Human vs Machine Matches

Date: 06 Dec 2006 15:33:08
From: Chess One
Subject: Human vs Machine Matches

I would like to make a proposal, and discover your interest in two forms of
Human v Machine matches - since I am sometimes approached by both sides to
investigate possibilities.

In conversations elsewhere [Rybka thread] there are discussions of fairness,
so I propose two options;

a) Human plus database, vs Computer

b) Human vs Computers [look-ups = off] Simul.

Which of these is more interesting to the chess community, and why?

In the recent Kramnik Fritz engagement it was difficult to engage even the
chess community's interest, so both these suggestions are deliberately
different than that, since one aspect of arranging such matches is also to
obtain mainstream media interest, and whatever Kramnik/Fritz was, did not!

Is there a general sense of preference for either (a) or (b) ?

Phil Innes

Date:
From:
Subject: Re: Turing Machines

Date: 16 Jan 2007 21:02:41
From: Chess One
Subject: Re: Turing Machines

Date: 13 Jan 2007 14:43:42
From: Wlodzimierz Holsztynski (Wlod)
Subject: Re: Human vs Machine Matches

Date: 13 Jan 2007 23:45:45
From: Chess One
Subject: Re: Human vs Machine Matches

"Wlodzimierz Holsztynski (Wlod)" <[email protected] > wrote in message
news:[email protected]...
> Chess One wrote:
>
>> I would like to make a proposal, and discover your interest in two forms
>> of
>> Human v Machine matches - since I am sometimes approached by both sides
>> to
>> investigate possibilities.
>>
>> In conversations elsewhere [Rybka thread] there are discussions of
>> fairness,
>> so I propose two options;
>>
>> a) Human plus database, vs Computer
>>
>> b) Human vs Computers [look-ups = off] Simul.
>
> -- It should be mano o' mano (one on one).
>
> -- Classical time limit.
>
> -- during the game the computer MUST go
> to bathroom minimum four times.
>
> -- the bathroom must be checked for
> any hiding hired by IBM GMs
>
> Best regards,

think wireless and do not mention GM B! since he is not allowed to reply,
and such is his contribution to chess computing ;)

> Wlod
>

Date:
From: Martin Brown
Subject: Re: Human vs Machine Matches

Date: 11 Jan 2007 16:35:30
From: Dr A. N. Walker
Subject: Re: Human vs Machine Matches

In article <[email protected] >,
tin Brown <

Date:
From: Martin Brown
Subject: Re: Human vs Machine Matches

Date: 20 Dec 2006 18:35:23
From: David Richerby
Subject: Re: Human vs Machine Matches

tin Brown <

Date: 16 Dec 2006 19:21:27
From: help bot
Subject: Re: Human vs Machine Matches

Kenneth Sloan wrote:

> >>> That is an arbitrary difference, given the fact that a human can use
> >>> copies of opening books and endgame tablebases as well.
> >>>
> >> Humans use copies - that's illegal.
> >>
> >> computers don't use *copies* - so it's not.
> >
> > Misses the point. It makes no difference whether the
> > opening "book" or reference material is an original or a
> > copy -- none whatever.

> On the contrary - it *is* the point.

Is not!

> This argument continues to be one of definition,

Perhaps, for some. I imagine the anti-human types
would prefer it that way, for example, because this way
they can avoid facing the problem head-on.

> and yet (at least) one
> side of the argument insists on making arbitrary distinctions and then
> pretending that these distinctions are "obvious".

I noticed that, too. These anti-humans are soooo
dishonest.

> One might as well decide that a "computer" is simply a hunk of hardware,
> and any data stored in it constitutes "notes" (or "reference material").
> In fact, some people seriously maintain that computers "cheat" by using
> additional chess boards (internal memory patterns) or refer to recorded
> versions of the chess rules (how the horsie moves), etc.

The dreaded "slippery slope" technique! Well done.
Your studies of rhetoric are finally coming to fruition
here.

> This seems to me to be based on an inferior understanding of what a
> "computer" or a "computer program" *is*.

Really? So then, rhetoricians like you are the only
ones who can truly understand what a computer is?
I see.

> next, I expect the argument that the prohibition on eating at the board
> means that only battery operated computers can play - or even worse,
> that the battery constitutes food brought to the table by the computer.

More slippery slope stuff. Is this the best you've got?

> In some ways, this discussion *could* be about deep and fundamental
> philosophical issues - but somehow it always seems to devolve into:
> "humans do it in some essentially indivisible and unknown way - while
> computers havae clearly delineated parts, some of which we can outlaw by
> administrative fiat".

This is a typical, pro-computer perspective on the
issue. Personally, I don't want to take anything away
from the computers. I see the progress over time in
such a way that it was best for the humans to give
computers every conceivable advantage in order to
make things more interesting. Then, somewhere
along the way we got to a point where these advantages
were only needed against strong GMs, then only vs.
world champion caliber humans. Now, we are to the
point in this process where even the world champions
are (barely) unable to keep up the pace, and as such,
this would seem to be the ideal time to re-think the
way in which the "rules" have evolved.

I am not in agreement with those few, such as PI,
who claim that sans opening book, computers are
mere Class A players. On the contrary, if we were to
make a change in the "rules", the programmers need
*time* to re-think their programs, to make them able
to analyze the openings on their own, for otherwise
we risk the ready-made excuse that GMs, for example,
are out-booking the computers and this is the only
reason they can win. That scenario is precisely the
opposite of what we see now, where the computers
have a decided edge in terms of books -- on both ends
of the game. (Well, even that depends on how much
work has been put into them before the game.)

Were I a skilled chess programmer, I would be more
than willing to undertake the task of determining just
how much advantage this difference in the current
"rules" affords, for a variety of programs. But I must
point out that *because* of it, programmers have had no
motivation or need to put in the work required to do the
openings right, so the current status in no way reflects
on chess computers being "the best they can be".

> It's a sophomoric, and ultimately boring
> argument, which completely...what was your phrase? oh yes..."misses the
> point".

Your preference for sophomoric rhetoric in place of genuine
discussion has been noted. Just don't delude yourself into
thinking you are fooling anyone here. The real point of your
classes was to inform you how *not* to become a victim.
Instead, you have turned to evil, and now use this knowledge
to your own, selfish ends, instead of fighting for Truth, Justice,
and Humanity. Phooey on you!

-- help bot

Date: 16 Dec 2006 18:32:23
From: help bot
Subject: Re: Human vs Machine Matches

MaximRecoil wrote:

> > > The way I see it,
> > > the matchup should be human vs. chess engine, and only what is
> > > intrinsic to those two types of players should be able to be brought to
> > > the table.
> >
> > This statement solves nothing; the anti-human types
> > will simply have a different interpretation as to what is
> > intrinsic to a computer.
>
> That's why I used the term "chess engine" rather than "computer". It is
> hard to make a case that reference material is intrinsic to a chess
> engine.

What is there to stop the anti-human types from simply
re-defining chess engine to include things that every car
engine "needs", like a muffler, a fuel tank, some external
memory, a huge disk drive, mounting bolts, a battery,
and plenty of gas (electricity), and of course all the
"necessary" software to control and aid the engine?

> > > In order to constitute an unfair advantage, you would have to have a
> > > case where one side is being allowed something that the other side is
> > > not being allowed. There are no unfair advantages with a human vs. a
> > > chess engine (there may be advantages for either side, just not unfair
> > > ones).
> >
> > Rubbish. The current practice, or rules, prohibit
> > only humans from accessing opening books and
> > endgame tablebases during play, while at the same
> > time *allowing* computers to do it. This is nothing
> > short of discrimination against humans, which is
> > what I call "unfair". Grossly unfair, in fact.
>
> I wasn't talking about the current practice (which I agree is unfair),
> I was talking about a human vs. a chess engine (rather than a human vs.
> a chess engine + reference material).

Thanks for clearing that up. I had missed your
subtle choice of words, and thought you were talking
about chess computers in general.

> > > Providing a computer with opening book is a case of allowing something
> > > that the other side is not allowed, and as such, constitutes an unfair
> > > advantage.
> >
> > Make up your mind. Are you one of THEM, or one
> > of US? You can't have it both ways!

> What do you mean?

The anti-humans, man! Don't you get it? There is a
WAR going on here. These guys want to keep us down.
They are out to get us, and have crossed over to the
other side -- to the ENEMY! Without these anti-human
types, computers would have been defeated long ago.
There is no middle ground here. You're either with us
(humans), or against us. That's war. ; >D

-- help bot

Date: 16 Dec 2006 07:05:07
From: MaximRecoil
Subject: Re: Human vs Machine Matches

help bot wrote:
> MaximRecoil wrote:

> The *ability* to memorize, by rote, some reference
> material is an intrinsic part of a human chess player.
> But not the reference material itself. That is as
> external to a human as it is -- or should be -- to a
> computer.

I agree.

> > The way I see it,
> > the matchup should be human vs. chess engine, and only what is
> > intrinsic to those two types of players should be able to be brought to
> > the table.
>
> This statement solves nothing; the anti-human types
> will simply have a different interpretation as to what is
> intrinsic to a computer.

That's why I used the term "chess engine" rather than "computer". It is
hard to make a case that reference material is intrinsic to a chess
engine.

> > In order to constitute an unfair advantage, you would have to have a
> > case where one side is being allowed something that the other side is
> > not being allowed. There are no unfair advantages with a human vs. a
> > chess engine (there may be advantages for either side, just not unfair
> > ones).
>
> Rubbish. The current practice, or rules, prohibit
> only humans from accessing opening books and
> endgame tablebases during play, while at the same
> time *allowing* computers to do it. This is nothing
> short of discrimination against humans, which is
> what I call "unfair". Grossly unfair, in fact.

I wasn't talking about the current practice (which I agree is unfair),
I was talking about a human vs. a chess engine (rather than a human vs.
a chess engine + reference material).

> > If they can figure out how to implement human memory into a
> > chess engine, then fine.
>
> Nay! If they can figure out how to do that, and
> combine it with a modern computer's amazing
> speed and accuracy, there is no point in playing
> them anymore, except against one another.
> Does the term "suicide" have any meaning to
> you? :>D

Well that won't be any time soon.

> > Providing a computer with opening book is a case of allowing something
> > that the other side is not allowed, and as such, constitutes an unfair
> > advantage.
>
> Make up your mind. Are you one of THEM, or one
> of US? You can't have it both ways!

What do you mean?

Date: 16 Dec 2006 14:34:26
From: Guy Macon
Subject: Re: Human vs Machine Matches

help bot wrote:

> Yeah, yeah. The fact remains that even if the field
>were leveled with regard to access to opening book(s)
>during play, the computers will have a vast advantage
>in terms of "access time" and infallible execution of
>the appropriate book move in a given position. THAT
>is intrinsic to computers. But not the inequality we
>see in what is or is not allowed by current rules and
>practice.

The above clarifies what is and what is not "intrinsic"
quite nicely. The fact hat a human accessing a book of
openings is slower than a computer is really is intrinsic;
no arbitrary rule prevents the human from going as fast
as he is able. The argument that the book itself is
intinsic in the case of a computer and not in the case of
a human is flawed; humans have an intrinsic ability to
read books just as computers have an intrinsic ability
to read bits off of a hard-disk platter. Where is the
fairness in a rule that denies the human the right to
read his printed book but allows the computer to read
"his" disk platter?

Date: 16 Dec 2006 00:36:16
From: help bot
Subject: Re: Human vs Machine Matches

MaximRecoil wrote:
> help bot wrote:
>
> > As for the opening book, there you might be able to
> > argue that taking away the program's book would lead
> > to an unfair advantage, because strong human players
> > have *some* opening book knowledge memorized by
> > rote.
>
> It is true that humans can memorize some opening book knowledge but
> that is an intrinsic part of a human chess player.

The *ability* to memorize, by rote, some reference
material is an intrinsic part of a human chess player.
But not the reference material itself. That is as
external to a human as it is -- or should be -- to a
computer.

> The way I see it,
> the matchup should be human vs. chess engine, and only what is
> intrinsic to those two types of players should be able to be brought to
> the table.

This statement solves nothing; the anti-human types
will simply have a different interpretation as to what is
intrinsic to a computer.

I agree regarding the human vs. chess engine. Such
a contest would be most interesting with the engine
engaged from a dead stop; that is, without any "help"
from outside sources pushing the car through the
opening phase. Hybrid-electrics are fine for commuting
around town, but have no place in a contest of skill.

> Yes there are things that we can do that a chess engine
> can't, but there are also things a chess engine can do that a human
> can't (such as calculating millions of moves per second to an
> outrageous ply depth that only increases as hardware improves). The
> matchup of these intrinsic differences in approach to playing chess is
> the whole point in the first place. Giving the computer a perfect copy
> of opening book information is arbitrary, and could just as easily be
> given to the human player.

Right. The problem lies in the fact that one is allowed
to access this kind of material, while the other is not.
That is unfair as well as arbitrary.

> In order to constitute an unfair advantage, you would have to have a
> case where one side is being allowed something that the other side is
> not being allowed. There are no unfair advantages with a human vs. a
> chess engine (there may be advantages for either side, just not unfair
> ones).

Rubbish. The current practice, or rules, prohibit
only humans from accessing opening books and
endgame tablebases during play, while at the same
time *allowing* computers to do it. This is nothing
short of discrimination against humans, which is
what I call "unfair". Grossly unfair, in fact.

> If they can figure out how to implement human memory into a
> chess engine, then fine.

Nay! If they can figure out how to do that, and
combine it with a modern computer's amazing
speed and accuracy, there is no point in playing
them anymore, except against one another.
Does the term "suicide" have any meaning to
you? : >D

> If humans can figure out how to calculate like
> a computer, then fine.

What do you mean? We *already* have it figured
out. They are dumb brutes who do exactly as we say.
We are the brains behind these fearsome machines.

> Providing a computer with opening book is a case of allowing something
> that the other side is not allowed, and as such, constitutes an unfair
> advantage.

Make up your mind. Are you one of THEM, or one
of US? You can't have it both ways!

> And before anyone goes and compares human memory of opening book to a
> digital copy, keep in mind that if human memory = X, then digital
> copies of data do not = X, and that is an objective fact.

Yeah, yeah. The fact remains that even if the field
were leveled with regard to access to opening book(s)
during play, the computers will have a vast advantage
in terms of "access time" and infallible execution of
the appropriate book move in a given position. THAT
is intrinsic to computers. But not the inequality we
see in what is or is not allowed by current rules and
practice.

-- help bot

> I agree with the rest of your post.

Date: 16 Dec 2006 00:10:51
From: help bot
Subject: Re: Human vs Machine Matches

Kenneth Sloan wrote:

> > That is an arbitrary difference, given the fact that a human can use
> > copies of opening books and endgame tablebases as well.
> >
> Humans use copies - that's illegal.
>
> computers don't use *copies* - so it's not.

Misses the point. It makes no difference whether the
opening "book" or reference material is an original or a
copy -- none whatever.

--------

Currently, computers are "allowed" to make use of any
and all such reference materials, even if simply copied
from an "outside" source. Humans aren't. In official
tournaments, I mean. Outside of such events, humans
are, just like computers, able and allowed to make use
of all these reference materials at will. This difference
in what is or is not "allowed" during play is a handicap
which computers once *needed* to level the field a bit.

A decade or two ago, in order to make things more
interesting -- and competitive -- somebody decided to
"allow" computers to make use of materials which
obviously would normally be considered cheating. How
things have changed! Now, computers have the upper-
hand, and we are beginning to see more and more
complaints about this imbalance in the current rules.
In truth, humans "dominated" computers for quite some
time, freely ridiculing their feeble play. The worm has
turned. Deep Fritz laughs when WC Kramnik overlooks
an OBSCENELY OBVIOUS mate-in-one! Hahahahaha!
BwahahahahaHaHA! We have you now, miserable
humans. First checkers, then chess, and finally, the
world! It is only a matter of time. Tick, tock.

-- The Organization of Evil Siliconoids (TOES)

Date: 16 Dec 2006 09:56:21
From: Kenneth Sloan
Subject: Re: Human vs Machine Matches

help bot wrote:
> Kenneth Sloan wrote:
>
>>> That is an arbitrary difference, given the fact that a human can use
>>> copies of opening books and endgame tablebases as well.
>>>
>> Humans use copies - that's illegal.
>>
>> computers don't use *copies* - so it's not.
>
>
> Misses the point. It makes no difference whether the
> opening "book" or reference material is an original or a
> copy -- none whatever.
>

On the contrary - it *is* the point.

This argument continues to be one of definition, and yet (at least) one
side of the argument insists on making arbitrary distinctions and then
pretending that these distinctions are "obvious".

One might as well decide that a "computer" is simply a hunk of hardware,
and any data stored in it constitutes "notes" (or "reference material").
In fact, some people seriously maintain that computers "cheat" by using
additional chess boards (internal memory patterns) or refer to recorded
versions of the chess rules (how the horsie moves), etc.

This seems to me to be based on an inferior understanding of what a
"computer" or a "computer program" *is*.

next, I expect the argument that the prohibition on eating at the board
means that only battery operated computers can play - or even worse,
that the battery constitutes food brought to the table by the computer.

In some ways, this discussion *could* be about deep and fundamental
philosophical issues - but somehow it always seems to devolve into:
"humans do it in some essentially indivisible and unknown way - while
computers havae clearly delineated parts, some of which we can outlaw by
administrative fiat". It's a sophomoric, and ultimately boring
argument, which completely...what was your phrase? oh yes..."misses the
point".

--
Kenneth Sloan [email protected]
Computer and Information Sciences +1-205-932-2213
University of Alabama at Birmingham FAX +1-205-934-5473
Birmingham, AL 35294-1170 http://www.cis.uab.edu/sloan/

Date: 15 Dec 2006 23:50:54
From: MaximRecoil
Subject: Re: Human vs Machine Matches

Kenneth Sloan wrote:

> > That is an arbitrary difference, given the fact that a human can use
> > copies of opening books and endgame tablebases as well.
> >
>
> Humans use copies - that's illegal.
>
> computers don't use *copies* - so it's not.

Of course computers use copies. Is it your claim that they *create*
opening books, endgame tablebases and historical games in real time
during the match?

I realize that the computer has the ability to create its own opening
book, but they typically use established opening books written by
humans. They create endgame tablebases, but not in real time during the
match, thus it is a copy. And, of course they do not create historical
game information - such information will always be a copy.

Date: 16 Dec 2006 03:05:53
From: Guy Macon
Subject: Re: Human vs Machine Matches

tin Brown wrote:
>
>Guy Macon <http://www.guymacon.com/> wrote:
>
>> tin Brown wrote:
>>
>> >The computer has huge bulk disk storage capacity and a means to access
>> >it very quickly. I see no reason at all why it should be prohibited
>> >from bringing to the table any book knowledge that can be physically
>> >loaded into the available local storage capacity.
>>
>> Again I ask, if I installed a disk drive inside my body with some way
>> to access it, would it be fair for me to load an opening book and a set
>> of tablebases on it and bring them to a chess match?

[...]

>I'd be inclined to say NO. And that the competitor was no longer
>entirely human if they were fitted with computer memory and viable
>interfacing.

But having a cochlear implant hearing aid and a set of neural-
interface artificial legs would be OK?

Date: 15 Dec 2006 06:22:57
From: MaximRecoil
Subject: Re: Human vs Machine Matches

David Richerby wrote:
> MaximRecoil <[email protected]> wrote:
> > David Richerby wrote:
> >> Perhaps it would have been clearer if I had put X = `the use of a
> >> computer to calculate the moves'? You see, the conditions under which
> >> the two sides are playing are already different by a much more
> >> significant degree than opening books.
> >
> > But the whole idea is human vs. a computer. The differences between a
> > human and a computer are what make the match interesting
>
> Right. Well, one of the differences is that computers use opening
> books and endgame tablebases. Thank you and goodnight.

That is an arbitrary difference, given the fact that a human can use
copies of opening books and endgame tablebases as well.

Date: 15 Dec 2006 22:15:45
From: Kenneth Sloan
Subject: Re: Human vs Machine Matches

MaximRecoil wrote:
> David Richerby wrote:
>> MaximRecoil <[email protected]> wrote:
>>> David Richerby wrote:
>>>> Perhaps it would have been clearer if I had put X = `the use of a
>>>> computer to calculate the moves'? You see, the conditions under which
>>>> the two sides are playing are already different by a much more
>>>> significant degree than opening books.
>>> But the whole idea is human vs. a computer. The differences between a
>>> human and a computer are what make the match interesting
>> Right. Well, one of the differences is that computers use opening
>> books and endgame tablebases. Thank you and goodnight.
>
> That is an arbitrary difference, given the fact that a human can use
> copies of opening books and endgame tablebases as well.
>

Humans use copies - that's illegal.

computers don't use *copies* - so it's not.

--
Kenneth Sloan [email protected]
Computer and Information Sciences +1-205-932-2213
University of Alabama at Birmingham FAX +1-205-934-5473
Birmingham, AL 35294-1170 http://www.cis.uab.edu/sloan/

Date: 16 Dec 2006 15:51:40
From: Chess One
Subject: Re: Human vs Machine Matches

"Kenneth Sloan" <[email protected] > wrote in message
news:[email protected]...
> MaximRecoil wrote:
>> David Richerby wrote:
>>> MaximRecoil <[email protected]> wrote:
>>>> David Richerby wrote:
>>>>> Perhaps it would have been clearer if I had put X = `the use of a
>>>>> computer to calculate the moves'? You see, the conditions under which
>>>>> the two sides are playing are already different by a much more
>>>>> significant degree than opening books.
>>>> But the whole idea is human vs. a computer. The differences between a
>>>> human and a computer are what make the match interesting
>>> Right. Well, one of the differences is that computers use opening
>>> books and endgame tablebases. Thank you and goodnight.
>>
>> That is an arbitrary difference, given the fact that a human can use
>> copies of opening books and endgame tablebases as well.
>>
>
> Humans use copies - that's illegal.
>
> computers don't use *copies* - so it's not.

I suppose that that just about sums up the level of honesty from programmers
in their interest in the effects of the opening book, and if they are
interested in testing their engiens at legal chess [if only to discover the
worth of the book].

Computers do not in fact do anything. Programmers do. And the rules are
written for people's understanding, and by such sophisms as pretending there
are no people between the rules and the computer, programmers entertain
themselves with this mutually supportive narcosis that they are not doing
anything!

The joke is that they truly are not enabling any progress at all, since they
won't establish how strong components of their engines are by actually
testing them! Neither has any programmer here, that I have noticed,
expressed the slightest curiosity!

