed operator using a clever system of folding partitions to remain hidden while the automaton's interior was open to view."

Another story about this

James Randi did a nice write up about this, with some great pictures and commentary about the machine on his site [randi.org]. You can find a direct link to the articles here [randi.org] and here [randi.org]. I especially enjoyed the artwork depicting how the person inside fit in the contraption and enabled it to play chess. This was a very, very clever little hoax!

Impossible?

"Kempelen's contraption was, of course, a hoax. It would have been impossible to build a genuine mechanical chess player using 18th-century clockwork technology."

Don't sell old technology too short. While a fully playing chess computer was beyond their reach, there were genuine automata in the 18th and early 19th century that could play end-games mechanically. Another examples of amizingly advanced automaton is the Swiss scribe, which can be programed to write a persons name with a quilt in long-hand, including pausing to dip the quilt in the ink well.

That would still be a challenging task for a robotic arm today.

Lastly the entire mechanism that allowed the chessmen to be grasped by a person from inside the Tuks was not replicated until a few decades back, again by "advanced" robotic research.

Ugh

So when's someone going to make an automaton that whacks people every time they build a website [theturkbook.com] with a background that makes it totally unreadable?

I read about it

This is not the only chess playing machine.. There were others, most of them containing a chess playing midget. But then some man got really frustrated when he lost the game and shot the machine with his gun..

Othello/Reversi

A Multi-Gigahertz processor.
1.5 GB of DDR RAM.
Dual 100 GB hard drives.
A half dozen fans to cool the whole thing.

AND I CAN'T FIND A GAME OF OTHELLO / REVERSI THAT CAN CONSISTENTLY BEAT ME.

Alan Turing

Turing talked a lot about the Babbage Engine in his famous essay "Can Machines Think?" While that fact has very little bearing upon the article, Turing's essay touches upon the meaning of what it means to be human and whether it can be replicated. The Babbage Engine was his way of disproving that electricity is what makes humans human. Effectively it also banished the notion that it is any physical or quantifiable thing that makes humans human.

Other purposes

Those machines were not built just to get rich:

In 1879 Mephisto (Gunsberg) went on tour, defeating every male player. However, when playing ladies, it would obtain a winning position, then lose the game, offering to shake hands afterwards

.. but also to get chicks!

does anyone else find it ironic

...that most of the first public appearances of computing technology appear to have been rigged demos?

It seems like some things never change.

but Charles Babbage is NOT the father of computing

We Americans would love to convince ourselves that we, rather Charles Babbage, invented the computer. The British have Allan Turing, and a Postal Inspector for their first computer, or so they like to think. However, the fact is that the first computer was invented by Konrad Zuse [about.com] (1910-1995) at the age of 28 (1938). Konrad was unfortunately living under a Nazi Dictatorship at the time. Turing was brilliant, and Zuse probably didn't hold a candle to Turing. However, I have to step in and make sure the bogus headline here on Slashdot does not perpetuate the silly myth. Konrad Zuse is the father of computing!

Speech synthesis

According to Randi's description:

"A small bellows and vibrating reed, a sort of artificial speaking mechanism, was incorporated whereby the operator could signal "check!" by forcing air through a tube. The approximation of the word "check" was said to lack clarity..."

Randi says this was an improvement made by Maelzel, who bought the machine after von Kempelen's death, but I think this idea, too, came from Kempelen's work, who spent his last years researching speech synthesis. Quiet successfully as he actually did build [ling.su.se] a speech synthetizer capable producing whole words and short sentences. And this machine was not a trick: it is exhibited in the Deutches Museum in Munich, and, according to the author of the link I mentioned, still functional.

and what about Blaise Pascal ?

an almost forgotten programming language
bears his name, because he was the one,
about 1660, to build the first adding
and multiplying machine....Babbage
was surely aware of his work !

Re:I read the Wired article

Yes, it was in Wired magazine. I can't remember if Wired put it online (Web version) or not.

Re:How is this news?

Charles Babbage, the pioneer of the mechanical computer, was another famous opponent; he lost two games to the Turk. Babbage was certain it was under human control, though he was not sure how. But he started to wonder whether a genuine chess-playing machine could, in fact, be constructed.

That is why it's here, not so much news but definately of interest to the slashdot computing crowd.

editorial quibble

I've found since then that shit nearly almost does.

Shouldn't that be "nearly always does"?

Re:I read the Wired article

The book's just been published here in the UK, and the weekend's papers have got reviews - including one that makes the same point (sort of) about Deep Blue. There's a good review here [guardian.co.uk] by Simon Singh, the guy that wrote Fermat's Last Theorem; he mentions that Edmund Cartwright set about building the first power weaving loom after seeing the Turk, reasoning that if a machine could play chess it must be possible to build one that could weave, and so contributing to the start of the industrial revolution.

BTW, the author of the Mechanical Turk is the technology correspondent of The Economist magazine, I see from his website.