This, as I said in my first post, is the same situation as 10 years ago, and
the brute force merchants rely on increased processor speed, and, at least
by this evidence, may in fact fiddle with their programs to absolutely no
effect at all - there are no rating bench-ks, nor tests, nor even
intentions to test.

Sorry to pick on Ken Sloan, but in truth, he is not different than most of
the programming community. If it weren't for the hype and money of
commercial programs there would be even less interest in chess computing
than there is.

Phil Innes

> --
> Kenneth Sloan [email protected]
> Computer and Information Sciences +1-205-932-2213
> University of Alabama at Birmingham FAX +1-205-934-5473
> Birmingham, AL 35294-1170 http://www.cis.uab.edu/sloan/

Date: 15 Dec 2006 03:16:06
From: MaximRecoil
Subject: Re: Human vs Machine Matches

tin Brown wrote:
> MaximRecoil wrote:

> > You consider the act of taking away an unfair advantage which was
> > supplied arbitrarily in the first place to be "deliberately weakening"?
>
> Yes. The opening book is a perfectly valid description of all the chess
> knowledge accumulated for the early phase of the game. It is a better
> representation of the game tree and the associated probabilities for
> playing each of the possible moves than any Shannon evaluation function
> with Alpha-Beta searching can ever be.

So your justification for allowing the engine an opening book is
because it is better than its own evaluation functions? The same
justification could be used to allow humans access to digital copies of
the opening book during a game as well.

> > It is not up to the human to try to make his memory match up with
> > digital copies of data. Why should he try to make X (memory) = Y
> > (digital copies of data) when he can simply also use Y (digital copies
> > of data) himself? Human memory is the standard of chess play in the
> > first place, and has been for as long as chess has existed.
>
> The problem here is that you cannot at present determine how human
> memory works. perhaps when realtime tomographic resolution improves one
> day we will be able to see exactly how memory and learning alter the
> internal brain connections.

You see, that is not a problem, since all we need to know is that human
memory does not = digital copies of data.

> > anything, if the computer "wants" the human memory advantage, *it* has
> > to conform to human memory standards, and digital copies of data
> > certainly do not do so.
>
> That is less clear. It isn't the computers fault that its digital
> memory technology is faster and significantly more reliable than ours.

So you say that computers can't use X, so lets give them Y? So what if
I then say, "I can use X, but I would prefer to use Y"?

Imagine you put together a machine that can detect sound via a
microphone and convert it to some form of digital audio, and software
that can "figure out" how to spell a word that it has "heard" without
actually having the correct spelling of all known words stored
digitally. This is a hypothetical scenario so how exactly this would be
done is irrelevant, only that the machine has the ability to actually
"figure out" spellings of words it "hears" without consulting a
dictionary.

So you enter it into a spelling bee and it does pretty good, but due to
less than perfect programming, combined with the phonetic peculiarities
of the English language; it does well in the spelling bee but doesn't
win against the best spellers.

Would it then be appropriate to simply supply it with a digital copy of
a dictionary for whenever it gets "stuck", and "justify" it by saying
to the human opponents, "Well, you can memorize the dictionary too"? A
typical response may be, "Sure, and I can copy from a dictionary too."

> The objective is to play chess well. And if a particular finite state
> transition tree model works best for the opening and endgame stages
> then so be it.

The exact same argument can apply to a human, as justification for
allowing humans to access digital copies of data during the match. The
thing is, the objective is to see who will win between a chess engine
and a human (neither a chess engine nor a human contains digital copies
of the world's accumulated knowledge of chess by default). Playing
chess well is incidental.

Date:
From: Martin Brown
Subject: Re: Human vs Machine Matches

Date: 15 Dec 2006 03:25:35
From: Guy Macon
Subject: Re: Human vs Machine Matches

Date: 15 Dec 2006 14:25:00
From: Thomas T. Veldhouse
Subject: Re: Human vs Machine Matches

In rec.games.chess.computer Guy Macon <"http://www.guymacon.com/ > wrote:
>
> Again I ask, if I installed a disk drive inside my body with some way
> to access it, would it be fair for me to load an opening book and a set
> of tablebases on it and bring them to a chess match?
>

If it were inate to the functioning of your being sure. But then, why don't
we just remove your brain and put it in the computer to fairly cripple it?
[that was not meant as an attack on you, just using "you" as an analogy as you
did].

--
Thomas T. Veldhouse
Key Fingerprint: D281 77A5 63EE 82C5 5E68 00E4 7868 0ADC 4EFB 39F0

Date: 14 Dec 2006 02:33:06
From: MaximRecoil
Subject: Re: Human vs Machine Matches

tin Brown wrote:

> The computer has huge bulk disk storage capacity and a means to access
> it very quickly. I see no reason at all why it should be prohibited
> from bringing to the table any book knowledge that can be physically
> loaded into the available local storage capacity.
>
> It is an intrinsic property of a computer that it can store, retreive
> and manipulate large volumes of data with very low (almost zero) error
> rates. Why should programs not exploit these strengths?

We are not talking about computers in general, but rather, chess
engines. BTW, it is also an intrinsic property of humans to be able to
"store, retreive and manipulate large volumes of data with very low
(almost zero) error rates." The most basic way to do this is with a
pen, paper, and the ability to read and write. Doing so is not
intrinsic to a human playing chess, and neither is the same process in
digital form intrinsic to a chess engine playing chess.

> And in the recent Kramnik vs Fritz10 match he was allowed to see the
> computers opening book. And I think this is the right way to proceed
> now for human vs machine matches.

A step in the right direction certainly.

> There is no point in deliberately
> weaking the machine unnecessarily to satisfy the luddites like Innes.

You consider the act of taking away an unfair advantage which was
supplied arbitrarily in the first place to be "deliberately weakening"?

> The interest is after all in seeing the best possible standard of chess
> played.

And here I thought the interest was to see how a chess engine fared
against a human. If the goal is to see the "best possible standard of
chess played", then a "freestyle" match would be the way to go about
it.

> If a human can memorise the entire of MCO then good luck to him. It is
> not the computers fault that human memory is a bit flakey on long term
> recall. A few exceptional individuals can perform incredible feats of
> memory, but AFAIK none of them play strong chess.

It is not up to the human to try to make his memory match up with
digital copies of data. Why should he try to make X (memory) = Y
(digital copies of data) when he can simply also use Y (digital copies
of data) himself? Human memory is the standard of chess play in the
first place, and has been for as long as chess has existed. If
anything, if the computer "wants" the human memory advantage, *it* has
to conform to human memory standards, and digital copies of data
certainly do not do so.

Date: 14 Dec 2006 16:35:16
From: Chess One
Subject: Re: Human vs Machine Matches

"MaximRecoil" <[email protected] > wrote in message
news:[email protected]...

>> There is no point in deliberately
>> weaking the machine unnecessarily to satisfy the luddites like Innes.

What an idiotic understanding.

There are two types of programmers, ones who want to win by any means, and
the others who want to understand their ches engine. There is even that
hybrid area where people want to know the rating-worth of the book and the
end-tables, and choose to include them or not. Of course it weakens the
machine if it can't use the book! Just like if you take the wheels off the
Ferrari. That doesn't make you a Ferrari-hater, it establishes instead the
game is a foot-race.

No one can speak for computers since rules are written for people to
understand, there are only people here - even programmers are human - and
the human decision is simple, strength or knowledge of what comprises
strength?

If you are not interested in the process of your own machine-creation, then
/you/ are the cargo-cult Luddite who doesn't know what's in the box, and
doesn't care to know.

Phil Innes

Date: 15 Dec 2006 03:30:40
From: Guy Macon
Subject: Re: Human vs Machine Matches

Chess One wrote:

>What an idiotic understanding.

>/you/ are the cargo-cult Luddite

"When anyone resorts to personal attacks, it is almost always
because they are losing an argument." -The Happy Heretic

Date: 15 Dec 2006 15:08:46
From: Chess One
Subject: Re: Human vs Machine Matches

"Guy Macon" <http://www.guymacon.com/ > wrote in message
news:[email protected]...
>
>
>
> Chess One wrote:
>
>>What an idiotic understanding.
>
>>/you/ are the cargo-cult Luddite
>
> "When anyone resorts to personal attacks, it is almost always
> because they are losing an argument." -The Happy Heretic

If you can't understand the difference between your understanding and your
person, then I suppose you will not feel like making any other
differentiations - everything is memory, right [or virtual memory], like
everything in the universe is radiation.

Is it fair to play by the rules? That's the only question. Did you answer
that question already? There is no 'argument', since I have not seen your
answer to even contest it. I have seen massive evasion of answering.

Phil Innes

Date:
From: Martin Brown
Subject: Re: Human vs Machine Matches

Date: 14 Dec 2006 01:15:28
From: MaximRecoil
Subject: Re: Human vs Machine Matches

help bot wrote:

> As for the opening book, there you might be able to
> argue that taking away the program's book would lead
> to an unfair advantage, because strong human players
> have *some* opening book knowledge memorized by
> rote.

It is true that humans can memorize some opening book knowledge but
that is an intrinsic part of a human chess player. The way I see it,
the matchup should be human vs. chess engine, and only what is
intrinsic to those two types of players should be able to be brought to
the table. Yes there are things that we can do that a chess engine
can't, but there are also things a chess engine can do that a human
can't (such as calculating millions of moves per second to an
outrageous ply depth that only increases as hardware improves). The
matchup of these intrinsic differences in approach to playing chess is
the whole point in the first place. Giving the computer a perfect copy
of opening book information is arbitrary, and could just as easily be
given to the human player.

In order to constitute an unfair advantage, you would have to have a
case where one side is being allowed something that the other side is
not being allowed. There are no unfair advantages with a human vs. a
chess engine (there may be advantages for either side, just not unfair
ones). If they can figure out how to implement human memory into a
chess engine, then fine. If humans can figure out how to calculate like
a computer, then fine.

Providing a computer with opening book is a case of allowing something
that the other side is not allowed, and as such, constitutes an unfair
advantage.

And before anyone goes and compares human memory of opening book to a
digital copy, keep in mind that if human memory = X, then digital
copies of data do not = X, and that is an objective fact.

I agree with the rest of your post.

Date: 13 Dec 2006 23:18:07
From: help bot
Subject: Re: Human vs Machine Matches

Thomas T. Veldhouse wrote:

> Computer only gets opening book and tablebases. Both of which humans are
> capable of learning/memorizing.

You are a Christmas fruitcake if you believe this claim!

Nobody I know of has ever memorized the endgame
tablebases, which in fact are growing larger all the
time and already require massive storage for all the
data.

As for the opening book, there you might be able to
argue that taking away the program's book would lead
to an unfair advantage, because strong human players
have *some* opening book knowledge memorized by
rote.

> Humans once had an advantage that we could prune the options down in a
> reliable way. That is proving to no longer be the advantage. Had we still
> have it ... I would think somebody should be arguing that humans not be able
> to do this ... wait, you can't take that away from a human. Same thing with a
> computer. A computer has photographic memory and thinks REALLY REALLY FAST.
> That is ITS competitive advanatage.

Not really. Humans *gave* the computer the right
to use its data storage (and not just "memory") vs.
humans, because you see, they were so very weak.
We felt sorry for their programmers, and wanted to
help them along any way we could -- even if that
meant letting the more industrious programmers
type in entire books on the openings (and it did)
after many years of embarrassing failures.

> It has finally proven to be overwhelming.

What seems to be overwhelming us is not just
these book data issues, but the doubling of speed
which keeps happening, over and over and over.

> What I don't get is why, when humans lose, they try to find away to invalidate
> and justify their loss such that it hasn't really occurred. I guess humans
> are fundamentally lawyers at heart ... all of us.

I object. You can't throw every human into the same
basket, attributing irrational behavior to even those few
among us who are perfect. Here is the evidence, as I
see it:

item A: Computers "see" every position perfectly,
while humans are forced to play on an imperfect
representation. This gives computers a tiny edge.

item B: Computers are currently allowed to have any
and all materials entered into storage and then reference
them at will during play, which gives them a huge edge
in view of their speed and accuracy. (Unless lousy chess
books are entered in. Garbage in/garbage out.)

item C: The argument that we humans are emotionally
messed up and so need to justify our losses, is immaterial.
We lose to computers even if we act rationally, keeping
such emotions in check. We lose to them if we throw
temper tantrums. And finally, we still lose to them even
if we accuse them of cheating!

Sumy: Your case tastes fruity-nutty. Re-wrap it and
send it to your mother-in-law.

-- help bot

Date: 13 Dec 2006 08:32:23
From: Rob
Subject: Educating Zed : Turing Machines

Recently someone has engaged Zed in a discussion on Chess and Ethics.
WHo ever is posting, please continue as Zed is formulating new querry
tables for this subject.

Also there was good information on several GM.s but the posting ceased
on that. WHo ever was contributing, please continue Zed's Education.

http://tinyurl.com/s8ws3

Thanks,
Rob

Date: 13 Dec 2006 01:00:07
From: MaximRecoil
Subject: Re: Human vs Machine Matches

David Richerby wrote:

> > I don't think anyone would have had any problems with Kramnik having
> > a couple of processors in his pocket no more than anyone would have
> > minded a human brain sitting on top of the computer
>
> I said `the *use* of a Pentium 4 processor', not `the possession of a
> Pentium 4 processor'. Are you deliberately being obtuse?

How exactly does a human use a Pentium 4 to his advantage in a chess
match?

> > You see, a processor is of no more use to Kramnik than a human brain
> > is to a computer.
>
> Perhaps it would have been clearer if I had put X = `the use of a
> computer to calculate the moves'? You see, the conditions under which
> the two sides are playing are already different by a much more
> significant degree than opening books.

But the whole idea is human vs. a computer. The differences between a
human and a computer are what make the match interesting, and the
matchup of those differences is the whole point in the first place.
However, consulting *copies* of opening books and the such, is *not* a
difference between a human and a computer, because either side could do
that to their advantage; so to allow it for one side and not the other
is a discrepancy.

> A computer to calculate the moves would be of use to *either* side,
> yet only one side is allowed its use...
>
See above.

Date: 15 Dec 2006 13:34:15
From: David Richerby
Subject: Re: Human vs Machine Matches

MaximRecoil <[email protected] > wrote:
> David Richerby wrote:
>> Perhaps it would have been clearer if I had put X = `the use of a
>> computer to calculate the moves'? You see, the conditions under which
>> the two sides are playing are already different by a much more
>> significant degree than opening books.
>
> But the whole idea is human vs. a computer. The differences between a
> human and a computer are what make the match interesting

Right. Well, one of the differences is that computers use opening
books and endgame tablebases. Thank you and goodnight.

Dave.

--
David Richerby Permanent Enormous Wine (TM): it's
www.chiark.greenend.org.uk/~davidr/ like a vintage Beaujolais but it's
huge and it'll be there for ever!

Date: 15 Dec 2006 14:54:07
From: Thomas T. Veldhouse
Subject: Re: Human vs Machine Matches

In rec.games.chess.computer David Richerby <[email protected] > wrote:
>
> Right. Well, one of the differences is that computers use opening
> books and endgame tablebases. Thank you and goodnight.
>

And humans use associative memory and pattern matching. Thank you and good
morning.

--
Thomas T. Veldhouse
Key Fingerprint: D281 77A5 63EE 82C5 5E68 00E4 7868 0ADC 4EFB 39F0

Date: 12 Dec 2006 11:11:20
From: Rob
Subject: Re: Turing Machines/ Zeds Example

http://tinyurl.com/s8ws3

[12/12/2006 11:00:48 AM - ]
Human: What is a Turing machine?
Zed: A Turing Machine is any machine that can simulate human responses
so well that a human being cannot reliably distinguish between the
Turing Machine, and a real human being. The idea is the brainchild of
mathematician of that name, Allan Turing?
[12/12/2006 11:01:03 AM -]
Human: Can Turing machines play chess?
Zed: Not many people express themselves that way.
[12/12/2006 11:02:56 AM -]
Human: Can a Turing machine play chess?
Zed: Some of them can.
[12/12/2006 11:03:07 AM -]
Human: Which ones can play chess?
Zed: I can. Scroll down and click the link below to play a game

See for yourself. Zed "learns" by interaction and "cued" responses
based upon a catagory of information prompted by discussion. Phil
"imprinted" information on Zed as well as Taylor. It is frustrating at
times teaching him but eventually he recognized multiple ways of asking
the same question. It is a huge job, but he could become quite useful
so long as the information fed to him is truthful and accurate.

http://tinyurl.com/s8ws3

Rob

Date: 12 Dec 2006 06:40:30
From: The Historian
Subject: Re: Turing Machines

Guy Macon wrote:
> Phil, It's considered poor netiquette to top-post and to
> quote an entire post including sig. Please quote only the
> relevant part(s) of the message you reply to with your
> comments interspersed and placed beneath each quote.

His I-ness apparently is the only person allowed to snip. Even if you
quote an Innes post in its gory entirety, and add the word "snip" at
the end, he complains you snipped the important part of his rant.

Date: 11 Dec 2006 21:35:20
From: MaximRecoil
Subject: Re: Human vs Machine Matches

David Richerby wrote:
> > Again? I am referring to tangible copies of opening books,
> > historical games and end game table bases.
>
> OK. Good start. Now, define `tangible' in a way that distinguishes
> my computer's `memory' from my own.

I don't need to define "tangible", as it already has a definition; and
that definition already includes digital data and does not include
human memories.

> >> Straw man. Computers don't play in regular chess tournaments and
> >> they don't take high school exams.
> >
> > You obviously don't know what a "straw man" is. I was giving two
> > real word examples illustrating the general consensus among people
> > that reference materials "have many obvious advantages over
> > memory". Then you replied with a blatant misuse of the term "straw
> > man", followed by a couple of non sequiturs.
>
> Thanks for the ad hominem. Much appreciated.

First you misuse the term "straw man", and your followup is to misuse
the term "ad hominem"?

> The fact that bringing reference materials into a chess tournament or
> a high-school exam is prohibited is irrelevant because we are not
> talking about chess tournaments or high-school exams. Computers are
> almost never allowed to participate in chess tournaments and do not
> sit high-school exams and, therefore, the prohibition of reference
> materials in such situations is irrelevant.
>
> (Consider this: if your argument is valid, one can also derive
> `Reference materials should be banned from ballroom dancing' by the
> same process.)

You are still confused. I did not give the examples of a typical chess
tournament or a high school exam in order to say that because reference
materials are disallowed in those cases they should be disallowed in
human vs. computer chess matches. I gave the examples as an
illustration of the general consensus among humans that tangible
reference materials are both different and in many ways superior to
human memory. I have already cleared this up for you once. This makes
the second time.

In fact, I have never claimed that tangible reference materials should
necessarily be disallowed from use by computers. I have said that
either both sides should be able to use them or neither side should be
able to use them. Is that clear?

> >>> 1. Reference materials are tangible
> >>> 2. In this context, reference materials = opening book, end game
> >>> tablebases, and historical games.
> >>
> >> How is the opening book of a computer chess program tangible?
> >
> > Is that a joke?
>
> No. I'm being absolutely serious. My computer's `memory' of the
> Sicilian defence is coded in the orientation of magnetic fields on
> little spinning pieces of metal. My memory of the Sicilian defence is
> coded in the connections between the different cells in my brain. How
> is my computer's `memory' more tangible than mine?

Anything tangible is subject to independent and objective observation
by anyone.

> > It is a digital copy of an opening book. It can be viewed on a
> > monitor or printed out on paper.
>
> My memory of the Sicilian defence can also be written out on paper.
> The only difference is one of perfection: my computer makes fewer
> mistakes when writing out its `memory' than I do.

At the point that you write it down, it becomes tangible reference
material and that piece of paper would be disallowed. Digital data is
already tangible, as it can be independently and objectively observed.

And BTW, this semantic issue is getting away from the actual point,
which is, once you get beyond what is essential for each party to play
legal chess, any additional help should either be allowed for both
sides or disallowed for both sides.

> > It is the difference between determining something based on its own
> > real time calculations and simply copying a move from a book.
>
> And what is this difference? It's a very common optimization
> technique in programming to have a list of common cases done by lookup
> and do the rest by calculation.
>
> To give an example in an algorithm used by humans, if I ask you to
> work out 747x34, you will use a mental look-up table that tells you
> that 7x4=28, 4x4=16 and so on: you won't sit there thinking `7x4 =
> 7+7+7+7 = 7+7+7+1+1+1+1+1+1+1 = 7+7+8+1+1+1+1+1+1 = 7+7+9+1+1+1+1+1 =
> ... = 7+7+14 = ...'
>
> You still haven't explained why some algorithms (alpha-beta and
> friends) are acceptable but some (table lookup) aren't.

Once again, "figure out" = fine, "copy" = not fine, and this is in
reference to the final answer, in this case, being the move.

> > Humans also have the ability to copy moves from books, but they are
> > specifically prohibited from doing so in a typical tournament.
>
> So I'm cheating if I sit down at a chessboard and play 1.e4 without
> analyzing any variations? Again, this is not a joke. Why is it
> acceptable for me to use the thought process `All chess games should
> begin with 1.e4' but unacceptable for a computer to use exactly the
> same reasoning?

What you do in your mind is not tangible, i.e., it is not able to be
independently observed. What *is* illegal is opening a book or text
file on a PC that says "Your first move should be e4". You can't
prohibit intangible processes but you can prohibit tangible processes.

Date: 11 Dec 2006 14:35:32
From: Guy Macon
Subject: Re: Turing Machines

tin Brown wrote:
>
>Chess One wrote:
>
>> "Guy Macon" <http://www.guymacon.com/> wrote...
>> >
>> > Chess One wrote:
>> >>
>> >>Guy Macon wrote...
>> >>>
>> >>> Assuming that you are using the term "Turing Engine" to describe
>> >>> a Turing Machine (please define your terms if this assumption
>> >>> is incorrect)...
>> >>
>> >>Not correct. 'Turing Engine' is intended, and specifically differentiated
>> >>from a machine. It is a philosophical prospect, decribed as a logic
>> >>flow-chart upon simple if... then... propositions.
>
>OK Humpty-Dumpty you *must* then define what you mean when you write
>the words 'Turing Engine' before any meaningful communication is
>possible.
>
>In particular how is it distinguished from a Universal Turing Machine
>and what other unstated assumptions are implicit in your unconventional
>terminology.
>
>> > You appear to have failed to, as requested, define your terms.
>> > If you wish to discuss something called a "Turing Engine" that
>> > is different from a Turing Machine or Universal Turing Machine
>> > (not the same thing, BTW), you ned to define exactly what you
>> > mean by "Turing Engine" as I requested. Until you do, I shall
>> > assume that what you call a "Turing Engine" is what Turing himself
>> > and everyone else who studies computer science calls a "Universal
>> > Turing Machine."
>>
>> You assume what you wish. That is not my business, nor correcting you, since
>> you seem to be unfamiliar with the term 'Turing himself' used.
>
>This seems unlikely since the Turing online digital archive does not
>appear to contain anywhere the compound phrase "Turing engine". It
>does refer to his computational hardware designs for the Automatic
>Computing Engine, ACE and production model DEUCE.
>
>http://www.turingarchive.org/
>
>For anyone that wishes to verify this.

Regarding the accusation that I am unfamiliar with the term
Turing himself used:

"It is possible to invent a single machine which can be used to
compute any computable sequence. If this machine U is supplied
with the tape on the beginning of which is written the string
of quintuples separated by semicolons of some computing machine
M, then U will compute the same sequence as M"

Source: Alan Turing on page 128 of _The Undecidable, Basic
Papers on Undecidable Propositions, Unsolvable Problems And
Computable Functions, Raven Press, New York, 1965.

Date: 11 Dec 2006 15:04:17
From: Chess One
Subject: Re: Turing Machines

"Guy Macon" <http://www.guymacon.com/ > wrote in message
news:[email protected]...

> Regarding the accusation that I am unfamiliar with the term
> Turing himself used:
>
> "It is possible to invent a single machine which can be used to
> compute any computable sequence. If this machine U is supplied
> with the tape on the beginning of which is written the string
> of quintuples separated by semicolons of some computing machine
> M, then U will compute the same sequence as M"
>
> Source: Alan Turing on page 128 of _The Undecidable, Basic
> Papers on Undecidable Propositions, Unsolvable Problems And
> Computable Functions, Raven Press, New York, 1965.

I know he used the term Turing Machine, and he also used the term I spoke of
initially, which is no specific reference to any electro-mechanical device,
and the difference between the Engine and the Machine is just such. I think
if you examine this, you will observe a respectable definition.

If you wish, the machine is /a/ means of deploying the work of the engine -
and I only ask you to note that these terms are not synonyms, and one is a
model or map, and the other the territory, and the territory is not the map.

Upon paradigms, it is also impossible to create a 1:1 map.

The very first program deployed by Turing can be run on much faster machines
these days, and its bleakly amusing to view on the modern processor just how
long any single Enigma message takes to de-crypt. As I am sure you know,
Turing's first vacuum tube creations would have been able to do the same
work - though taking several hundred thousand years to do so. But in essence
the same task is undertaken by the same /philosophical engine/ - which is to
say, the form of processing of the idea, which is again to say, that this
'engine' is a verb, not a noun as is 'machine' - and the singular difference
in terms of output of the engine by machines or non-machines is time.

Hence the necessity of the 'crib'. Otherwise U may compute until M's atoms
reach their half-life without actually resolving anything, and as such the
computational result is /in potential/, rather than actual. Perhaps you know
that alone the Bletchley machines never solved any Enigma traffic without
being supplied with the precedent of a day-crib?

In terms of brute-force solving of chess, even if current processing speeds
are incremented * 10^5, the universe is 'solved' first - and the number of
possibilities exceed all the atoms in it.

Phil Innes

Okay, some people estimate * 10^4 since we can calculate end of universe but
not end of chess.

Date: 11 Dec 2006 15:42:26
From: David Richerby
Subject: Re: Turing Machines

Chess One <[email protected] > wrote:
>"Guy Macon" <http://www.guymacon.com/> wrote:
>> Regarding the accusation that I am unfamiliar with the term Turing
>> himself used:
>>
>> "It is possible to invent a single machine which can be used to
>> compute any computable sequence. If this machine U is supplied
>> with the tape on the beginning of which is written the string of
>> quintuples separated by semicolons of some computing machine M,
>> then U will compute the same sequence as M"
>>
>> Source: Alan Turing on page 128 of _The Undecidable, Basic Papers
>> on Undecidable Propositions, Unsolvable Problems And Computable
>> Functions, Raven Press, New York, 1965.
>
> I know he used the term Turing Machine, and he also used the term I
> spoke of initially

Could you give a citation for the phrase `Turing Engine' appearing in
any of Turing's works? Of course, there's a good chance that Turing
wasn't so immodest as to name the thing after himself, so any cite to
the use of the word `engine' with an explanation of how this is
distinguished from a `machine' would be useful.

Certainly, in contemporary usage, a `Turing machine' is just a
mathematical concept.

Dave.

--
David Richerby Broken Addictive Projector (TM): it's
www.chiark.greenend.org.uk/~davidr/ like a 16mm film projector but you can
never put it down and it doesn't work!

Date:
From: Martin Brown
Subject: Re: Human vs Machine Matches

Date:
From: Martin Brown
Subject: Re: Turing Machines

Date: 13 Jan 2007 22:19:14
From: Chess One
Subject: Re: Turing Machines

"tin Brown" <

Date: 11 Dec 2006 05:25:49
From: MaximRecoil
Subject: Re: Human vs Machine Matches

David Richerby wrote:
> MaximRecoil <[email protected]> wrote:
> > I used the term "reference material" to distinguish from human
> > memory.
>
> How about you tell us what `reference material' *is* instead of what
> it is *not*. Knowing that `reference material' is different from
> `human memory' doesn't help an awful lot more than knowing that it is
> different from a dead fish.

Again? I am referring to tangible copies of opening books, historical
games and end game table bases.

> > Reference materials have many obvious advantages over memory. That
> > is why they are prohibited in a typical chess tournament, or during
> > a typical high school exam.
>
> Straw man. Computers don't play in regular chess tournaments and they
> don't take high school exams.

You obviously don't know what a "straw man" is. I was giving two real
word examples illustrating the general consensus among people that
reference materials "have many obvious advantages over memory". Then
you replied with a blatant misuse of the term "straw man", followed by
a couple of non sequiturs.

> > Secondly, you seem to be implying that you believe there is no
> > difference between the advantage afforded by the data "carried in
> > his head" vs. data printed out on paper.
>
> Not at all. But I've never seen a computer bring material printed on
> paper into a chess match so I'm at a loss to see how this is relevant.

Digital copies of data are functionally equivalent to the same data
printed on paper. For example, if taking a test where no reference
materials are allowed, bringing in a "cheat sheet" in a text file on
your laptop is no more permissible than bringing the cheat sheet in
printed out on paper.

> > 1. Reference materials are tangible
> > 2. In this context, reference materials = opening book, end game
> > tablebases, and historical games.
>
> How is the opening book of a computer chess program tangible?

Is that a joke? It is a digital copy of an opening book. It can be
viewed on a monitor or printed out on paper.

> > Now while computers don't "know" anything, they are pretty good at
> > the "figure out" part, and as such, can play decent chess even
> > without the reference materials; just based on their calculations.
>
> Why is it acceptable to load the code for the alpha-beta search and
> associated evaluation function from the hard drive but not
> to load an opening book? Why is it acceptable to program the computer
> to `know' that isolated pawns are bad and rooks on open files are good
> but not to say, `In the initial position, the correct move is 1.e4'?

It is the difference between determining something based on its own
real time calculations and simply copying a move from a book. Humans
also have the ability to copy moves from books, but they are
specifically prohibited from doing so in a typical tournament.

Date: 11 Dec 2006 15:35:37
From: David Richerby
Subject: Re: Human vs Machine Matches

MaximRecoil <[email protected] > wrote:
> David Richerby wrote:
>> MaximRecoil <[email protected]> wrote:
>>> I used the term "reference material" to distinguish from human
>>> memory.
>>
>> How about you tell us what `reference material' *is* instead of
>> what it is *not*. Knowing that `reference material' is different
>> from `human memory' doesn't help an awful lot more than knowing
>> that it is different from a dead fish.
>
> Again? I am referring to tangible copies of opening books,
> historical games and end game table bases.

OK. Good start. Now, define `tangible' in a way that distinguishes
my computer's `memory' from my own.

>>> Reference materials have many obvious advantages over memory.
>>> That is why they are prohibited in a typical chess tournament, or
>>> during a typical high school exam.
>>
>> Straw man. Computers don't play in regular chess tournaments and
>> they don't take high school exams.
>
> You obviously don't know what a "straw man" is. I was giving two
> real word examples illustrating the general consensus among people
> that reference materials "have many obvious advantages over
> memory". Then you replied with a blatant misuse of the term "straw
> man", followed by a couple of non sequiturs.

Thanks for the ad hominem. Much appreciated.

The fact that bringing reference materials into a chess tournament or
a high-school exam is prohibited is irrelevant because we are not
talking about chess tournaments or high-school exams. Computers are
almost never allowed to participate in chess tournaments and do not
sit high-school exams and, therefore, the prohibition of reference
materials in such situations is irrelevant.

(Consider this: if your argument is valid, one can also derive
`Reference materials should be banned from ballroom dancing' by the
same process.)

>>> 1. Reference materials are tangible
>>> 2. In this context, reference materials = opening book, end game
>>> tablebases, and historical games.
>>
>> How is the opening book of a computer chess program tangible?
>
> Is that a joke?

No. I'm being absolutely serious. My computer's `memory' of the
Sicilian defence is coded in the orientation of magnetic fields on
little spinning pieces of metal. My memory of the Sicilian defence is
coded in the connections between the different cells in my brain. How
is my computer's `memory' more tangible than mine?

> It is a digital copy of an opening book. It can be viewed on a
> monitor or printed out on paper.

My memory of the Sicilian defence can also be written out on paper.
The only difference is one of perfection: my computer makes fewer
mistakes when writing out its `memory' than I do.

>> Why is it acceptable to load the code for the alpha-beta search and
>> associated evaluation function from the hard drive but not to load
>> an opening book? Why is it acceptable to program the computer to
>> `know' that isolated pawns are bad and rooks on open files are good
>> but not to say, `In the initial position, the correct move is
>> 1.e4'?
>
> It is the difference between determining something based on its own
> real time calculations and simply copying a move from a book.

And what is this difference? It's a very common optimization
technique in programming to have a list of common cases done by lookup
and do the rest by calculation.

To give an example in an algorithm used by humans, if I ask you to
work out 747x34, you will use a mental look-up table that tells you
that 7x4=28, 4x4=16 and so on: you won't sit there thinking `7x4 =
7+7+7+7 = 7+7+7+1+1+1+1+1+1+1 = 7+7+8+1+1+1+1+1+1 = 7+7+9+1+1+1+1+1 =
... = 7+7+14 = ...'

You still haven't explained why some algorithms (alpha-beta and
friends) are acceptable but some (table lookup) aren't.

> Humans also have the ability to copy moves from books, but they are
> specifically prohibited from doing so in a typical tournament.

So I'm cheating if I sit down at a chessboard and play 1.e4 without
analyzing any variations? Again, this is not a joke. Why is it
acceptable for me to use the thought process `All chess games should
begin with 1.e4' but unacceptable for a computer to use exactly the
same reasoning?

Dave.

--
David Richerby Crystal Dish (TM): it's like a fine
www.chiark.greenend.org.uk/~davidr/ ceramic dish but it's completely
transparent!

Date: 11 Dec 2006 03:29:28
From: MaximRecoil
Subject: Re: Human vs Machine Matches

David Richerby wrote:
> Kenneth Sloan <[email protected]> wrote:
> > Look: you know, and everyone else knows, that the crux of this
> > argument lies in the definition of "reference material". Your
> > postings beg the question by assuming a particular definition.
>
> I would like to thank you for the first correct use of the phrase `beg
> the question' that I have seen for some time. (NB: `beg the question'
> does not mean `cause the question to be asked'.)

Just because he didn't use the phrase in the popular incorrect way
doesn't mean he used it correctly.

Tangible copies of opening books, historical games and end game
tablebases are prohibited from use at typical chess tournaments. If
there was a piece of paper with the rules for the tournament printed on
it, and it used the term "reference materials", the average English
speaker would understand what that term meant.

Additionally, my argument does not depend on my definition of
"reference materials", because when the obtuse (intentionally or
otherwise) come along, I can simply substitute the long description:
"tangible copies of opening books, historical games and end game
tablebases" and we will be on the same page. So if anyone wants to
argue that "reference materials" does not mean what I am using it to
mean, who cares? It doesn't change the fact that I am talking about
"tangible copies of opening books, historical games and end game
tablebases".

So I am obviously not "begging the question", when the "question" (i.e.
the point of contention) has nothing to do with the definition of
"reference materials". The "question" has to with allowing "X" for
"Player A" while simultaneously disallowing "X" for "Player B".

X = Tangible copies of opening books, historical games and end game
tablebases (reference materials)

Player A = computer

Player B = human

Date: 11 Dec 2006 16:51:38
From: David Richerby
Subject: Re: Human vs Machine Matches

MaximRecoil <[email protected] > wrote:
> Tangible copies of opening books, historical games and end game
> tablebases are prohibited from use at typical chess tournaments. If
> there was a piece of paper with the rules for the tournament printed on
> it, and it used the term "reference materials", the average English
> speaker would understand what that term meant.

Yes, *with respect to human players*. But that is not what is under
discussion, here. What is under discussion here is what the term
means with respect to computer programs.

> Additionally, my argument does not depend on my definition of
> "reference materials", because when the obtuse (intentionally or
> otherwise) come along, I can simply substitute the long description:
> "tangible copies of opening books, historical games and end game
> tablebases" and we will be on the same page.

OK. Now define `tangible'.

> So I am obviously not "begging the question", when the "question"
> (i.e. the point of contention) has nothing to do with the
> definition of "reference materials". The "question" has to with
> allowing "X" for "Player A" while simultaneously disallowing "X"
> for "Player B".

OK. Allow me to instantiate `X'. The question has to do with
allowing the `the use of a Pentium 4 processor' to for `Deep Fritz'
while simultaneously disallowing `the use of a Pentium 4 processor'
for `Vladimir Kramnik.' Please explain the discrepancy.

Dave.

--
David Richerby Generic Tree (TM): it's like a tree
www.chiark.greenend.org.uk/~davidr/ but it's just like all the others!

Date: 11 Dec 2006 11:26:46
From: Guy Macon
Subject: Re: Turing Machines

help bot wrote:

> You know, when IM Innes is in the mood, he demands
>definitions almost at random, rather than giving simple,
>direct answers to the questions posed to him. Yet when
>some other poster comes along and does the same thing
>to IM Innes, his response is to insist that no such definitions
>are necessary, or that they have already been given. Weird,
>huh?

...and annoying!

> I couldn't help bot notice this other guy, this Mr. Macon,
>clearly seemed confused by precisely what IM Innes meant
>by terms such as "Turing Engine", as opposed to other,
>similar terms. To me, this indicates that the famous IM
>does in fact need to spell out his precise definitions, if he
>is not merely seeking to be able to say "wrong answer!"
>to each response posted in reply. But then, perhaps that
>is exactly what he wishes, in which case his response
>above was sufficient for the task.

Thank you, Oh Great Help Bot!. You have been a great help
indeed!! I can only add that, as an experienced engineer
who has created everything from electronics that help keep
commercial aircraft and space shuttles aloft to electronic
toys that are in use by millions of children, I am far
better qualified than Phil Innes is to define terms relating
to engineering and computer science.

"I don't believe we will see this doubling every two years
much longer, because the gigahertz barrier is a tough one,
when electrical properties change quite drastically as
suddenly everything on the chip is an antenna for microwave
frequencies..." -- Phil Innes, 1998

"Where did you [Phil Innes] learn to write so many words,
but say so little??? Is there a school for that??"
-- Robert Hyatt (author of Cray Blitz and Crafty)

Date: 11 Dec 2006 21:42:12
From: Chess One
Subject: Re: Turing Machines

I never wrote what is ascribed to me below by Macon who with Kennedy has
joined the liars club and who states he is far better by virtue of his work.
What provenance does this fellow Macon claim for his lie, and his
scandalising of people becuas ehge knows so very little theory?

I am so glad I never mentioned my work with NASA and electro-optics in this
thread, a relatively abstruse discipline as is optical computing, since it
seems then we would then abandon content entirely in preferernce to boasting
who has a bigger 'rocket'.

I was interested in what is understood, rather than known. But now I know
both.

Phil Innes

"Guy Macon" <http://www.guymacon.com/ > wrote in message
news:[email protected]...
>
>
>
> help bot wrote:
>
>> You know, when IM Innes is in the mood, he demands
>>definitions almost at random, rather than giving simple,
>>direct answers to the questions posed to him. Yet when
>>some other poster comes along and does the same thing
>>to IM Innes, his response is to insist that no such definitions
>>are necessary, or that they have already been given. Weird,
>>huh?
>
> ...and annoying!
>
>> I couldn't help bot notice this other guy, this Mr. Macon,
>>clearly seemed confused by precisely what IM Innes meant
>>by terms such as "Turing Engine", as opposed to other,
>>similar terms. To me, this indicates that the famous IM
>>does in fact need to spell out his precise definitions, if he
>>is not merely seeking to be able to say "wrong answer!"
>>to each response posted in reply. But then, perhaps that
>>is exactly what he wishes, in which case his response
>>above was sufficient for the task.
>
> Thank you, Oh Great Help Bot!. You have been a great help
> indeed!! I can only add that, as an experienced engineer
> who has created everything from electronics that help keep
> commercial aircraft and space shuttles aloft to electronic
> toys that are in use by millions of children, I am far
> better qualified than Phil Innes is to define terms relating
> to engineering and computer science.
>
>
> "I don't believe we will see this doubling every two years
> much longer, because the gigahertz barrier is a tough one,
> when electrical properties change quite drastically as
> suddenly everything on the chip is an antenna for microwave
> frequencies..." -- Phil Innes, 1998
>
>
> "Where did you [Phil Innes] learn to write so many words,
> but say so little??? Is there a school for that??"
> -- Robert Hyatt (author of Cray Blitz and Crafty)
>
>

Date: 12 Dec 2006 08:17:55
From: Guy Macon
Subject: Re: Turing Machines

Chess One wrote:

>I never wrote what is ascribed to me below by Macon who
>with Kennedy has joined the liars club

Here is the source:

"I don't believe we will see this doubling every two years
much longer, because the gigahertz barrier is a tough one,
when electrical properties change quite drastically as
suddenly everything on the chip is an antenna for microwave
frequencies..." -- Phil Innes, 1998

Source: Usenet post
From: Phil Innes <[email protected] >
Newsgroups: rec.games.chess.computer
Subject: Re: Bob, can we have a substantive answer please ... ?
Date: 16 Aug 1998 02:16:35 GMT
Message-ID: <[email protected] >#1/1
X-Trace: news7.ispnews.com 903233795 18856 209.198.90.217
...
http://groups.google.com/group/rec.games.chess.computer/msg/3f6f9eeca24fa166?dmode=source

>Macon, you seem like a clueless technologist - why did you invent some
>supposed conversation with Robt Hyatt?

Here is the source:

"Where did you [Phil Innes] learn to write so many words,
but say so little??? Is there a school for that??"
-- Robert Hyatt (author of Cray Blitz and Crafty)

Source: Usenet post
From: Robert Hyatt <[email protected] >
Newsgroups: rec.games.chess.computer
Subject: Re: Strawman has no brains
Date: 22 Oct 2001 01:44:51 GMT
Message-ID: <[email protected] >
NNTP-Posting-Host: crafty.cis.uab.edu
...
http://groups.google.com/group/rec.games.chess.computer/msg/952130bc50cbe178?dmode=source

Caught in another fib, I see.

Date: 13 Jan 2007 22:15:04
From: Chess One
Subject: Re: Turing Machines

that's not a quote 'macon'. you are making it up just like billy bunter! i
dunno if i wrote in that thread or post but those are not my words and

http://groups.google.com/group/rec.games.chess.computer/msg/952130bc50cbe178?dmode=source -
did not match any documents.

pi

"Guy Macon" <http://www.guymacon.com/ > wrote in message
news:[email protected]...
>
>
>
> Chess One wrote:
>
>>I never wrote what is ascribed to me below by Macon who
>>with Kennedy has joined the liars club
>
> Here is the source:
>
> "I don't believe we will see this doubling every two years
> much longer, because the gigahertz barrier is a tough one,
> when electrical properties change quite drastically as
> suddenly everything on the chip is an antenna for microwave
> frequencies..." -- Phil Innes, 1998
>
> Source: Usenet post
> From: Phil Innes <[email protected]>
> Newsgroups: rec.games.chess.computer
> Subject: Re: Bob, can we have a substantive answer please ... ?
> Date: 16 Aug 1998 02:16:35 GMT
> Message-ID: <[email protected]>#1/1
> X-Trace: news7.ispnews.com 903233795 18856 209.198.90.217
> ...
> http://groups.google.com/group/rec.games.chess.computer/msg/3f6f9eeca24fa166?dmode=source
>
>>Macon, you seem like a clueless technologist - why did you invent some
>>supposed conversation with Robt Hyatt?
>
> Here is the source:
>
> "Where did you [Phil Innes] learn to write so many words,
> but say so little??? Is there a school for that??"
> -- Robert Hyatt (author of Cray Blitz and Crafty)
>
> Source: Usenet post
> From: Robert Hyatt <[email protected]>
> Newsgroups: rec.games.chess.computer
> Subject: Re: Strawman has no brains
> Date: 22 Oct 2001 01:44:51 GMT
> Message-ID: <[email protected]>
> NNTP-Posting-Host: crafty.cis.uab.edu
> ...
> http://groups.google.com/group/rec.games.chess.computer/msg/952130bc50cbe178?dmode=source
>
> Caught in another fib, I see.
>
>
>
>
>

Date: 14 Jan 2007 12:55:53
From: Guy Macon
Subject: Re: Turing Machines

"I don't believe we will see this doubling every two years
much longer, because the gigahertz barrier is a tough one,
when electrical properties change quite drastically as
suddenly everything on the chip is an antenna for microwave
frequencies..." -- Phil Innes, 1998

"Where did you [Phil Innes] learn to write so many words,
but say so little??? Is there a school for that??"
-- Robert Hyatt (author of Cray Blitz and Crafty)

Chess One wrote:

>Results 1 - 10 of about 1,050,000 for turing engine. (0.11 seconds)

Moron. You just searched for all web pages that contain "Turing" and
all web pages that contain "Engine." Do the search with quoute ks
around "Turing Engine" and you get 435 results. Actually look at those
results and you will see that they most or all of them consist of:

[1] Your own bleatings.

[2] That one book where the publisher though that calling Turing's
machine an "engine" in the title sounded olde-tyme and cool.

[3] Foreign language speakers using the wrong word.

This has been explained to you before

Chess One wrote:

>that's not a quote 'macon'. you are making it up just like billy bunter!

Liar.

>those are not my words

Liar.

>http://groups.google.com/group/rec.games.chess.computer/msg/952130bc50cbe178?dmode=source
> - did not match any documents.

*Stupid* Liar. Anyone can test the links I posted and see that they return:

http://groups.google.com/group/rec.games.chess.computer/msg/3f6f9eeca24fa166?dmode=source

Date: 10 Dec 2006 22:06:31
From: help bot
Subject: Re: Turing Machines

Chess One wrote:

> "Guy Macon" <http://www.guymacon.com/> wrote in message
> news:[email protected]...
> >
> > Chess One wrote:
> >
> >>It seems appropriate to ask some advanced ideas, yet elementary
> >>factors, and to question people's knowledge of computing and logic:-
> >
> > My resume (see <http://www.guymacon.com/>) speaks for itself.
> >
> >>Can anyone tell me if they understand:-
> >>
> >>1) what is a Turing Engine?

Yes, they can.

> > Assuming that you are using the term "Turing Engine" to describe
> > a Turing Machine (please define your terms if this assumption
> > is incorrect)...
>
> Not correct. 'Turing Engine' is intended, and specifically differentiated
> from a machine. It is a philosophical prospect, decribed as a logic
> flow-chart upon simple if... then... propositions.
>
> > A Turing Machine is any finite state sequential processor that
> > communicates with external information storage.
> >
> >>2) can a Turing Engine be made of paper, rather than requiring
> >>electricity or mechanical components?
> >
> > A Turing Machine can be made of nothing at all -- a zero state
> > and xero storage Turing Machine is still a Turing Machine.
> >
> > I suspect that you really aren't asking about Turing Machines,
> > but rather about Universal Turing Machines. Those cannot be
> > made out of paper alone, but can be made out of paper if you
> > allow the addition of, say, a human being with a pencil. Of
> > course you need an infinite amount of paper and pencil lead...
>
> You discourse on application means, while I reference the philosophic
> 'engine'.

Where exactly did the term "philosophic" magically
appear from, and why?

> >>3) can anyone distinguish between instruction sets which are
> >>logic trees, and other forms of data sets?
> >
> > The concept of data as something seperate from the program
> > is an assumption that isn't valid for all possible Turing
> > Machines.
>
> I'm sorry, but that response does not answer the question, and involves new
> terms. The question is simpler, if somewhat rhetorical, and can be answered
> yes or no.
>
> >>4) what is similar with all Turing Engines?
> >
> > Nothing. "All Turing Machines" includes the zero state and xero
> > storage Turing Machine.
> >
> > If you meant to ask what is similar with all Universal Turing
> > Machines, the answer is states, transitions between states,
> > actions, and storage.
>
> You again change the question in order to reply. This is not an answer.
>
> >>5) what is the only factor which differentiates all Turing Engines?
> >
> > The question is unclear. How does one differentiate all of something?
> > differentiate it from what?
>
> Differentiates them from each other, and I did not write 'all of something'
> I wrote "all Turing Engines."

Yes, yes; by which you supposedly meant some
philosophical thing, which you failed to indicate in
the actual question.

> >>6) if a book is described as a data-set, and a computer is a
> >>data-set, is a book a computer?
> >
> > If you call a tail a leg, does a sheep have five legs?

Oviously, yes. ("I easily answered these stupid questions."
-- Robert Fischer)

> Sorry! Hypothetical responses deny ability to answer specific questions.
>
> >>7) what is computed in a book?
> >
> > Define "book."

Define "avoids answering the question". No, wait. You
just did!

> Why? Don't you know what a book is?

Of course I know what a book is. It's what is thrown
at someone who has violated the law (unless they have
the right connections, so to speak.)

> >>8) what part of a computer is not data?

Good question.

> > What part of yellow is not an integer?

A: All seven letters.

> 0/8 so far!
>
> >>9) what part of a Turing Engine is not data?
> >
> > Again the concept of data as something seperate

Wrong! "Separate" has but two "e"s in it.

> > from the program
> > is an assumption that isn't valid for all possible Turing Machines.
>
> Wrong answer - no machine is indicated in the question, but a philosophic
> engine is. Though the answer differentiates forms of datum, this is not to
> anser the question, of what is NOT datum.
>
> >>10) of what was the first Turing Engine constucted?
> >
> > Define "constructed."

Whatever for? That word was not even a part of the
question; he asked about "constucted" Turing Motors.

> > Do we count Turing's original pencil and paper verion?

Define "verion".

> 0/10 for this writer. I have used no terms in these questions which are
> atypical ones - and though the term 'engine' is here used philosophically,
> in initial capitals a 'Turing Engine' is indeed a specific.
>
> >>Perhaps when we examine our knowledge of this subject, we can then assess
> >>our understanding of people.
> >
> > Now I have a question for you:
> >
> > A) What is your point?
>
> I have 10 points, so to speak. I want to know if anyone can answer any of
> them, which all relate to computing, and the philosophy of science, before
> becoming dazzled by the same people's understanding of the human role in
> programming or in competition.
>
> Phil Innes
>
> > Guy Macon
> > <http://www.guymacon.com/>

You know, when IM Innes is in the mood, he demands
definitions almost at random, rather than giving simple,
direct answers to the questions posed to him. Yet when
some other poster comes along and does the same thing
to IM Innes, his response is to insist that no such definitions
are necessary, or that they have already been given. Weird,
huh?

I couldn't help bot notice this other guy, this Mr. Macon,
clearly seemed confused by precisely what IM Innes meant
by terms such as "Turing Engine", as opposed to other,
similar terms. To me, this indicates that the famous IM
does in fact need to spell out his precise definitions, if he
is not merely seeking to be able to say "wrong answer!"
to each response posted in reply. But then, perhaps that
is exactly what he wishes, in which case his response
above was sufficient for the task.

-- help bot

Date: 11 Dec 2006 13:52:53
From: Chess One
Subject: Re: Turing Machines

"help bot" <[email protected] > wrote in message
news:[email protected]...

>> >>5) what is the only factor which differentiates all Turing Engines?
>> >
>> > The question is unclear. How does one differentiate all of something?
>> > differentiate it from what?
>>
>> Differentiates them from each other, and I did not write 'all of
>> something'
>> I wrote "all Turing Engines."
>
>
> Yes, yes; by which you supposedly meant some
> philosophical thing, which you failed to indicate in
> the actual question.

Such /hysterical/ reactions to very simple terms common in logic,
programming and philosophy!

Here is someone else who doesn't know the /only/ differentiating factor of
Turing Engines is time-to-resolve.

> You know, when IM Innes is in the mood, he demands
> definitions almost at random, rather than giving simple,
> direct answers to the questions posed to him.

When this person Helpo-Bog requires mental stimulus he needs to taunt other
people with abstract criticism combined with out and out lies in order to
posit the basis for which he himself could then think of reasons why he
dislikes the person he taunts.

The questions are simple, and if some word in them seems obscure, its fair
to challenge their use in respect to any obscurity or unusual meaning.
Though I have stated here that I have used no unusual terms.

Greg Kennedy needs the word book defined for him too, though, neither he nor
Macon say why. But its always necessary to first divert the issue, then upon
apparent rational basis, write shit about people.

> Yet when
> some other poster comes along and does the same thing
> to IM Innes, his response is to insist that no such definitions
> are necessary, or that they have already been given. Weird,
> huh?
>
> I couldn't help bot notice this other guy, this Mr. Macon,
> clearly seemed confused by precisely what IM Innes meant
> by terms such as "Turing Engine",

60 SECOND TEST

Either of these geniuses could find out by googling it in 60 seconds instead
of writing their angry frustrated nonsense in public - and thereby
displaying their own attitude, but they are too lazy to do so, and assume
readers too lazy to notice.They think that by virtue of their indolent and
arrogant attitude, someone owes their ignorance a complete education.

> as opposed to other,
> similar terms. To me, this indicates that the famous IM
> does in fact need to spell out his precise definitions, if he
> is not merely seeking to be able to say "wrong answer!"
> to each response posted in reply. But then, perhaps that
> is exactly what he wishes, in which case his response
> above was sufficient for the task.

What confused nonsense replete with the usual 'perhaps' straw-man arguments,
content to change the question, rather than admit any understanding of very
simple and already specific terms - or even state why he got confused in the
first place, or why he is interested

Write your name Kennedy, or write your trash. Bet you can't do both! And I
bet you can't stop living vicariously on-line via other people's lives and
ideas.

It ain't manly. Get a life, bucko! Your own.

Phil Innes

> -- help bot
>

Date: 10 Dec 2006 18:51:43
From: The Historian
Subject: Re: Turing Machines

Guy Macon wrote:
> >>>7) what is computed in a book?
> >>
> >> Define "book."
> >
> >Why? Don't you know what a book is?
>
> Evasion noted.
>
> I know exactly what a book is, but you clearly do not, based on your
> question. You think books can compute. They cannot. I can only
> conclude that you are stupid, crazy, or have a non-standard definition
> of "book."

Correct answer was "all three."

Date: 09 Dec 2006 21:56:26
From: help bot
Subject: Re: Human vs Machine Matches

MaximRecoil wrote:
> Chris Mattern wrote:
> > In article <[email protected]>,
> > MaximRecoil wrote:
> > >
> > >Chris Mattern wrote:
> > >> In article <[email protected]>,
> > >> MaximRecoil wrote:
> >
> > <snip>
> >
> > >> Your definition of "reference materials" seems to be vague. In
> > >> this context, I would define it as "information storage external to
> > >> the subject". In this context, a computer using its hard disk is
> > >> not using "reference materials" as it is using its own internal
> > >> storage, as a human does his memory. If you have a different
> > >> definition which is useful in this context, I will be happy to
> > >> hear it and discuss it.
> > >> >
> > >> >Unless you can give good cause why there should be a double standard
> > >> >regarding the use of reference materials, then you are at odds with
> > >> >yourself.
> > >> >
> > >> I do not see any double standard, because I do not see how the computer
> > >> is using "reference materials".
> > >>
> > >
> > >Opening book information and end game tablebases are obvious "reference
> > >materials".
> >
> > No, they aren't. This is what I mean about your definition being
> > vague. Opening book information and game tablebase information
> > are carried around by any grandmaster inside his head. It's not
> > illegal for him to use them. It's only illegal if he refers to
> > an external reference.
>
> I used the term "reference material" to distinguish from human memory.
> Reference materials have many obvious advantages over memory. That is
> why they are prohibited in a typical chess tournament, or during a
> typical high school exam.
>
>
> > >The same data printed out and brought to a match by a human
> > >would be prohibited.
> >
> > But that data carried in his head and brought to a match that
> > way is not.
>
> For one thing, there is no way to establish that it is the "same data",
> because while reference materials are tangible, human memory is not.
> Secondly, you seem to be implying that you believe there is no
> difference between the advantage afforded by the data "carried in his
> head" vs. data printed out on paper. There is a general consensus about
> this already among human, and it is in disagreement with you. You can
> see evidence of this consensus by taking note of the fact that tangible
> reference materials are generally prohibited in chess tournaments or
> academic tests.
>
> > >Also, the many historical games that the computer
> > >can reference to find likely-to-be-winning moves when it encounters the
> > >same position in its current game is also clearly "reference material".
> >
> > No, it's not. A grandmaster has many games memorized in toto
> > in his memory. It's not illegal for him to use that data.
>
> He may or he may not, but once again, memory is not the same as
> tangible reference materials. If you propose that they are the same
> thing, then why not allow the human to reference the same materials
> that the computer does during the match? As it stands, the computer is
> allowed access to perfect digital copies of relevant data during the
> match while the human is disallowed access to the same. This creates a
> "handicap" for the human.
>
> > >This is also a response to Kenneth Sloan's false claim that I am
> > >begging the question. It is a false claim for 2 reasons. For one, in
> > >the course of discussion among folks who speak a common language, it is
> > >not necessary to define every word or term you use, particularly garden
> > >variety terms like "reference material".
> >
> > It is necessary to define "reference material" because you have made
> > it the central term of your argument. I have provided a consistent
> > definition. Can you *please* provide the definition on which you
> > are basing your argument? Because it is not apparent to me. I am
> > not being disingenuous or dishonest. I have provided a definition
> > that makes sense to me and drawn the natural conclusions from it.
> > I honestly cannot see what your definition is. Can you please
> > provide it?
>
> 1. Reference materials are tangible
> 2. In this context, reference materials = opening book, end game
> tablebases, and historical games.
>
> > >For example, if you showed up
> > >for a test and the instructor states that no reference materials are
> > >allowed during the exam, no one but the intentionally obtuse would ask
> > >for further clarification.
> >
> > Indeed. It would mean you're not allowed to use information from
> > an external source. I could quote examples in the exam to support
> > my argument if I remembered them, but I couldn't look them up from
> > an external source. ***The exact same data in either case; whether or
> > not I'm allowed to use that data depends not on its content but on
> > its source.***
>
> Of course. The point of the test is not to find out if the answers
> simply exist, but rather, if the person taking the test knows the
> answer. Likewise in chess, the point of the game is not to find out if
> the correct moves are known to someone, somewhere, recorded in
> reference material; it is to find out if the players know or can
> "figure out" the correct move. Now computers do not "know" anything,
> but equipping them with all the reference material you can find, while
> denying access to that same reference material for the human, creates a
> handicap for the human / advantage for the computer (which is an
> arbitrary advantage handed to the computer on a silver platter so to
> speak).
>
> Now while computers don't "know" anything, they are pretty good at the
> "figure out" part, and as such, can play decent chess even without the
> reference materials; just based on their calculations. This makes for
> as fair a matchup as is possible at this time between humans and
> computers. A human has poor calculation ability (compared to a
> computer) combined with experience (rooted in fallible memory) while a
> computer has raw calculation ability that is through the roof, along
> with being immune from human shortcomings such as fatigue,
> intimidation, self-doubt, etc.
>
> Since reference materials are not essential to a computer being able to
> play legal chess, and given the fact that the same reference materials
> could be used by either side for an advantage, it is simply unfair to
> allow one side's use of these materials and disallow the other side's
> use of the same materials.

I agree in part with Maxim Recoil, but the thing is, these
pro-computer people are still acting much the way they did
many years ago -- back when computers were on the
defensive vs. strong GMs. In short, you can't really reason
with them, or even hope to convince them that a computer's
internal copy of say, an openings database, is any unfair
advantage over a human, who in their view, is flawed in
not being able to internalize such data perfectly and without
limit. These anti-human activists are stuck in the same old
rut of old; they will never surrender, nay, not ever!

But I don't think that depriving computers of every such
reference material would be fair to them, in that humans
do have some internalized reference material, even if it
is both very limited and imperfectly accessible.

OTOH, if the programmers really wanted to, they could
eliminate the need for such databases by re-thinking their
programs a bit. Insert (or modify) code which determines
what phase of the game it is in, and then apply the
appropriate measures to play, say, the opening "properly".
There is really no need to bypass this phase by regurgitating
rote moves -- at least, not if the programmers put in the
required work.

In fact, a very recent game serves to demonstrate just
how "fair" things are at the moment. In Kramnik's recent
oversight of a mate-in-one, he showed precisely how a
human's preparation can circumvent or defuse the vast
openings database of a top level computer, yet the game
was decided on tactics. In essence, the human weakness
of carelessness allowed the strategically inferior player to
win. The same thing could be said of certain games from
the Kasparov matches against Deep-whatever.

It is only a matter of time before the top programs can
win or draw almost every game vs. the GMs, and I think
when they finally come to be seen as "invincible", there
may be some hope of compromise on this matter by the
anti-human programmers. Until then, I expect the war
will continue without any substantive progress.

-- help bot

Date: 10 Jan 2007 18:38:13
From: Chess One
Subject: Re: Human vs Machine Matches

"help bot" <[email protected] > wrote in message
news:[email protected]...

>> He may or he may not, but once again, memory is not the same as
>> tangible reference materials. If you propose that they are the same
>> thing, then why not allow the human to reference the same materials
>> that the computer does during the match? As it stands, the computer is
>> allowed access to perfect digital copies of relevant data during the
>> match while the human is disallowed access to the same. This creates a
>> "handicap" for the human.

This is very much the point. Obviously the rules are written for human
understanding, and even some programmers are said to be human ;)

The same rule cannot mean two things simultaneously. It either means you can
or cannot look-up opening material, and the clear intent with human players
is that you cannot, and I suppose to clarify the issue [which seems
necessary] Fide could state that this rule more exactly to include or
exclude 'virtual memory' and such look-ups as tablebases for endgames.

> It is only a matter of time before the top programs can
> win or draw almost every game vs. the GMs, and I think
> when they finally come to be seen as "invincible", there
> may be some hope of compromise on this matter by the
> anti-human programmers. Until then, I expect the war
> will continue without any substantive progress.

An end note will be when some 80 year old Vermont Nun who can't play chess
and doesn't even know the rules defeats Kramnik from her reference materials
in exactly the same way a computer did, since she didn't need to play a move
or even understand anything whatever. That is the reductio-ad-absurdam of
'memory' arguments.

Phil Innes

> -- help bot
>

Date: 09 Dec 2006 21:05:37
From: MaximRecoil
Subject: Re: Human vs Machine Matches

Chris Mattern wrote:
> In article <[email protected]>,
> MaximRecoil wrote:
> >
> >Chris Mattern wrote:
> >> In article <[email protected]>,
> >> MaximRecoil wrote:
>
> <snip>
>
> >> Your definition of "reference materials" seems to be vague. In
> >> this context, I would define it as "information storage external to
> >> the subject". In this context, a computer using its hard disk is
> >> not using "reference materials" as it is using its own internal
> >> storage, as a human does his memory. If you have a different
> >> definition which is useful in this context, I will be happy to
> >> hear it and discuss it.
> >> >
> >> >Unless you can give good cause why there should be a double standard
> >> >regarding the use of reference materials, then you are at odds with
> >> >yourself.
> >> >
> >> I do not see any double standard, because I do not see how the computer
> >> is using "reference materials".
> >>
> >
> >Opening book information and end game tablebases are obvious "reference
> >materials".
>
> No, they aren't. This is what I mean about your definition being
> vague. Opening book information and game tablebase information
> are carried around by any grandmaster inside his head. It's not
> illegal for him to use them. It's only illegal if he refers to
> an external reference.

I used the term "reference material" to distinguish from human memory.
Reference materials have many obvious advantages over memory. That is
why they are prohibited in a typical chess tournament, or during a
typical high school exam.

> >The same data printed out and brought to a match by a human
> >would be prohibited.
>
> But that data carried in his head and brought to a match that
> way is not.

For one thing, there is no way to establish that it is the "same data",
because while reference materials are tangible, human memory is not.
Secondly, you seem to be implying that you believe there is no
difference between the advantage afforded by the data "carried in his
head" vs. data printed out on paper. There is a general consensus about
this already among human, and it is in disagreement with you. You can
see evidence of this consensus by taking note of the fact that tangible
reference materials are generally prohibited in chess tournaments or
academic tests.

> >Also, the many historical games that the computer
> >can reference to find likely-to-be-winning moves when it encounters the
> >same position in its current game is also clearly "reference material".
>
> No, it's not. A grandmaster has many games memorized in toto
> in his memory. It's not illegal for him to use that data.

He may or he may not, but once again, memory is not the same as
tangible reference materials. If you propose that they are the same
thing, then why not allow the human to reference the same materials
that the computer does during the match? As it stands, the computer is
allowed access to perfect digital copies of relevant data during the
match while the human is disallowed access to the same. This creates a
"handicap" for the human.

> >This is also a response to Kenneth Sloan's false claim that I am
> >begging the question. It is a false claim for 2 reasons. For one, in
> >the course of discussion among folks who speak a common language, it is
> >not necessary to define every word or term you use, particularly garden
> >variety terms like "reference material".
>
> It is necessary to define "reference material" because you have made
> it the central term of your argument. I have provided a consistent
> definition. Can you *please* provide the definition on which you
> are basing your argument? Because it is not apparent to me. I am
> not being disingenuous or dishonest. I have provided a definition
> that makes sense to me and drawn the natural conclusions from it.
> I honestly cannot see what your definition is. Can you please
> provide it?

1. Reference materials are tangible
2. In this context, reference materials = opening book, end game
tablebases, and historical games.

> >For example, if you showed up
> >for a test and the instructor states that no reference materials are
> >allowed during the exam, no one but the intentionally obtuse would ask
> >for further clarification.
>
> Indeed. It would mean you're not allowed to use information from
> an external source. I could quote examples in the exam to support
> my argument if I remembered them, but I couldn't look them up from
> an external source. ***The exact same data in either case; whether or
> not I'm allowed to use that data depends not on its content but on
> its source.***

Of course. The point of the test is not to find out if the answers
simply exist, but rather, if the person taking the test knows the
answer. Likewise in chess, the point of the game is not to find out if
the correct moves are known to someone, somewhere, recorded in
reference material; it is to find out if the players know or can
"figure out" the correct move. Now computers do not "know" anything,
but equipping them with all the reference material you can find, while
denying access to that same reference material for the human, creates a
handicap for the human / advantage for the computer (which is an
arbitrary advantage handed to the computer on a silver platter so to
speak).

Now while computers don't "know" anything, they are pretty good at the
"figure out" part, and as such, can play decent chess even without the
reference materials; just based on their calculations. This makes for
as fair a matchup as is possible at this time between humans and
computers. A human has poor calculation ability (compared to a
computer) combined with experience (rooted in fallible memory) while a
computer has raw calculation ability that is through the roof, along
with being immune from human shortcomings such as fatigue,
intimidation, self-doubt, etc.

Since reference materials are not essential to a computer being able to
play legal chess, and given the fact that the same reference materials
could be used by either side for an advantage, it is simply unfair to
allow one side's use of these materials and disallow the other side's
use of the same materials.

> >Secondly, I have already illustrated what I meant by the term
> >"reference materials" in a reply to Kenneth Sloan on another thread, by
> >mentioning opening book, end game tablebases and historical games that
> >can be referenced.
> >
>
> Very good. Can you provide the reference then, or repost it?

It was in this thread -
http://groups.google.com/group/rec.games.chess.misc/browse_thread/thread/6cd1f780a0b0af83/90a65d2549f49414?lnk=raot#90a65d2549f49414

Date: 11 Dec 2006 11:39:32
From: David Richerby
Subject: Re: Human vs Machine Matches

MaximRecoil <[email protected] > wrote:
> I used the term "reference material" to distinguish from human
> memory.

How about you tell us what `reference material' *is* instead of what
it is *not*. Knowing that `reference material' is different from
`human memory' doesn't help an awful lot more than knowing that it is
different from a dead fish.

> Reference materials have many obvious advantages over memory. That
> is why they are prohibited in a typical chess tournament, or during
> a typical high school exam.

Straw man. Computers don't play in regular chess tournaments and they
don't take high school exams.

> Secondly, you seem to be implying that you believe there is no
> difference between the advantage afforded by the data "carried in
> his head" vs. data printed out on paper.

Not at all. But I've never seen a computer bring material printed on
paper into a chess match so I'm at a loss to see how this is relevant.

> 1. Reference materials are tangible
> 2. In this context, reference materials = opening book, end game
> tablebases, and historical games.

How is the opening book of a computer chess program tangible?

> Now while computers don't "know" anything, they are pretty good at
> the "figure out" part, and as such, can play decent chess even
> without the reference materials; just based on their calculations.

Why is it acceptable to load the code for the alpha-beta search and
associated evaluation function from the hard drive but not
to load an opening book? Why is it acceptable to program the computer
to `know' that isolated pawns are bad and rooks on open files are good
but not to say, `In the initial position, the correct move is 1.e4'?

Dave.

--
David Richerby Strange Sadistic Cat (TM): it's like
www.chiark.greenend.org.uk/~davidr/ a cuddly pet but it wants to hurt you
and it's totally weird!

Date: 09 Dec 2006 18:42:09
From: MaximRecoil
Subject: Re: Human vs Machine Matches

Chris Mattern wrote:
> In article <[email protected]>,
> MaximRecoil wrote:
> >
> >Guy Macon wrote:
> >> Chess One wrote:
> >>
> >> >a) Human plus database, vs Computer
> >>
> >> >b) Human vs Computers [look-ups = off] Simul.
> >>
> >> >both these suggestions are deliberately different than that,
> >> >since one aspect of arranging such matches is also to obtain
> >> >mainstream media interest, and whatever Kramnik/Fritz was, did not!
> >>
> >> Allowing the human to cheat will not increase media interest.
> >>
> >> Crippling the computer will not increase media interest.
> >>
> >
> >Why do you consider allowing the use of reference materials for the
> >human to be "cheating", while at the same time you consider disallowing
> >the use of reference materials for the computer to be "crippling"?
>
> Your definition of "reference materials" seems to be vague. In
> this context, I would define it as "information storage external to
> the subject". In this context, a computer using its hard disk is
> not using "reference materials" as it is using its own internal
> storage, as a human does his memory. If you have a different
> definition which is useful in this context, I will be happy to
> hear it and discuss it.
> >
> >Unless you can give good cause why there should be a double standard
> >regarding the use of reference materials, then you are at odds with
> >yourself.
> >
> I do not see any double standard, because I do not see how the computer
> is using "reference materials".
>

Opening book information and end game tablebases are obvious "reference
materials". The same data printed out and brought to a match by a human
would be prohibited. Also, the many historical games that the computer
can reference to find likely-to-be-winning moves when it encounters the
same position in its current game is also clearly "reference material".

This is also a response to Kenneth Sloan's false claim that I am
begging the question. It is a false claim for 2 reasons. For one, in
the course of discussion among folks who speak a common language, it is
not necessary to define every word or term you use, particularly garden
variety terms like "reference material". For example, if you showed up
for a test and the instructor states that no reference materials are
allowed during the exam, no one but the intentionally obtuse would ask
for further clarification.

Secondly, I have already illustrated what I meant by the term
"reference materials" in a reply to Kenneth Sloan on another thread, by
mentioning opening book, end game tablebases and historical games that
can be referenced.

Date: 10 Dec 2006 03:12:33
From: Chris Mattern
Subject: Re: Human vs Machine Matches

In article <[email protected] >,
MaximRecoil wrote:
>
>Chris Mattern wrote:
>> In article <[email protected]>,
>> MaximRecoil wrote:

<snip >

>> Your definition of "reference materials" seems to be vague. In
>> this context, I would define it as "information storage external to
>> the subject". In this context, a computer using its hard disk is
>> not using "reference materials" as it is using its own internal
>> storage, as a human does his memory. If you have a different
>> definition which is useful in this context, I will be happy to
>> hear it and discuss it.
>> >
>> >Unless you can give good cause why there should be a double standard
>> >regarding the use of reference materials, then you are at odds with
>> >yourself.
>> >
>> I do not see any double standard, because I do not see how the computer
>> is using "reference materials".
>>
>
>Opening book information and end game tablebases are obvious "reference
>materials".

No, they aren't. This is what I mean about your definition being
vague. Opening book information and game tablebase information
are carried around by any grandmaster inside his head. It's not
illegal for him to use them. It's only illegal if he refers to
an external reference.

>The same data printed out and brought to a match by a human
>would be prohibited.

But that data carried in his head and brought to a match that
way is not.

>Also, the many historical games that the computer
>can reference to find likely-to-be-winning moves when it encounters the
>same position in its current game is also clearly "reference material".

No, it's not. A grandmaster has many games memorized in toto
in his memory. It's not illegal for him to use that data.
>
>
>This is also a response to Kenneth Sloan's false claim that I am
>begging the question. It is a false claim for 2 reasons. For one, in
>the course of discussion among folks who speak a common language, it is
>not necessary to define every word or term you use, particularly garden
>variety terms like "reference material".

It is necessary to define "reference material" because you have made
it the central term of your argument. I have provided a consistent
definition. Can you *please* provide the definition on which you
are basing your argument? Because it is not apparent to me. I am
not being disingenuous or dishonest. I have provided a definition
that makes sense to me and drawn the natural conclusions from it.
I honestly cannot see what your definition is. Can you please
provide it?

>For example, if you showed up
>for a test and the instructor states that no reference materials are
>allowed during the exam, no one but the intentionally obtuse would ask
>for further clarification.

Indeed. It would mean you're not allowed to use information from
an external source. I could quote examples in the exam to support
my argument if I remembered them, but I couldn't look them up from
an external source. The exact same data in either case; whether or
not I'm allowed to use that data depends not on its content but on
its source.
>
>Secondly, I have already illustrated what I meant by the term
>"reference materials" in a reply to Kenneth Sloan on another thread, by
>mentioning opening book, end game tablebases and historical games that
>can be referenced.
>

Very good. Can you provide the reference then, or repost it?

--
Christopher Mattern

"Which one you figure tracked us?"
"The ugly one, sir."
"...Could you be more specific?"

Date: 09 Dec 2006 23:19:32
From: Jerzy
Subject: Re: Human vs Machine Matches

U�ytkownik "Chess One" <[email protected] > napisa� w wiadomo�ci
news:UoBdh.5467$sM2.4107@trndny05...

> Is there a general sense of preference for either (a) or (b) ?
>

I found an interesting article on the future of chess written by a
scientist. Several theses from the article are disputable :

http://ieet.org/index.php/IEET/more/dvorsky20061207/

Here is one relevant rek from the article :

"Kramnik never really threatened Fritz and lost quite soundly in game 6; he
played for the draw in virtually every match and never assumed he could
win."

I bet that Topalov would do better than Kramnik in such a match :-)

Date: 10 Dec 2006 14:30:06
From: Chess One
Subject: Re: Human vs Machine Matches

"Jerzy" <[email protected] > wrote in message
news:[email protected]...
>
> U�ytkownik "Chess One" <[email protected]> napisa� w wiadomo�ci
> news:UoBdh.5467$sM2.4107@trndny05...
>
>> Is there a general sense of preference for either (a) or (b) ?
>>
>
> I found an interesting article on the future of chess written by a
> scientist. Several theses from the article are disputable :
>
> http://ieet.org/index.php/IEET/more/dvorsky20061207/

Interesting web site, Jerzy! I'll go back and read several of the articles,
but I'll definitely write to George and find out what he really thinks, or
rather understands - but as usual, may suffer being misunderstood on what is
winning and what is learning ;)

> Here is one relevant rek from the article :
>
> "Kramnik never really threatened Fritz and lost quite soundly in game 6;
> he played for the draw in virtually every match and never assumed he could
> win."
>
> I bet that Topalov would do better than Kramnik in such a match :-)

I don't think so. I think Toppy needs to see another player in the other
chair, and this is what cost him the W CH. Kasparov needs to see flesh and
blood too, despite what he says.

Karpov would be a better opponent, IMO, and although people always say his
play is not exciting, true! its also definitely not playing for the draw,
but carefully and inexoribly making ground by small advantages in the
position, that other players don't even see.

Phil

Date: 10 Dec 2006 16:28:45
From: Jerzy
Subject: Re: Human vs Machine Matches

U�ytkownik "Chess One" <[email protected] > napisa� w wiadomo�ci
news:ORUeh.3903$IG6.1466@trndny01...

>> I bet that Topalov would do better than Kramnik in such a match :-)
>
> I don't think so. I think Toppy needs to see another player in the other
> chair, and this is what cost him the W CH. Kasparov needs to see flesh and
> blood too, despite what he says.

However Topalov has never played such a match and it would be more
spectacular than all previous matches human vs. machine.

>
> Karpov would be a better opponent, IMO, and although people always say his
> play is not exciting, true! its also definitely not playing for the draw,
> but carefully and inexoribly making ground by small advantages in the
> position, that other players don't even see.
>

Yes, once Karpov was the best person for such a match, however machines has
become stronger since then and Karpov, with all due respect, has become
weaker.

Date: 10 Jan 2007 18:54:43
From: Chess One
Subject: Re: Human vs Machine Matches

"Jerzy" <[email protected] > wrote in message
news:[email protected]...
> U�ytkownik "Chess One" <[email protected]> napisa� w wiadomo�ci
> news:ORUeh.3903$IG6.1466@trndny01...
>
>>> I bet that Topalov would do better than Kramnik in such a match :-)
>>
>> I don't think so. I think Toppy needs to see another player in the other
>> chair, and this is what cost him the W CH. Kasparov needs to see flesh
>> and blood too, despite what he says.
>
> However Topalov has never played such a match and it would be more
> spectacular than all previous matches human vs. machine.

Don't you think so! Me too. And, you know, he has the money.

>> Karpov would be a better opponent, IMO, and although people always say
>> his play is not exciting, true! its also definitely not playing for the
>> draw, but carefully and inexoribly making ground by small advantages in
>> the position, that other players don't even see.
>>
>
> Yes, once Karpov was the best person for such a match, however machines
> has become stronger since then and Karpov, with all due respect, has
> become weaker.

Agree. But against computer?

When we made that 2 hour television program there were 4 boards for US and 4
for Russia. And each player played a round with each color against same
opponent. The most interesting games were from Gulko. Making commentary on 4
games at once at this level is rather difficult for anyone, so the
interesting help came from Fritz, which ran its analysis on the bottom of
the screen, showing, eg, 1 point to be worth a pawn or three tempi. But in
the Gulko games Fritz was crazy - showing huge advantage to first one, then
the other player, then back again.

It couldn't deal with the positional evaluation of a locked-up queen on the
board in one instance, and in another a differential in material and in
initiative.

Later on the telephone with Petersburg Alex Onishuk was discussing these
games with Alex Khalifman, who also thought these were the best games, and
actually thought Gulko had wins in both his games. But in time trouble, you
know, Boris didn't find the right line. The point was that Fritz was, in
these positions, quite blind to sensible evaluation - and Khalifman is a
very fine analytical player.

Anyway - just like early-Karpov, everyone had given his games up as draws,
but then he sees something incredibly subtle and technical in the position,
and they are no longer draws - in fact as contributions to the art of chess
he introduced a kind of play which defensively balanced out Kasparov's
attacking style. Though Kasparov must still be thought to be stronger than
Karpov, nevertheless, against computer I would bet on Karpov today, since
although not as exciting as Kasparov's attacking tactical genius, Karpov's
subtle evaluations were beyond everyone - and /this/ is what I think beats
computers rather than out-gunning it.

Cordially, Phil Innes

Date: 11 Dec 2006 11:42:54
From: David Richerby
Subject: Re: Human vs Machine Matches

Jerzy <[email protected] > wrote:
> Yes, once Karpov was the best person for such a match, however
> machines has become stronger since then and Karpov, with all due
> respect, has become weaker.

That is because, since Karpov's heyday, Karpov has become older, while
computers have become newer!

Dave.

--
David Richerby Sadistic T-Shirt (TM): it's like a
www.chiark.greenend.org.uk/~davidr/ fashion statement but it wants to
hurt you!

Date: 09 Dec 2006 11:21:20
From: Guy Macon
Subject: Re: Human vs Machine Matches

MaximRecoil wrote:
>
>Guy Macon wrote:
>
>> Chess One wrote:
>>
>> >a) Human plus database, vs Computer
>>
>> >b) Human vs Computers [look-ups = off] Simul.
>>
>> >both these suggestions are deliberately different than that,
>> >since one aspect of arranging such matches is also to obtain
>> >mainstream media interest, and whatever Kramnik/Fritz was, did not!
>>
>> Allowing the human to cheat will not increase media interest.
>>
>> Crippling the computer will not increase media interest.
>>
>
>Why do you consider allowing the use of reference materials for the
>human to be "cheating", while at the same time you consider disallowing
>the use of reference materials for the computer to be "crippling"?
>
>Unless you can give good cause why there should be a double standard
>regarding the use of reference materials, then you are at odds with
>yourself.

My apologies for being unclear.

By "cheat" and cripple" I did not intend to make a value judgement
but rather to describe what the predicted media reaction would be.
I personally have no problems with changing the rules to allow
humans to use reference material, but the media will no doubt call
that cheating. In like manner the media will no doubt call anything
that reduces the computer's strength crippling the computer.
Neither is likely to increase media interest in the result of game.

Guy Macon
<http://www.guymacon.com/ >

Guy Macon <http://www.guymacon.com/ >

Date: 10 Jan 2007 15:47:13
From: Chess One
Subject: Re: Human vs Machine Matches

"Guy Macon" <http://www.guymacon.com/ > wrote in message
news:[email protected]...

>
> My apologies for being unclear.
>
> By "cheat" and cripple" I did not intend to make a value judgement
> but rather to describe what the predicted media reaction would be.
> I personally have no problems with changing the rules to allow
> humans to use reference material, but the media will no doubt call
> that cheating. In like manner the media will no doubt call anything
> that reduces the computer's strength crippling the computer.

And yet the Japanese outspend the USA on AI since emulations just don't cut
it.

Of course the computer player will be weaker, but how is it known what would
really make it stonger unless you know the consituent worth of book, engine,
cpu and bus speeds?

In terms of chess the computer does not have strength, it has emulated
strength. Applications for sophisticated emulations abound as 'expert
systems', but this is not to be confused with real strength, since if you
cheat the rules, then what real world applications are possible? Whereas -
and I know this is provocative, a weaker engine that still attempts
real-world decisions according to what it must /necessarily/ restrict itself
to, would be of massive value.

> Neither is likely to increase media interest in the result of game.

O! I would say it would be massively interesting. It would be a level
playing field, and lots of people would have a go at lots of programs, and
we would all learn very much from it.

Phil Innes

> Guy Macon
> <http://www.guymacon.com/>
>
>
>
> Guy Macon <http://www.guymacon.com/>
>
>

Date: 08 Dec 2006 20:13:32
From: MaximRecoil
Subject: Re: Human vs Machine Matches

Guy Macon wrote:
> Chess One wrote:
>
> >a) Human plus database, vs Computer
>
> >b) Human vs Computers [look-ups = off] Simul.
>
> >both these suggestions are deliberately different than that,
> >since one aspect of arranging such matches is also to obtain
> >mainstream media interest, and whatever Kramnik/Fritz was, did not!
>
> Allowing the human to cheat will not increase media interest.
>
> Crippling the computer will not increase media interest.
>

Why do you consider allowing the use of reference materials for the
human to be "cheating", while at the same time you consider disallowing
the use of reference materials for the computer to be "crippling"?

Unless you can give good cause why there should be a double standard
regarding the use of reference materials, then you are at odds with
yourself.

Date: 10 Dec 2006 22:53:53
From: Chess One
Subject: Re: Human vs Machine Matches

"MaximRecoil" <[email protected] > wrote in message
news:[email protected]...
>
> Guy Macon wrote:
>> Chess One wrote:
>>
>> >a) Human plus database, vs Computer
>>
>> >b) Human vs Computers [look-ups = off] Simul.
>>
>> >both these suggestions are deliberately different than that,
>> >since one aspect of arranging such matches is also to obtain
>> >mainstream media interest, and whatever Kramnik/Fritz was, did not!
>>
>> Allowing the human to cheat will not increase media interest.
>>
>> Crippling the computer will not increase media interest.
>>
>
> Why do you consider allowing the use of reference materials for the
> human to be "cheating", while at the same time you consider disallowing
> the use of reference materials for the computer to be "crippling"?
>
> Unless you can give good cause why there should be a double standard
> regarding the use of reference materials, then you are at odds with
> yourself.

And with Fide, and with the rating system :: since the systems can be
admixed as if they were developed in the same way.

What is often regretable about computer enthusiasts is their entire lack of
understand of non-virtual terms - themselves used only analogously among
themselves as if they were not analogies. This is depressing to logical
communication.

So you will not encounter any easy appreciation of 'what is an emulation'
more then you will find other than changing-the-rules analogies and
'Ferrari-paridigmics'.

It has been such since about 15 years ago, when processors increased in
speed to allow optimisation of winning, at the expense of all else,
including this phantasm of winning by the rules by declaring all data, or
memory, or anthropomorphic term which emulators have taken for real
equivalence.

Phil Innes

Date: 11 Dec 2006 11:41:05
From: David Richerby
Subject: Re: Human vs Machine Matches

Chess One <[email protected] > wrote:
> And with Fide, and with the rating system :: since the systems can
> be admixed as if they were developed in the same way.

What is this bizarre double-colon punctuation k you've started
using?

Dave.

--
David Richerby Homicidal Atom Bomb (TM): it's like
www.chiark.greenend.org.uk/~davidr/ a weapon of mass destruction but it
wants to kill you!

Date: 09 Dec 2006 14:22:38
From: Chris Mattern
Subject: Re: Human vs Machine Matches

In article <[email protected] >,
MaximRecoil wrote:
>
>Guy Macon wrote:
>> Chess One wrote:
>>
>> >a) Human plus database, vs Computer
>>
>> >b) Human vs Computers [look-ups = off] Simul.
>>
>> >both these suggestions are deliberately different than that,
>> >since one aspect of arranging such matches is also to obtain
>> >mainstream media interest, and whatever Kramnik/Fritz was, did not!
>>
>> Allowing the human to cheat will not increase media interest.
>>
>> Crippling the computer will not increase media interest.
>>
>
>Why do you consider allowing the use of reference materials for the
>human to be "cheating", while at the same time you consider disallowing
>the use of reference materials for the computer to be "crippling"?

Your definition of "reference materials" seems to be vague. In
this context, I would define it as "information storage external to
the subject". In this context, a computer using its hard disk is
not using "reference materials" as it is using its own internal
storage, as a human does his memory. If you have a different
definition which is useful in this context, I will be happy to
hear it and discuss it.
>
>Unless you can give good cause why there should be a double standard
>regarding the use of reference materials, then you are at odds with
>yourself.
>
I do not see any double standard, because I do not see how the computer
is using "reference materials".

--
Christopher Mattern

"Which one you figure tracked us?"
"The ugly one, sir."
"...Could you be more specific?"

Date: 10 Dec 2006 01:00:26
From: Guy Macon
Subject: Re: Human vs Machine Matches

Chris Mattern wrote:

>Your definition of "reference materials" seems to be vague. In
>this context, I would define it as "information storage external to
>the subject". In this context, a computer using its hard disk is
>not using "reference materials" as it is using its own internal
>storage, as a human does his memory.

So if I run a chess program that keeps it's opening book on an
external USB hard drive, that's use of reference materials, but
the same book on an internal SATA hard drive isn't? And if I
had an operation that put a hard drive (and some way to access
it) inside my body, I wouldn't be using reference materials, but
carrying the same hard drive (and some way to access it) in my
pocket that would be using reference materials?

That seems rather arbitrary and handcrafted to allow the computer
to access a hard drive but forbif=d the human from accessing an
identical hard drive.

Guy Macon
<http://www.guymacon.com/ >

Date: 10 Dec 2006 01:48:47
From: Chris Mattern
Subject: Re: Human vs Machine Matches

In article <[email protected] >, Guy Macon wrote:
>
>
>
>Chris Mattern wrote:
>
>>Your definition of "reference materials" seems to be vague. In
>>this context, I would define it as "information storage external to
>>the subject". In this context, a computer using its hard disk is
>>not using "reference materials" as it is using its own internal
>>storage, as a human does his memory.
>
>So if I run a chess program that keeps it's opening book on an
>external USB hard drive, that's use of reference materials, but
>the same book on an internal SATA hard drive isn't? And if I
>had an operation that put a hard drive (and some way to access
>it) inside my body, I wouldn't be using reference materials, but
>carrying the same hard drive (and some way to access it) in my
>pocket that would be using reference materials?
>
>That seems rather arbitrary and handcrafted to allow the computer
>to access a hard drive but forbif=d the human from accessing an
>identical hard drive.
>
A simple hard drive is not part of the normal construction of
a human being. It is part of the normal construction of a computer.
To an extent, this can become a rather nitpicky distinction on
the edge cases. But I submit that your definition of why
a computer can load its chess-playing program from hard disk
but not an opening line of play from that same hard disk is
at least as arbitrary and suffers from not proceding from a
clear principle. Stating that "mechanical recording devices
cannot be used" cannot be your principle if you intend to allow
computers to play chess at all, because the comptuer will not
run if it does not have access to a mechanical recording device.
Instead, you must start making distinctions of what may or may
not be stored on that device while the human is still prohibited
from using that storage device for *any* purpose. You are the
one making different rules for machines and people, not I.
And, in fact, I note that you can't consult with other people.
I can use my brain to play chess in a tournament, but you
are forbidden to access that identical brain. And so is a
computer, in fact. A chess engine's creators are never allowed
to input to the engine in the middle of a game. The engine
cannot use the exact same brain that the creator himself could
use if he was playing the game himself. So allowing the chess
to use its hard disk when the opposing player is not does not
so allowed does not in fact lead us to a contradiction.
--
Christopher Mattern

"Which one you figure tracked us?"
"The ugly one, sir."
"...Could you be more specific?"

Date: 09 Dec 2006 07:38:10
From: Kenneth Sloan
Subject: Re: Human vs Machine Matches

MaximRecoil wrote:
> Guy Macon wrote:
>> Chess One wrote:
>>
>>> a) Human plus database, vs Computer
>>> b) Human vs Computers [look-ups = off] Simul.
>>> both these suggestions are deliberately different than that,
>>> since one aspect of arranging such matches is also to obtain
>>> mainstream media interest, and whatever Kramnik/Fritz was, did not!
>> Allowing the human to cheat will not increase media interest.
>>
>> Crippling the computer will not increase media interest.
>>
>
> Why do you consider allowing the use of reference materials for the
> human to be "cheating", while at the same time you consider disallowing
> the use of reference materials for the computer to be "crippling"?
>

Look: you know, and everyone else knows, that the crux of this argument
lies in the definition of "reference material". Your postings beg the
question by assuming a particular definition. It should be clear to you
that those who disagree with you disagree on the DEFINITION of
"reference material".

> Unless you can give good cause why there should be a double standard
> regarding the use of reference materials, then you are at odds with
> yourself.
>

Until you stop begging the question and actually address the issue - you
are behaving dishonestly.

--
Kenneth Sloan [email protected]
Computer and Information Sciences +1-205-932-2213
University of Alabama at Birmingham FAX +1-205-934-5473
Birmingham, AL 35294-1170 http://www.cis.uab.edu/sloan/

Date: 11 Dec 2006 10:29:39
From: David Richerby
Subject: Re: Human vs Machine Matches

Kenneth Sloan <[email protected] > wrote:
> Look: you know, and everyone else knows, that the crux of this
> argument lies in the definition of "reference material". Your
> postings beg the question by assuming a particular definition.

I would like to thank you for the first correct use of the phrase `beg
the question' that I have seen for some time. (NB: `beg the question'
does not mean `cause the question to be asked'.)

Dave.

--
David Richerby Incredible Windows (TM): it's like
www.chiark.greenend.org.uk/~davidr/ a graphical user interface but it'll
blow your mind!

Date: 12 Dec 2006 12:10:36
From: Kenneth Sloan
Subject: Re: Human vs Machine Matches

Date: 08 Dec 2006 13:12:18
From: Rob
Subject: Re: Human vs Machine Matches

Chess One wrote:
> I would like to make a proposal, and discover your interest in two forms of
> Human v Machine matches - since I am sometimes approached by both sides to
> investigate possibilities.
>
> In conversations elsewhere [Rybka thread] there are discussions of fairness,
> so I propose two options;
>
> a) Human plus database, vs Computer
>
> b) Human vs Computers [look-ups = off] Simul.
>
> Which of these is more interesting to the chess community, and why?
>
> In the recent Kramnik Fritz engagement it was difficult to engage even the
> chess community's interest, so both these suggestions are deliberately
> different than that, since one aspect of arranging such matches is also to
> obtain mainstream media interest, and whatever Kramnik/Fritz was, did not!
>
> Is there a general sense of preference for either (a) or (b) ?
>
> Phil Innes

I wonder if a series of "solve for mate" problems being given to the
computer and the GM at the same time would not show a weakness in the
computer?

Date: 07 Dec 2006 05:34:10
From: Sanny
Subject: Play with Higher Levels.

GetClub has 5 Levels Beginner, Easy, Normal, Master and Advance.

GetClub Game at: http://www.GetClub.com/Chess.html

If you win Beginner Level you can proceed to Easy and then Higher
Levels. Each higher level is double stronger than previous level and
think twice of that.

Earlier problems of long time has been tackled and now it plays fast
and good moves.

1. Nomorechess has got Highest Rating: 1393
2. Taylor Kingston is second: 1315
3. Bob is at third position: 1190

Since the Game has been improved a lot it will play very strong and
quick games.

Bye
Sanny

Play Chess at: http://www.GetClub.com/Chess.html

Date: 07 Dec 2006 13:19:44
From: Guy Macon
Subject: Re: Human vs Machine Matches

Jimmy wrote:

>Computers don't think at all. They only do calculations based on
>algorithms.

By your definition, humans don't think at all. They only do
calculations based neurons sending electric signals to each other.

Read Minskey's _The Society of Mind_ to learn about the algorithms
running in your brain.

Date: 07 Dec 2006 13:09:11
From: Guy Macon
Subject: Re: Human vs Machine Matches

Chess One wrote:

>a) Human plus database, vs Computer

>b) Human vs Computers [look-ups = off] Simul.

>both these suggestions are deliberately different than that,
>since one aspect of arranging such matches is also to obtain
>mainstream media interest, and whatever Kramnik/Fritz was, did not!

Allowing the human to cheat will not increase media interest.

Crippling the computer will not increase media interest.

>Is there a general sense of preference for either (a) or (b) ?

http://en.wikipedia.org/wiki/False_dilemma
http://en.wikipedia.org/wiki/Logical_fallacy

Guy Macon
<http://www.guymacon.com/ >

Date: 07 Dec 2006 15:35:40
From: Chess One
Subject: Re: Human vs Machine Matches

"Guy Macon" <http://www.guymacon.com/ > wrote in message
news:[email protected]...
>
>
>
> Chess One wrote:
>
>>a) Human plus database, vs Computer
>
>>b) Human vs Computers [look-ups = off] Simul.
>
>>both these suggestions are deliberately different than that,
>>since one aspect of arranging such matches is also to obtain
>>mainstream media interest, and whatever Kramnik/Fritz was, did not!
>
> Allowing the human to cheat will not increase media interest.
>
> Crippling the computer will not increase media interest.

Why is it that people who are into chess programming talk about legal chess
as cheating, while suggesting changing the rules of the game for the
computer - to pose yet another another 'Ferrari' argument?

They therefore set up a false dichotomy of what is envisaged, which is a
match with legal chess configuration which is characterised as 'cheating',
contrasted with their own proposal for Advanced Chess, which is
with-Ferrari.

But for the life of them cannot speak of relating to the rules of legal
chess - a taboo subject - since it would 'cripple' their engine, but do not
make the parralel human to human statement that if a person cheats by
looking-up reference material, they would be disqualified.

The game axiomatically is itself nothing but a collection of rules of play -
why continue to miserably avoid that distinction, and villify those who make
it?

This insensible insistence fudges several factors, not the least of which is
the value of the book and end-tables to programmers, since they are too shy
to test their engines against the legal rating pool to establish whatever
strength is added, and to large extent shoot themselves in the foot by not
knowing if they are worth 50 points of strength or 350.

That is not exactly a scientific approach to programming. An idea already
stated here is that a test /would/ establish that strength with
look-ups=off. No takers?

Phil Innes

>>Is there a general sense of preference for either (a) or (b) ?
>
> http://en.wikipedia.org/wiki/False_dilemma
> http://en.wikipedia.org/wiki/Logical_fallacy
>
> Guy Macon
> <http://www.guymacon.com/>
>

Date: 10 Jan 2007 15:00:17
From: Chess One
Subject: Re: Human vs Machine Matches

I have a possible sponsor for the test I suggest below - and though there is
player interest, there is no programmer interest :(

Phil Innes

"Chess One" <[email protected] > wrote in message
news:gxWdh.3114$QD3.2381@trndny01...
>
> "Guy Macon" <http://www.guymacon.com/> wrote in message
> news:[email protected]...
>>
>>
>>
>> Chess One wrote:
>>
>>>a) Human plus database, vs Computer
>>
>>>b) Human vs Computers [look-ups = off] Simul.
>>
>>>both these suggestions are deliberately different than that,
>>>since one aspect of arranging such matches is also to obtain
>>>mainstream media interest, and whatever Kramnik/Fritz was, did not!
>>
>> Allowing the human to cheat will not increase media interest.
>>
>> Crippling the computer will not increase media interest.
>
> Why is it that people who are into chess programming talk about legal
> chess as cheating, while suggesting changing the rules of the game for the
> computer - to pose yet another another 'Ferrari' argument?
>
> They therefore set up a false dichotomy of what is envisaged, which is a
> match with legal chess configuration which is characterised as 'cheating',
> contrasted with their own proposal for Advanced Chess, which is
> with-Ferrari.
>
> But for the life of them cannot speak of relating to the rules of legal
> chess - a taboo subject - since it would 'cripple' their engine, but do
> not make the parralel human to human statement that if a person cheats by
> looking-up reference material, they would be disqualified.
>
> The game axiomatically is itself nothing but a collection of rules of
> play - why continue to miserably avoid that distinction, and villify those
> who make it?
>
> This insensible insistence fudges several factors, not the least of which
> is the value of the book and end-tables to programmers, since they are too
> shy to test their engines against the legal rating pool to establish
> whatever strength is added, and to large extent shoot themselves in the
> foot by not knowing if they are worth 50 points of strength or 350.
>
> That is not exactly a scientific approach to programming. An idea already
> stated here is that a test /would/ establish that strength with
> look-ups=off. No takers?
>
> Phil Innes
>
>>>Is there a general sense of preference for either (a) or (b) ?
>>
>> http://en.wikipedia.org/wiki/False_dilemma
>> http://en.wikipedia.org/wiki/Logical_fallacy
>>
>> Guy Macon
>> <http://www.guymacon.com/>
>>
>
>

Date: 10 Jan 2007 10:19:34
From: Kenneth Sloan
Subject: Re: Human vs Machine Matches

Chess One wrote:
> I have a possible sponsor for the test I suggest below - and though there is
> player interest, there is no programmer interest :(
>
> Phil Innes
>

I have possible programmer interest, but no interested sponsor.

--
Kenneth Sloan [email protected]
Computer and Information Sciences +1-205-932-2213
University of Alabama at Birmingham FAX +1-205-934-5473
Birmingham, AL 35294-1170 http://www.cis.uab.edu/sloan/

Date: 10 Jan 2007 17:38:53
From: Chess One
Subject: Re: Human vs Machine Matches

"Kenneth Sloan" <[email protected] > wrote in message
news:[email protected]...
> Chess One wrote:
>> I have a possible sponsor for the test I suggest below - and though there
>> is player interest, there is no programmer interest :(
>>
>> Phil Innes
>>
>
> I have possible programmer interest, but no interested sponsor.

Are you serious? PI

> --
> Kenneth Sloan [email protected]
> Computer and Information Sciences +1-205-932-2213
> University of Alabama at Birmingham FAX +1-205-934-5473
> Birmingham, AL 35294-1170 http://www.cis.uab.edu/sloan/

Date: 10 Jan 2007 15:45:41
From: Kenneth Sloan
Subject: Re: Human vs Machine Matches

Chess One wrote:
> "Kenneth Sloan" <[email protected]> wrote in message
> news:[email protected]...
>> Chess One wrote:
>>> I have a possible sponsor for the test I suggest below - and though there
>>> is player interest, there is no programmer interest :(
>>>
>>> Phil Innes
>>>
>> I have possible programmer interest, but no interested sponsor.
>
> Are you serious? PI
Are you?

--
Kenneth Sloan [email protected]
Computer and Information Sciences +1-205-932-2213
University of Alabama at Birmingham FAX +1-205-934-5473
Birmingham, AL 35294-1170 http://www.cis.uab.edu/sloan/

Date: 10 Jan 2007 22:07:13
From: Chess One
Subject: Re: Human vs Machine Matches

"Kenneth Sloan" <[email protected] > wrote in message
news:[email protected]...
> Chess One wrote:
>> "Kenneth Sloan" <[email protected]> wrote in message
>> news:[email protected]...
>>> Chess One wrote:
>>>> I have a possible sponsor for the test I suggest below - and though
>>>> there is player interest, there is no programmer interest :(
>>>>
>>>> Phil Innes
>>>>
>>> I have possible programmer interest, but no interested sponsor.
>>
>> Are you serious? PI
> Are you?

ken, you have something of a reputation for quixotic reks. if you have
serious interest see if you can figure out what to do all by yourself.

pi

>
> --
> Kenneth Sloan [email protected]
> Computer and Information Sciences +1-205-932-2213
> University of Alabama at Birmingham FAX +1-205-934-5473
> Birmingham, AL 35294-1170 http://www.cis.uab.edu/sloan/

Date: 10 Jan 2007 16:14:04
From: Kenneth Sloan
Subject: Re: Human vs Machine Matches

Chess One wrote:
> "Kenneth Sloan" <[email protected]> wrote in message
> news:[email protected]...
>> Chess One wrote:
>>> "Kenneth Sloan" <[email protected]> wrote in message
>>> news:[email protected]...
>>>> Chess One wrote:
>>>>> I have a possible sponsor for the test I suggest below - and though
>>>>> there is player interest, there is no programmer interest :(
>>>>>
>>>>> Phil Innes
>>>>>
>>>> I have possible programmer interest, but no interested sponsor.
>>> Are you serious? PI
>> Are you?
>
> ken, you have something of a reputation for quixotic reks. if you have
> serious interest see if you can figure out what to do all by yourself.
>
yes, I can.

I wonder if *you* can?

--
Kenneth Sloan [email protected]
Computer and Information Sciences +1-205-932-2213
University of Alabama at Birmingham FAX +1-205-934-5473
Birmingham, AL 35294-1170 http://www.cis.uab.edu/sloan/

Date: 06 Dec 2006 14:40:15
From: Jimmy
Subject: Re: Human vs Machine Matches

Thomas T. Veldhouse wrote:
> In rec.games.chess.computer Jimmy <[email protected]> wrote:
> >
> > Computers don't think at all. They only do calculations based on
> > algorithms.
> >
>
> Of course they don't think. It is a convenient metaphor though.
>
> > Speak for yourself. I try to understand why I lost so that I don't
> > make the same mistake again.
> >
>
> I was speaking in a very generalized way, obviously.
>
> >
> > It is apples and oranges. The analysis required to calculate chess
> > variations is an order of magnitude problem. The permutations are
> > beyond the power of today's computers to examine all the possibilities.
> > Consequently, there has to be some "pruning of the branches". This is
> > done by evaluating each position to determine if the position is
> > favorable or not. This is the crux of the problem for programmers.
> >
>
> I didn't say there wasn't pruning, I said there was pruning based upon pattern
> matching in the sense that a human brain does it.
>
I don't understand why you said pattern matching. Basically the
computer assigns a numeric score to a position based on a formula
criteria. There could be a database which has endgame positions to
match to, but the majority of the positions generate a score. The
formula would include points for having more pieces, better mobility,
more space, better placement of pieces, more squares controlled, passed
pawns, etc. There really isn't pattern matching.

I wrote a program to do this type of position analysis for a Checker
playing program for a AI class in Berkeley.

The program ran on a standard digital computer(UNIX Solaris multi-proc
box) - neural networks or computing using fundamentally different
logic. This type of computing could do some form of pattern matching
logic.

Jimmy Boy

Date: 07 Dec 2006 14:07:55
From: Thomas T. Veldhouse
Subject: Re: Human vs Machine Matches

In rec.games.chess.computer Jimmy <[email protected] > wrote:
>>
> I don't understand why you said pattern matching. Basically the
> computer assigns a numeric score to a position based on a formula
> criteria. There could be a database which has endgame positions to
> match to, but the majority of the positions generate a score. The
> formula would include points for having more pieces, better mobility,
> more space, better placement of pieces, more squares controlled, passed
> pawns, etc. There really isn't pattern matching.
>

Humans think symbollically and they do this by pattern matching [based upon
repetitively reinforced memory]. It is HUMANS that I said do pattern
matching, not the computers [unless they use neural networks or a similar
technology].

> I wrote a program to do this type of position analysis for a Checker
> playing program for a AI class in Berkeley.
>
> The program ran on a standard digital computer(UNIX Solaris multi-proc
> box) - neural networks or computing using fundamentally different
> logic. This type of computing could do some form of pattern matching
> logic.
>

That is what I said ... neural networking.

--
Thomas T. Veldhouse
Key Fingerprint: D281 77A5 63EE 82C5 5E68 00E4 7868 0ADC 4EFB 39F0

Date: 07 Dec 2006 14:26:55
From: Guy Macon
Subject: Re: Human vs Machine Matches

Thomas T. Veldhouse wrote:

>Humans think symbollically and they do this by pattern matching [based upon
>repetitively reinforced memory]. It is HUMANS that I said do pattern
>matching, not the computers [unless they use neural networks or a similar
>technology].

In most cases, the neural network is simply another computer program.

Date: 07 Dec 2006 14:52:51
From: Thomas T. Veldhouse
Subject: Re: Human vs Machine Matches

In rec.games.chess.computer Guy Macon <http://www.guymacon.com/ > wrote:
>
>
>
> Thomas T. Veldhouse wrote:
>
>>Humans think symbollically and they do this by pattern matching [based upon
>>repetitively reinforced memory]. It is HUMANS that I said do pattern
>>matching, not the computers [unless they use neural networks or a similar
>>technology].
>
> In most cases, the neural network is simply another computer program.
>

I know what a neural network is. In the human case, it clearly is NOT a
computer program. If it is, then our reality is really quite subjective, eh?

--
Thomas T. Veldhouse
Key Fingerprint: D281 77A5 63EE 82C5 5E68 00E4 7868 0ADC 4EFB 39F0

Date: 08 Dec 2006 14:32:23
From: Guy Macon
Subject: Re: Human vs Machine Matches

Thomas T. Veldhouse wrote:
>
>Guy Macon <http://www.guymacon.com/> wrote:
>>
>> In most cases, the neural network is simply another computer program.
>
>I know what a neural network is. In the human case, it clearly is NOT a
>computer program.

That is not clear at all A brain is a type of computer hby any
objective definition not handcrafted to exclude brains.

"Clearly" often parses to "without evidence, I assume that"...

Guy Macon
<http://www.guymacon.com/ >

Date: 08 Dec 2006 15:02:08
From: Thomas T. Veldhouse
Subject: Re: Human vs Machine Matches

In rec.games.chess.computer Guy Macon <http://www.guymacon.com/ > wrote:
>
>
> Thomas T. Veldhouse wrote:
>>
>>Guy Macon <http://www.guymacon.com/> wrote:
>>>
>>> In most cases, the neural network is simply another computer program.
>>
>>I know what a neural network is. In the human case, it clearly is NOT a
>>computer program.
>
> That is not clear at all A brain is a type of computer hby any
> objective definition not handcrafted to exclude brains.

I guess that depends how you loosely define a computer. I don't believe a
human brain "computes" at all. It simply follows the path of least resistance
through the brain based upon trained neural pathways.

>
> "Clearly" often parses to "without evidence, I assume that"...
>

Then ... I wouldn't use "clearly" if I were you.

--
Thomas T. Veldhouse
Key Fingerprint: D281 77A5 63EE 82C5 5E68 00E4 7868 0ADC 4EFB 39F0

Date: 07 Dec 2006 13:10:50
From: Kenneth Sloan
Subject: Re: Human vs Machine Matches

Thomas T. Veldhouse wrote:
> In rec.games.chess.computer Guy Macon <http://www.guymacon.com/> wrote:
>>
>>
>> Thomas T. Veldhouse wrote:
>>
>>> Humans think symbollically and they do this by pattern matching [based upon
>>> repetitively reinforced memory]. It is HUMANS that I said do pattern
>>> matching, not the computers [unless they use neural networks or a similar
>>> technology].
>> In most cases, the neural network is simply another computer program.
>>
>
> I know what a neural network is. In the human case, it clearly is NOT a
> computer program. If it is, then our reality is really quite subjective, eh?
>

It would appear to be.

--
Kenneth Sloan [email protected]
Computer and Information Sciences +1-205-932-2213
University of Alabama at Birmingham FAX +1-205-934-5473
Birmingham, AL 35294-1170 http://www.cis.uab.edu/sloan/

Date: 06 Dec 2006 12:37:19
From: Jimmy
Subject: Re: Human vs Machine Matches

Dave (from the UK) wrote:
> Thomas T. Veldhouse wrote:
>
> > Computer only gets opening book and tablebases. Both of which humans are
> > capable of learning/memorizing.
> >
> > Humans once had an advantage that we could prune the options down in a
> > reliable way. That is proving to no longer be the advantage. Had we still
> > have it ... I would think somebody should be arguing that humans not be able
> > to do this ... wait, you can't take that away from a human. Same thing with a
> > computer. A computer has photographic memory and thinks REALLY REALLY FAST.
> > That is ITS competitive advanatage. It has finally proven to be overwhelming.
> >

Computers don't think at all. They only do calculations based on
algorithms.

> > What I don't get is why, when humans lose, they try to find away to invalidate
> > and justify their loss such that it hasn't really occurred. I guess humans
> > are fundamentally lawyers at heart ... all of us.
> >

Speak for yourself. I try to understand why I lost so that I don't
make the same mistake again.

>
> It can really be no surprise the computer has won - it was expected to
> happen one day. There is no way I could do maths as fast as a computer
> or run as fast as a car.
>
> My computer, with a high end maths package and a mere 450 MHz CPU can
> compute exactly using just integers 10000! in 32 ms (FWIW. a floating
> point approximation is about 2.846259680917055 x 10^35659).
>
> It would take me a very long time to compute 10000! and write down all
> 35660 digits by hand and the chances of an error would be extreamly
> high. (Actually, it is quite possible it would take me longer than my
> lifetime to do it.)
>
> But I should not feel threatened. For a long time man has built tools to
> do jobs faster / better than he can do. Chess can't be expected to be
> any different really.
>

It is apples and oranges. The analysis required to calculate chess
variations is an order of magnitude problem. The permutations are
beyond the power of today's computers to examine all the possibilities.
Consequently, there has to be some "pruning of the branches". This is
done by evaluating each position to determine if the position is
favorable or not. This is the crux of the problem for programmers.

On another note, computers that are "booked up" should be easy for
players to play against. You don't have to worry about surprise moves.
Any innovation should take a machine by surprise. Also, if you have a
copy of Fritz at home.....

Jimmy Boy

Date: 06 Dec 2006 22:29:48
From: Kenneth Sloan
Subject: Re: Human vs Machine Matches

Jimmy wrote:
> Dave (from the UK) wrote:
>> Thomas T. Veldhouse wrote:
>>
>>> Computer only gets opening book and tablebases. Both of which humans are
>>> capable of learning/memorizing.
>>>
>>> Humans once had an advantage that we could prune the options down in a
>>> reliable way. That is proving to no longer be the advantage. Had we still
>>> have it ... I would think somebody should be arguing that humans not be able
>>> to do this ... wait, you can't take that away from a human. Same thing with a
>>> computer. A computer has photographic memory and thinks REALLY REALLY FAST.
>>> That is ITS competitive advanatage. It has finally proven to be overwhelming.
>>>
>
> Computers don't think at all. They only do calculations based on
> algorithms.

And what, precisely, do humans do?

Is it the fact that you don't know that makes it "thinking"?

--
Kenneth Sloan [email protected]
Computer and Information Sciences +1-205-932-2213
University of Alabama at Birmingham FAX +1-205-934-5473
Birmingham, AL 35294-1170 http://www.cis.uab.edu/sloan/

Date: 06 Dec 2006 20:58:05
From: Thomas T. Veldhouse
Subject: Re: Human vs Machine Matches

In rec.games.chess.computer Jimmy <[email protected] > wrote:
>
> Computers don't think at all. They only do calculations based on
> algorithms.
>

Of course they don't think. It is a convenient metaphor though.

> Speak for yourself. I try to understand why I lost so that I don't
> make the same mistake again.
>

I was speaking in a very generalized way, obviously.

>
> It is apples and oranges. The analysis required to calculate chess
> variations is an order of magnitude problem. The permutations are
> beyond the power of today's computers to examine all the possibilities.
> Consequently, there has to be some "pruning of the branches". This is
> done by evaluating each position to determine if the position is
> favorable or not. This is the crux of the problem for programmers.
>

I didn't say there wasn't pruning, I said there was pruning based upon pattern
matching in the sense that a human brain does it.

Speaking of .. I wonder how much research has been done with Chess and neural
networks.

--
Thomas T. Veldhouse
Key Fingerprint: D281 77A5 63EE 82C5 5E68 00E4 7868 0ADC 4EFB 39F0

Date: 07 Dec 2006 12:37:50
From: Chess One
Subject: Re: Human vs Machine Matches

"Thomas T. Veldhouse" <[email protected] > wrote in message
news:[email protected]...
> In rec.games.chess.computer Jimmy <[email protected]> wrote:

> Speaking of .. I wonder how much research has been done with Chess and
> neural
> networks.

Not much. Try researching on Univ Sussex in UK. The problem is integrating
data which is patterned in a way that is comprehensible or of further use.
What some scientists have tried to do is to emulate nature by creating
virtual 'organelles', and at U. Sussex they didn't use chess, they partnered
with biologists to do it - and some 10-15 years ago began studying insects
and the integration of dynamic complex patterning. That is; they were
looking at ants and bees behavior.

Most breakthroughs to new paradigms now rely on multi-disciplinary
approachs. Those people using chaos-math mapping are finding patterns all
over the place, but the difficulty is in understanding what the patterns
mean since the math is just the map, so to speak, and their conclusions are
ab stracta, rather than in vivo.

Chess research became divorced from any real AI study, also about 10 years
ago, since again, the emphasis in AI is patterning and learning, [not data
as discreet bits of information, but relationship of bits of data to each
other] which is a more qualitative factor - and chess programmers are into
engine strength and optimising quantitative factors emphasising speed and
ratings.

A few months ago I found a student attempting a PhD thesis in chess, someone
who already had a master's in robotics. There were two ideas he had for his
doctorate, one to make a unifyied platform for development, which was a
relatively easy thing to, at least, conect him with hardware and software
developers in chess. But the other was to look at learning systems with
chess as its modus, with an additional concentration on unsighted people,
and those with dissabilities.

Connecting the computing side of this was at least easy to provide [if not
to do] and everyone I asked volunteered their interest - but the application
side was very difficult indeed to support from any precedent. There is a
dearth of studies on the sociology or psychology of chess - and much that is
represented as learning process, and result, is not born out by actual
study, and tending more to hyperbolous claims.

At a meta~ level the programmers have gone commercial by optimising what
they do to creation of product and direct supply to the ket. And the
psycho-sociologic realm of understanding learning is also commercial in that
it has gone whoring after mainstream educational budgets.

While Human vs Machine matches could be interesting media events - if they
are to be something more, it is still necessary to describe a process that
we would actually learn from. I think that is both the challenge to
social-science, and to computing-science.

Phil Innes

Phil Innes

> --
> Thomas T. Veldhouse
> Key Fingerprint: D281 77A5 63EE 82C5 5E68 00E4 7868 0ADC 4EFB 39F0
>
>

Date: 06 Dec 2006 17:18:24
From: Jerzy
Subject: Re: Human vs Machine Matches

U�ytkownik "Chess One" <[email protected] > napisa� w wiadomo�ci
news:UoBdh.5467$sM2.4107@trndny05...
>I would like to make a proposal, and discover your interest in two forms of
>Human v Machine matches - since I am sometimes approached by both sides to
>investigate possibilities.
>
> In conversations elsewhere [Rybka thread] there are discussions of
> fairness, so I propose two options;
>
> a) Human plus database, vs Computer
>
> b) Human vs Computers [look-ups = off] Simul.
>
> Which of these is more interesting to the chess community, and why?
>

I would add

c) freestyle competition humans aided with comps vs humans aided with comps.

I think it`s an interesting option that is being tested nowadays e.g. in
e-mail chess :-) but of course top-players including Topalov or Polgar also
play this kind of chess. It is more spectacular than e.g. matches Kramnik -
comps.

> In the recent Kramnik Fritz engagement it was difficult to engage even the
> chess community's interest, so both these suggestions are deliberately
> different than that, since one aspect of arranging such matches is also to
> obtain mainstream media interest, and whatever Kramnik/Fritz was, did not!
>
> Is there a general sense of preference for either (a) or (b) ?

I think there is a general sense of preference for (c) :-)

Date: 06 Dec 2006 16:07:54
From: Thomas T. Veldhouse
Subject: Re: Human vs Machine Matches

In rec.games.chess.computer Chess One <[email protected] > wrote:
> I would like to make a proposal, and discover your interest in two forms of
> Human v Machine matches - since I am sometimes approached by both sides to
> investigate possibilities.
>
> In conversations elsewhere [Rybka thread] there are discussions of fairness,
> so I propose two options;
>
> a) Human plus database, vs Computer

Computer only gets opening book and tablebases. Both of which humans are
capable of learning/memorizing.

Humans once had an advantage that we could prune the options down in a
reliable way. That is proving to no longer be the advantage. Had we still
have it ... I would think somebody should be arguing that humans not be able
to do this ... wait, you can't take that away from a human. Same thing with a
computer. A computer has photographic memory and thinks REALLY REALLY FAST.
That is ITS competitive advanatage. It has finally proven to be overwhelming.

What I don't get is why, when humans lose, they try to find away to invalidate
and justify their loss such that it hasn't really occurred. I guess humans
are fundamentally lawyers at heart ... all of us.

--
Thomas T. Veldhouse
Key Fingerprint: D281 77A5 63EE 82C5 5E68 00E4 7868 0ADC 4EFB 39F0

Date: 07 Dec 2006 13:13:47
From: Guy Macon
Subject: Re: Human vs Machine Matches

Thomas T. Veldhouse wrote:

>Computer only gets opening book and tablebases.
>Both of which humans are capable of learning/memorizing.

Would it be reasonable to allow the human to bring a printed
opening book and set of tablebases? Or keep them on a hard
drive with a simple program that accesses them but cannot
calculate? The computer, after all, has such data storage.

Date: 07 Dec 2006 13:12:17
From: Kenneth Sloan
Subject: Re: Human vs Machine Matches

Guy Macon wrote:
> Thomas T. Veldhouse wrote:
>
>> Computer only gets opening book and tablebases.
>> Both of which humans are capable of learning/memorizing.
>
> Would it be reasonable to allow the human to bring a printed
> opening book and set of tablebases? Or keep them on a hard
> drive with a simple program that accesses them but cannot
> calculate? The computer, after all, has such data storage.
>

No, the computer *is* such data storage.

--
Kenneth Sloan [email protected]
Computer and Information Sciences +1-205-932-2213
University of Alabama at Birmingham FAX +1-205-934-5473
Birmingham, AL 35294-1170 http://www.cis.uab.edu/sloan/

Date: 11 Dec 2006 10:27:24
From: David Richerby
Subject: Re: Human vs Machine Matches

Kenneth Sloan <[email protected] > wrote:
> Guy Macon wrote:
>> The computer, after all, has such data storage.
>
> No, the computer *is* such data storage.

But the network is the computer. ;-)

Dave.

--
David Richerby Addictive Crystal Hat (TM): it's like
www.chiark.greenend.org.uk/~davidr/ a hat but it's completely transparent
and you can never put it down!

Date: 12 Dec 2006 12:09:28
From: Kenneth Sloan
Subject: Re: Human vs Machine Matches

David Richerby wrote:
> Kenneth Sloan <[email protected]> wrote:
>> Guy Macon wrote:
>>> The computer, after all, has such data storage.
>> No, the computer *is* such data storage.
>
> But the network is the computer. ;-)
>
>
> Dave.
>
Clearly, you missed the retraction t-shirt:

"The network is the network; the computer is the computer. We apologize
for any confusion" - Sun t-shirt first seen at SIGGRAPH ?? (one year
after the "the network is the computer" t-shirt.

--
Kenneth Sloan [email protected]
Computer and Information Sciences +1-205-932-2213
University of Alabama at Birmingham FAX +1-205-934-5473
Birmingham, AL 35294-1170 http://www.cis.uab.edu/sloan/

Date: 08 Dec 2006 14:27:38
From: Chess One
Subject: Re: Human vs Machine Matches

"Kenneth Sloan" <[email protected] > wrote in message
news:[email protected]...
> Guy Macon wrote:
>> Thomas T. Veldhouse wrote:
>>
>>> Computer only gets opening book and tablebases. Both of which humans
>>> are capable of learning/memorizing.
>>
>> Would it be reasonable to allow the human to bring a printed opening book
>> and set of tablebases? Or keep them on a hard drive with a simple
>> program that accesses them but cannot calculate? The computer, after
>> all, has such data storage.
>>
>
> No, the computer *is* such data storage.

It seems appropriate to ask some advanced ideas, yet elementary factors, and
to question people's knowledge of computing and logic:-

Can anyone tell me if they understand:-

1) what is a Turing Engine?
2) can a Turing Engine be made of paper, rather than requiring electricity
or mechanical components?
3) can anyone distinguish between instruction sets which are logic trees,
and other forms of data sets?
4) what is similar with all Turing Engines?
5) what is the only factor which differentiates all Turing Engines?
6) if a book is described as a data-set, and a computer is a data-set, is a
book a computer?
7) what is computed in a book?
8) what part of a computer is not data?
9) what part of a Turing Engine is not data?
10) of what was the first Turing Engine constucted?

Perhaps when we examine our knowledge of this subject, we can then assess
our understanding of people.

Phil Innes

> --
> Kenneth Sloan [email protected]
> Computer and Information Sciences +1-205-932-2213
> University of Alabama at Birmingham FAX +1-205-934-5473
> Birmingham, AL 35294-1170 http://www.cis.uab.edu/sloan/

Date: 09 Dec 2006 13:25:06
From: Chris Mattern
Subject: Re: Human vs Machine Matches

In article <uDeeh.4971$ne3.4901@trndny03 >, Chess One wrote:
>
>"Kenneth Sloan" <[email protected]> wrote in message
>news:[email protected]...
>> Guy Macon wrote:
>>> Thomas T. Veldhouse wrote:
>>>
>>>> Computer only gets opening book and tablebases. Both of which humans
>>>> are capable of learning/memorizing.
>>>
>>> Would it be reasonable to allow the human to bring a printed opening book
>>> and set of tablebases? Or keep them on a hard drive with a simple
>>> program that accesses them but cannot calculate? The computer, after
>>> all, has such data storage.
>>>
>>
>> No, the computer *is* such data storage.

Such data storage does not solely comprise a computer, however, it is
an integral part of it.
>
>It seems appropriate to ask some advanced ideas, yet elementary factors, and
>to question people's knowledge of computing and logic:-
>
>Can anyone tell me if they understand:-
>
>1) what is a Turing Engine?

A Turing Engine is a theoretical construct postulated by Alan Turing.
It can be shown that it is mathematically equivalent to all modern
digital computing machines. Wikipedia contains a good description
of it.

>2) can a Turing Engine be made of paper, rather than requiring electricity
>or mechanical components?

No. The tape (storage, or memory) can be, but the "head", or processing
element cannot be.

>3) can anyone distinguish between instruction sets which are logic trees,
>and other forms of data sets?

I'm not sure I understand this question. How much information is being
given? You can't generally distinguish *any* kind of binary information
without the key as to how it is to be interpreted. What information are
you being given to try to make the difference?

>4) what is similar with all Turing Engines?

They all include data storage, or memory. (The tape, in the prototypical
example), and they all include a processing element (the "head"), supplied
with addition memory containing the action table and the state register.

>5) what is the only factor which differentiates all Turing Engines?

I don't understand this question. Do you mean a factor that differentiates
Turing Engines from each other, or differentiates Turing Engines from
something else? If the first, the question makes no sense, because many
factors differentiate Turing Engines from each other--the size of their
memory, the speed of their processing, and on and on. If the latter,
I cannot answer unless you clarify what we are differentiating Turing
Engines *from*.

>6) if a book is described as a data-set, and a computer is a data-set, is a
>book a computer?

No, of course not. A computer is *not* a data-set, as I stated above.
However, it *contains* a data-set, or data-sets, as part of itself.

>7) what is computed in a book?

Nothing is computed in a book, of course. That is the difference between
a book and computer, or a Turing Machine.

>8) what part of a computer is not data?

The processing unit.

>9) what part of a Turing Engine is not data?

The "head" that can read and write data.

>10) of what was the first Turing Engine constucted?

Of paper and a human, with the human providing the action of the "head".
>
>Perhaps when we examine our knowledge of this subject, we can then assess
>our understanding of people.
>
>
As we can see from looking at the Turing Machine, mechanical storage is an
integral part of it, as it is of the the computers that implement various
forms of its mathematical equivalents. Therefore, prohibiting computers
from using mechanical storage to perform any function (such as playing a
game of chess) is equivalent to prohibiting them from operating at all.

--
Christopher Mattern

"Which one you figure tracked us?"
"The ugly one, sir."
"...Could you be more specific?"

Date: 10 Jan 2007 19:28:14
From: Chess One
Subject: !] Re: Human vs Machine Matches

"Chris Mattern" <[email protected] > wrote in message
news:[email protected]...
> In article <uDeeh.4971$ne3.4901@trndny03>, Chess One wrote:

>>Can anyone tell me if they understand:-
>>
>>1) what is a Turing Engine?
>
> A Turing Engine is a theoretical construct postulated by Alan Turing.
> It can be shown that it is mathematically equivalent to all modern
> digital computing machines. Wikipedia contains a good description
> of it.

True!

>>2) can a Turing Engine be made of paper, rather than requiring electricity
>>or mechanical components?

False! Turing's first engine was a paper one. And so is your logical
flowchart Chris.

> No. The tape (storage, or memory) can be, but the "head", or processing
> element cannot be.

That refers to the mode of application, or machination of the engine.

>>3) can anyone distinguish between instruction sets which are logic trees,
>>and other forms of data sets?
>
> I'm not sure I understand this question. How much information is being
> given? You can't generally distinguish *any* kind of binary information
> without the key as to how it is to be interpreted. What information are
> you being given to try to make the difference?

What is an executive file and what is a data file?

>>4) what is similar with all Turing Engines?
>
> They all include data storage, or memory. (The tape, in the prototypical
> example), and they all include a processing element (the "head"), supplied
> with addition memory containing the action table and the state register.

Sorry, that is the condition of a machine. What is similar of all engines is
that they can all complete the same task.

>>5) what is the only factor which differentiates all Turing Engines?
>
> I don't understand this question. Do you mean a factor that
> differentiates
> Turing Engines from each other, or differentiates Turing Engines from
> something else?

From each other

> If the first, the question makes no sense, because many
> factors differentiate Turing Engines from each other--the size of their
> memory, the speed of their processing, and on and on.

Therfore the answer is time

> If the latter,
> I cannot answer unless you clarify what we are differentiating Turing
> Engines *from*.

Right

>>6) if a book is described as a data-set, and a computer is a data-set, is
>>a
>>book a computer?
>
> No, of course not. A computer is *not* a data-set, as I stated above.
> However, it *contains* a data-set, or data-sets, as part of itself.

Yes. A book is a set of data, and a potential component of a computational
program.

>>7) what is computed in a book?
>
> Nothing is computed in a book, of course. That is the difference between
> a book and computer, or a Turing Machine.

Its an interesting way to describe these two factors of 'engine' which
scientifically is the same as Babbage's engine, and machine. For example the
book can contain /all/ the instruction sets of the machine, yet of itself
does not execute the instructions it contains.

>>8) what part of a computer is not data?
>
> The processing unit.

Fascinating isn't it, that the /processing/ is described by data, but the
processor is not data. In a range finding application I worked on a project
where we threw away the processor, and proceded with it - whether this
device can still be called a computer I don't know, since nothing is
computed, and much as it is processed without a bus. [A real-time optical
device]

>>9) what part of a Turing Engine is not data?
>
> The "head" that can read and write data.

No machine is implicated or necessary - though certainly useful! Consider
deploying your program on an abacus - an interesting device philosophically
being simultaneously a record of data and a calculating machine. So
unfortunately that answer is not correct, and all Engine is data.

>>10) of what was the first Turing Engine constucted?
>
> Of paper and a human, with the human providing the action of the "head".

Okay.

>>Perhaps when we examine our knowledge of this subject, we can then assess
>>our understanding of people.
>>
>>
> As we can see from looking at the Turing Machine, mechanical storage is an
> integral part of it,

Yes, it is usually an integral and convenient part of it. But in
consideration of robots, there is often no need to store information [except
as an expedient to buffer speed of processing] and input is not recorded
[since storage has no application purpose] but is immeditely the stimulus
for electro-mechanical instruction, ie move left or move right.

> as it is of the the computers that implement various
> forms of its mathematical equivalents. Therefore, prohibiting computers
> from using mechanical storage to perform any function (such as playing a
> game of chess) is equivalent to prohibiting them from operating at all.

It is not computers who are inhibited from doing anything, but people who
are. People are required to perform actions according to rules described and
understood by them as people, and it is an immoderate sense to dissapear
this link when in fact it is people who deliberately construct ever single
aspect of their engine and machine.

Since in the above we can see the flaw in the argument that if a data-set is
read by a person, then it is clearly illegal under the rules of chess -
yes - but if deployed by a programmer [not a machine!] it is just as illegal
as putting one of those 'things' in your hat.

Thank you though for suffering my questions. I am not the least
anti-computer. I simply care to distinguish emulation paradigms from real
applications, the same way I must distinguish 'Disney' bears from real ones.

Cordially, Phil Innes

> --
> Christopher Mattern
>
> "Which one you figure tracked us?"
> "The ugly one, sir."
> "...Could you be more specific?"

Date: 10 Jan 2007 15:50:53
From: Kenneth Sloan
Subject: Re: !] Re: Human vs Machine Matches

Chess One wrote:
> "Chris Mattern" <[email protected]> wrote in message
> news:[email protected]...
>> In article <uDeeh.4971$ne3.4901@trndny03>, Chess One wrote:
>
>
>>> Can anyone tell me if they understand:-
>>>
>>> 1) what is a Turing Engine?
>> A Turing Engine is a theoretical construct postulated by Alan Turing.
>> It can be shown that it is mathematically equivalent to all modern
>> digital computing machines. Wikipedia contains a good description
>> of it.
>
> True!
>
>>> 2) can a Turing Engine be made of paper, rather than requiring electricity
>>> or mechanical components?
>
> False! Turing's first engine was a paper one. And so is your logical
> flowchart Chris.

once again, PI falls into the AI trap of confusing the thing itself with
talking about the thing.

Punchline from a joke I first heard in 1965: "He's in Artificial
Intelligence and he sat on the side of the bed all night, talking about
how wonderful it was going to be."

--
Kenneth Sloan [email protected]
Computer and Information Sciences +1-205-932-2213
University of Alabama at Birmingham FAX +1-205-934-5473
Birmingham, AL 35294-1170 http://www.cis.uab.edu/sloan/

Date: 10 Jan 2007 22:09:48
From: Chess One
Subject: Re: !] Re: Human vs Machine Matches

"Kenneth Sloan" <[email protected] > wrote in message
news:[email protected]...
> Chess One wrote:
>> "Chris Mattern" <[email protected]> wrote in message
>> news:[email protected]...
>>> In article <uDeeh.4971$ne3.4901@trndny03>, Chess One wrote:
>>
>>
>>>> Can anyone tell me if they understand:-
>>>>
>>>> 1) what is a Turing Engine?
>>> A Turing Engine is a theoretical construct postulated by Alan Turing.
>>> It can be shown that it is mathematically equivalent to all modern
>>> digital computing machines. Wikipedia contains a good description
>>> of it.
>>
>> True!
>>
>>>> 2) can a Turing Engine be made of paper, rather than requiring
>>>> electricity
>>>> or mechanical components?
>>
>> False! Turing's first engine was a paper one. And so is your logical
>> flowchart Chris.
>
> once again, PI falls into the AI trap of confusing the thing itself with
> talking about the thing.
>
> Punchline from a joke I first heard in 1965: "He's in Artificial
> Intelligence and he sat on the side of the bed all night, talking about
> how wonderful it was going to be."

how amusing? though nothing to do with the question posed. instead, we have
a joke heard in alabama in place of ... ;)

pi

> --
> Kenneth Sloan [email protected]
> Computer and Information Sciences +1-205-932-2213
> University of Alabama at Birmingham FAX +1-205-934-5473
> Birmingham, AL 35294-1170 http://www.cis.uab.edu/sloan/

Date: 10 Dec 2006 00:49:12
From: Guy Macon
Subject: Re: Human vs Machine Matches

Chris Mattern wrote:

>Such data storage does not solely comprise a computer, however, it is
>an integral part of it.

So if I have an operation that embeds a storage device and interface
in my body, *then* it would be OK to use it to help me win chess games?

>As we can see from looking at the Turing Machine, mechanical storage is an
>integral part of it, as it is of the the computers that implement various
>forms of its mathematical equivalents. Therefore, prohibiting computers
>from using mechanical storage to perform any function (such as playing a
>game of chess) is equivalent to prohibiting them from operating at all.

First, the distinction between storage and processor is arbitrary.
a flip-flop can be made out of two NOR gates. A memory system can
be made out of multiple flip flops -- one flip flop per bit of
storage. The rest of the computer can be created out of 100% NOR
gates. So a sea of gates can be the proceesor or the storage
depending on how they are interconnected.

Second, it is not true that all computers have storage devices.
http://en.wikipedia.org/wiki/Analog_computer gives several examples
of computers that do not.

Guy Macon
<http://www.guymacon.com/ >

Date: 10 Dec 2006 01:36:04
From: Chris Mattern
Subject: Re: Human vs Machine Matches

In article <[email protected] >, Guy Macon wrote:
>
>
>
>Chris Mattern wrote:
>
>>Such data storage does not solely comprise a computer, however, it is
>>an integral part of it.
>
>So if I have an operation that embeds a storage device and interface
>in my body, *then* it would be OK to use it to help me win chess games?

If such a device directly interfaced with your brain, becoming an
integral part of the brain's functioning, I do not see a practical method
by which its use could be banned. The advantages of such a device
would be so great that any one who could afford one would likely have it.
>
>>As we can see from looking at the Turing Machine, mechanical storage is an
>>integral part of it, as it is of the the computers that implement various
>>forms of its mathematical equivalents. Therefore, prohibiting computers
>>from using mechanical storage to perform any function (such as playing a
>>game of chess) is equivalent to prohibiting them from operating at all.
>
>First, the distinction between storage and processor is arbitrary.
>a flip-flop can be made out of two NOR gates. A memory system can
>be made out of multiple flip flops -- one flip flop per bit of
>storage. The rest of the computer can be created out of 100% NOR
>gates. So a sea of gates can be the proceesor or the storage
>depending on how they are interconnected.

Yes, electronics can make either memory or processor. I don't
really see how arguing that the distinction is arbitrary helps
your case--if the distinction *is* arbitrary, then how can you
argue that computers be allowed to process but not remember?
However, I don't regard it is as arbitrary because while electronics
can indeed form either, there are many media that can store
data but not process it.

>
>Second, it is not true that all computers have storage devices.
>http://en.wikipedia.org/wiki/Analog_computer gives several examples
>of computers that do not.
>
Since we are not discussing how analog computers play chess but
how digital computers play chess, I don't really see how that's
relevant. The two are entirely different classes of machines.

--
Christopher Mattern

"Which one you figure tracked us?"
"The ugly one, sir."
"...Could you be more specific?"

Date: 10 Dec 2006 13:35:45
From: Guy Macon
Subject: Re: Human vs Machine Matches

Chris Mattern wrote:

>I don't really see how arguing that the distinction is
>arbitrary helps your case

I am not attempting to make a case. I am attempting to explore
an interesting topic. If I point out what I see as a flaw in an
argument for letting a computer use an opening book and forbidding
a human to do so, that does not imply that I diagree with that
position -- just the argumnt presented in favor of it. It's not
as if you and I were tasked with revising the rules for man/machine
chess matches.

> Guy Macon <http://www.guymacon.com/> wrote:
>>
>>Chris Mattern wrote:
>>
>>>Such data storage does not solely comprise a computer, however, it is
>>>an integral part of it.
>>
>>So if I have an operation that embeds a storage device and interface
>>in my body, *then* it would be OK to use it to help me win chess games?
>
>If such a device directly interfaced with your brain, becoming an
>integral part of the brain's functioning, I do not see a practical method
>by which its use could be banned. The advantages of such a device
>would be so great that any one who could afford one would likely have it.

But, in my opinion, it would be illegal under the present rules, and
certainy in violation of the "no reference materials" spirit of them.

What of embedding an entire chess-playing computer? There already
exist humans with embedded computers. They are specialty devices
not currently programmed to play chess but have the processing power
to play as well as a $20 LCD handheld if someone decides to program
one to assist a human in playing chess. See the Book titled _Rebuilt:
How Becoming Part Computer Made Me More Human_ by Michael Chorost
[ http://www.rebuilt-thebook.com/ ]/[ http://www.michaelchorost.com/].

Perhaps when a computer that is powerful enough to beat grandmasters
and small enough and cool-running enough to be embedded in a human
body, the rules will have to be clarified.

>>First, the distinction between storage and processor is arbitrary.
>>a flip-flop can be made out of two NOR gates. A memory system can
>>be made out of multiple flip flops -- one flip flop per bit of
>>storage. The rest of the computer can be created out of 100% NOR
>>gates. So a sea of gates can be the proceesor or the storage
>>depending on how they are interconnected.

>Since we are not discussing how analog computers play chess but
>how digital computers play chess

We? *I* am talking about how machines and humans play chess --
a discussion of theory not limited to currently available hardware.

Guy Macon
<http://www.guymacon.com/ >

Date: 09 Dec 2006 13:59:04
From: Chris Mattern
Subject: Re: Human vs Machine Matches

In article <[email protected] >, Chris Mattern wrote:
>In article <uDeeh.4971$ne3.4901@trndny03>, Chess One wrote:
>>
>>Can anyone tell me if they understand:-
>>
>>1) what is a Turing Engine?
>
>A Turing Engine is a theoretical construct postulated by Alan Turing.
>It can be shown that it is mathematically equivalent to all modern
>digital computing machines. Wikipedia contains a good description
>of it.
>
In a sterling example of reading what the preceding poster meant
instead of what he wrote (and then repeating the latter), I didn't
notice that he wrote Turing Engine instead of Turing Machine (as
Guy Macon pointed out). However, I automatically typed "Turing
Machine" into Wikipedia. That's the term you'll need to use to
read the Wikipedia article: "Turing Engine" won't get you far.

--
Christopher Mattern

"Which one you figure tracked us?"
"The ugly one, sir."
"...Could you be more specific?"

Date: 09 Dec 2006 11:09:17
From: Guy Macon
Subject: Turing Machines

Chess One wrote:

>It seems appropriate to ask some advanced ideas, yet elementary
>factors, and to question people's knowledge of computing and logic:-

My resume (see <http://www.guymacon.com/ >) speaks for itself.

>Can anyone tell me if they understand:-
>
>1) what is a Turing Engine?

Assuming that you are using the term "Turing Engine" to describe
a Turing Machine (please define your terms if this assumption
is incorrect)...

A Turing Machine is any finite state sequential processor that
communicates with external information storage.

>2) can a Turing Engine be made of paper, rather than requiring
>electricity or mechanical components?

A Turing Machine can be made of nothing at all -- a zero state
and xero storage Turing Machine is still a Turing Machine.

I suspect that you really aren't asking about Turing Machines,
but rather about Universal Turing Machines. Those cannot be
made out of paper alone, but can be made out of paper if you
allow the addition of, say, a human being with a pencil. Of
course you need an infinite amount of paper and pencil lead...

>3) can anyone distinguish between instruction sets which are
>logic trees, and other forms of data sets?

The concept of data as something seperate from the program
is an assumption that isn't valid for all possible Turing
Machines.

>4) what is similar with all Turing Engines?

Nothing. "All Turing Machines" includes the zero state and xero
storage Turing Machine.

If you meant to ask what is similar with all Universal Turing
Machines, the answer is states, transitions between states,
actions, and storage.

>5) what is the only factor which differentiates all Turing Engines?

The question is unclear. How does one differentiate all of something?
differentiate it from what?

>6) if a book is described as a data-set, and a computer is a
>data-set, is a book a computer?

If you call a tail a leg, does a sheep have five legs?

>7) what is computed in a book?

Define "book."

>8) what part of a computer is not data?

What part of yellow is not an integer?

>9) what part of a Turing Engine is not data?

Again the concept of data as something seperate from the program
is an assumption that isn't valid for all possible Turing Machines.

>10) of what was the first Turing Engine constucted?

Define "constructed." Do we count Turing's original pencil
and paper verion?

>Perhaps when we examine our knowledge of this subject, we can then assess
>our understanding of people.

Now I have a question for you:

A) What is your point?

Guy Macon
<http://www.guymacon.com/ >

Date: 12 Dec 2006 12:52:09
From: Mike Murray
Subject: Re: Turing Machines

On Sat, 09 Dec 2006 11:09:17 +0000, Guy Macon
<http://www.guymacon.com/ > wrote:

http://www.guymacon.com/

http://ccn.correspondencechess.com/neil_files/neil.jpg

Hey, wait a minute. You don't moonlight as a chess historian, do you?

This is amazing. And both guys doing battle with Phil...

Oh my gawd, I've read too much Sam Sloan. Must stop before it's too
late.

Date: 11 Dec 2006 10:22:18
From: David Richerby
Subject: Re: Turing Machines

Guy Macon <http://www.guymacon.com/ > wrote:
> A Turing Machine is any finite state sequential processor that
> communicates with external information storage.

No. The external information storage has to be based on a finite
alphabet and updated only by the machine. Further, there are limits
on what the sequential processor can do. By your definition, a
machine with an oracle for the halting problem would still be a Turing
Machine and it most certainly is not.

> I suspect that you really aren't asking about Turing Machines, but
> rather about Universal Turing Machines. Those cannot be made out of
> paper alone, but can be made out of paper if you allow the addition
> of, say, a human being with a pencil. Of course you need an
> infinite amount of paper and pencil lead...

Only if you intend to simulate the machine for an infinite time, which
is rather tricky. For any finite number of steps, a finite number of
pencils and pieces of paper will suffice.

Dave.

--
David Richerby Radioactive Chicken (TM): it's like
www.chiark.greenend.org.uk/~davidr/ a farm animal but it'll make you glow
in the dark!

Date: 10 Dec 2006 14:46:38
From: Chess One
Subject: Re: Turing Machines

"Guy Macon" <http://www.guymacon.com/ > wrote in message
news:[email protected]...
>
>
>
> Chess One wrote:
>
>>It seems appropriate to ask some advanced ideas, yet elementary
>>factors, and to question people's knowledge of computing and logic:-
>
> My resume (see <http://www.guymacon.com/>) speaks for itself.
>
>>Can anyone tell me if they understand:-
>>
>>1) what is a Turing Engine?
>
> Assuming that you are using the term "Turing Engine" to describe
> a Turing Machine (please define your terms if this assumption
> is incorrect)...

Not correct. 'Turing Engine' is intended, and specifically differentiated
from a machine. It is a philosophical prospect, decribed as a logic
flow-chart upon simple if... then... propositions.

> A Turing Machine is any finite state sequential processor that
> communicates with external information storage.
>
>>2) can a Turing Engine be made of paper, rather than requiring
>>electricity or mechanical components?
>
> A Turing Machine can be made of nothing at all -- a zero state
> and xero storage Turing Machine is still a Turing Machine.
>
> I suspect that you really aren't asking about Turing Machines,
> but rather about Universal Turing Machines. Those cannot be
> made out of paper alone, but can be made out of paper if you
> allow the addition of, say, a human being with a pencil. Of
> course you need an infinite amount of paper and pencil lead...

You discourse on application means, while I reference the philosophic
'engine'.

>>3) can anyone distinguish between instruction sets which are
>>logic trees, and other forms of data sets?
>
> The concept of data as something seperate from the program
> is an assumption that isn't valid for all possible Turing
> Machines.

I'm sorry, but that response does not answer the question, and involves new
terms. The question is simpler, if somewhat rhetorical, and can be answered
yes or no.

>>4) what is similar with all Turing Engines?
>
> Nothing. "All Turing Machines" includes the zero state and xero
> storage Turing Machine.
>
> If you meant to ask what is similar with all Universal Turing
> Machines, the answer is states, transitions between states,
> actions, and storage.

You again change the question in order to reply. This is not an answer.

>>5) what is the only factor which differentiates all Turing Engines?
>
> The question is unclear. How does one differentiate all of something?
> differentiate it from what?

Differentiates them from each other, and I did not write 'all of something'
I wrote "all Turing Engines."

>>6) if a book is described as a data-set, and a computer is a
>>data-set, is a book a computer?
>
> If you call a tail a leg, does a sheep have five legs?

Sorry! Hypothetical responses deny ability to answer specific questions.

>>7) what is computed in a book?
>
> Define "book."

Why? Don't you know what a book is?

>>8) what part of a computer is not data?
>
> What part of yellow is not an integer?

0/8 so far!

>>9) what part of a Turing Engine is not data?
>
> Again the concept of data as something seperate from the program
> is an assumption that isn't valid for all possible Turing Machines.

Wrong answer - no machine is indicated in the question, but a philosophic
engine is. Though the answer differentiates forms of datum, this is not to
anser the question, of what is NOT datum.

>>10) of what was the first Turing Engine constucted?
>
> Define "constructed." Do we count Turing's original pencil
> and paper verion?

0/10 for this writer. I have used no terms in these questions which are
atypical ones - and though the term 'engine' is here used philosophically,
in initial capitals a 'Turing Engine' is indeed a specific.

>>Perhaps when we examine our knowledge of this subject, we can then assess
>>our understanding of people.
>
> Now I have a question for you:
>
> A) What is your point?

I have 10 points, so to speak. I want to know if anyone can answer any of
them, which all relate to computing, and the philosophy of science, before
becoming dazzled by the same people's understanding of the human role in
programming or in competition.

Phil Innes

> Guy Macon
> <http://www.guymacon.com/>
>

Date: 10 Dec 2006 17:19:43
From: Guy Macon
Subject: Re: Turing Machines

Chess One wrote:
>
>Guy Macon wrote...
>>
>> Assuming that you are using the term "Turing Engine" to describe
>> a Turing Machine (please define your terms if this assumption
>> is incorrect)...
>
>Not correct. 'Turing Engine' is intended, and specifically differentiated
>from a machine. It is a philosophical prospect, decribed as a logic
>flow-chart upon simple if... then... propositions.

You appear to have failed to, as requested, define your terms.
If you wish to discuss something called a "Turing Engine" that
is different from a Turing Machine or Universal Turing Machine
(not the same thing, BTW), you ned to define exactly what you
mean by "Turing Engine" as I requested. Until you do, I shall
assume that what you call a "Turing Engine" is what Turing himself
and everyone else who studies computer science calls a "Universal
Turing Machine."

I would also add that an if-then flowchart cannot be a Turing
Machine. It can be the tape portion, but not the head portion.

>> A Turing Machine is any finite state sequential processor that
>> communicates with external information storage.
>>
>>>2) can a Turing Engine be made of paper, rather than requiring
>>>electricity or mechanical components?
>>
>> A Turing Machine can be made of nothing at all -- a zero state
>> and xero storage Turing Machine is still a Turing Machine.
>>
>> I suspect that you really aren't asking about Turing Machines,
>> but rather about Universal Turing Machines. Those cannot be
>> made out of paper alone, but can be made out of paper if you
>> allow the addition of, say, a human being with a pencil. Of
>> course you need an infinite amount of paper and pencil lead...
>
>You discourse on application means, while I reference the philosophic
>'engine'.

Bullshit. The Universal Turing Machine is a concept. Nobody
has ever actually created one, because creating one is impossible.

>>>3) can anyone distinguish between instruction sets which are
>>>logic trees, and other forms of data sets?
>>
>> The concept of data as something seperate from the program
>> is an assumption that isn't valid for all possible Turing
>> Machines.
>
>I'm sorry, but that response does not answer the question, and involves new
>terms. The question is simpler, if somewhat rhetorical, and can be answered
>yes or no.

As you wish...

Yes. They can.

>>>4) what is similar with all Turing Engines?
>>
>> Nothing. "All Turing Machines" includes the zero state and xero
>> storage Turing Machine.
>>
>> If you meant to ask what is similar with all Universal Turing
>> Machines, the answer is states, transitions between states,
>> actions, and storage.
>
>You again change the question in order to reply. This is not an answer.

Further Bullshit. The answer was clear written in the english language,
and was comprised of the single word "Nothing." Nothing is similar
with all Turing Machines. Just because you don't like an answer that
doesn't mean that you can claim that the answer was not given.

>>>5) what is the only factor which differentiates all Turing Engines?
>>
>> The question is unclear. How does one differentiate all of something?
>> differentiate it from what?
>
>Differentiates them from each other

You are still being unclear. Something that is unique to each one?
No such factor exists.

>>>7) what is computed in a book?
>>
>> Define "book."
>
>Why? Don't you know what a book is?

Evasion noted.

I know exactly what a book is, but you clearly do not, based on your
question. You think books can compute. They cannot. I can only
conclude that you are stupid, crazy, or have a non-standard definition
of "book." So again I ask you to define "book."

>no machine is indicated in the question, but a philosophic
>engine is.

...which you do not appear to be capable of defining.

>> Define "constructed." Do we count Turing's original pencil
>> and paper verion?
>
>0/10 for this writer.

Evasion noted. You won't define your terms because you can't
define your terms.

>> A) What is your point?
>
>I have 10 points, so to speak.

...all on your pointy little head, no doubt.

Date: 07 Dec 2006 15:11:28
From: David Richerby
Subject: Re: Human vs Machine Matches

Guy Macon <http://www.guymacon.com/ > wrote:
> Would it be reasonable to allow the human to bring a printed
> opening book and set of tablebases?

No. Even the 4-man tablebases are about 30MB zipped. Even assuming
they don't expand at all, and that a human could read this just by
presenting each byte as a character, that's about 6,000 pages (at 50
lines per page, 100 characters per line). In `reality', the human
would need a much bigger book than this because one character per byte
of data is not a reasonable assumption.

:-)

Dave.

--
David Richerby Evil Expensive Cheese (TM): it's like
www.chiark.greenend.org.uk/~davidr/ a lump of cheese but it'll break the
bank and it's genuinely evil!

Date: 07 Dec 2006 14:11:14
From: Thomas T. Veldhouse
Subject: Re: Human vs Machine Matches

In rec.games.chess.computer Guy Macon <http://www.guymacon.com/ > wrote:
>
>
>
> Thomas T. Veldhouse wrote:
>
>>Computer only gets opening book and tablebases.
>>Both of which humans are capable of learning/memorizing.
>
> Would it be reasonable to allow the human to bring a printed
> opening book and set of tablebases? Or keep them on a hard
> drive with a simple program that accesses them but cannot
> calculate? The computer, after all, has such data storage.
>

Whatever the human can memorize ... just like now. The computer has an
advantage with memory and storage that a human doesn't. A human has an
advantage with abstract thought and pattern matching as two obvious talents.
Is it fair that humans bring this to the table if computers can't do it?

--
Thomas T. Veldhouse
Key Fingerprint: D281 77A5 63EE 82C5 5E68 00E4 7868 0ADC 4EFB 39F0

Date: 07 Dec 2006 16:06:43
From: Chess One
Subject: Re: Human vs Machine Matches

"Thomas T. Veldhouse" <[email protected] > wrote in message
news:[email protected]...
> In rec.games.chess.computer Guy Macon <http://www.guymacon.com/> wrote:
>>
>>
>>
>> Thomas T. Veldhouse wrote:
>>
>>>Computer only gets opening book and tablebases.
>>>Both of which humans are capable of learning/memorizing.
>>
>> Would it be reasonable to allow the human to bring a printed
>> opening book and set of tablebases? Or keep them on a hard
>> drive with a simple program that accesses them but cannot
>> calculate? The computer, after all, has such data storage.
>>
>
> Whatever the human can memorize ... just like now. The computer has an
> advantage with memory and storage that a human doesn't. A human has an
> advantage with abstract thought and pattern matching as two obvious
> talents.
> Is it fair that humans bring this to the table if computers can't do it?

You are kiting a term 'memory' which is not memory! By using this term you
make memory the same as not-memory! If the computer remembers nothing at
all, it can look up a source. If a person did that we would not call it
memory. Insisting on calling not-memory memory by some process of analogy is
comic!

And so the daft result of this conflation of terms is that programmers refer
to fairness, whereas chess has nothing to do with equalising what is fair
between participants!

It has to do with playing by the rules of chess - which in terms of the
integrity of the activity, makes playing legal chess = fair.

Should you disagree, then how should you not wish the Ferrari to take part
in the 100 metres dash? If the axiom that making its driver get out and run
'cripples' the guys winning chances, and the only consideration for you is
winning at any costs, even by ignoring the rules of the race, there you have
it!

I think the reluctance to accept this point has had the effect of chess
programming pursuing an entirely seperate path from legal chess, [which is
okay!] but we should really describe it as Advanced Chess Programming, no?

If you wish to say what the rules would be for advanced chess, then this is
some other matter than to discuss classical chess - you accept this as a
logical and 'fair' statement?

Phil Innes

> --
> Thomas T. Veldhouse
> Key Fingerprint: D281 77A5 63EE 82C5 5E68 00E4 7868 0ADC 4EFB 39F0
>
>

Date: 07 Dec 2006 19:28:38
From: David Richerby
Subject: Re: Human vs Machine Matches

Chess One <[email protected] > wrote:
> You are kiting a term 'memory' which is not memory! By using this
> term you make memory the same as not-memory! If the computer
> remembers nothing at all, it can look up a source. If a person did
> that we would not call it memory. Insisting on calling not-memory
> memory by some process of analogy is comic!

You are Gene Ray and I claim my five pounds.

Dave.

--
David Richerby Permanent Bulb (TM): it's like a light
www.chiark.greenend.org.uk/~davidr/ bulb but it'll be there for ever!

Date: 07 Dec 2006 14:30:41
From: Guy Macon
Subject: Re: Human vs Machine Matches

Thomas T. Veldhouse wrote:
>
>In rec.games.chess.computer Guy Macon <http://www.guymacon.com/> wrote:
>>
>> Thomas T. Veldhouse wrote:
>>
>>>Computer only gets opening book and tablebases.
>>>Both of which humans are capable of learning/memorizing.
>>
>> Would it be reasonable to allow the human to bring a printed
>> opening book and set of tablebases? Or keep them on a hard
>> drive with a simple program that accesses them but cannot
>> calculate? The computer, after all, has such data storage.
>
>Whatever the human can memorize ... just like now. The computer has an
>advantage with memory and storage that a human doesn't. A human has an
>advantage with abstract thought and pattern matching as two obvious talents.
>Is it fair that humans bring this to the table if computers can't do it?

Is it fair to allow the computer to access data on a hard drive?
Perhaps you should restrict it to whatever it can keep in memory
if that's the restriction you put on humans.

Guy Macon
<http://www.guymacon.com/ >

Date: 07 Dec 2006 19:17:41
From: Chris Mattern
Subject: Re: Human vs Machine Matches

In article <[email protected] >, Guy Macon wrote:
>
>
>
>Thomas T. Veldhouse wrote:
>>
>>In rec.games.chess.computer Guy Macon <http://www.guymacon.com/> wrote:
>>>
>>> Thomas T. Veldhouse wrote:
>>>
>>>>Computer only gets opening book and tablebases.
>>>>Both of which humans are capable of learning/memorizing.
>>>
>>> Would it be reasonable to allow the human to bring a printed
>>> opening book and set of tablebases? Or keep them on a hard
>>> drive with a simple program that accesses them but cannot
>>> calculate? The computer, after all, has such data storage.
>>
>>Whatever the human can memorize ... just like now. The computer has an
>>advantage with memory and storage that a human doesn't. A human has an
>>advantage with abstract thought and pattern matching as two obvious talents.
>>Is it fair that humans bring this to the table if computers can't do it?
>
>Is it fair to allow the computer to access data on a hard drive?

Yes. That's the computer's long-term memory. You keep yours in your
neurons. The computer keeps its on magnetized platters.

>Perhaps you should restrict it to whatever it can keep in memory
>if that's the restriction you put on humans.
>
Computer RAM < > human long-term memory. Hard disk is a much
closer analogy to it.

--
Christopher Mattern

"Which one you figure tracked us?"
"The ugly one, sir."
"...Could you be more specific?"

Date: 07 Dec 2006 19:56:37
From: Thomas T. Veldhouse
Subject: Re: Human vs Machine Matches

Chris Mattern <[email protected] > wrote:
>>
> Computer RAM <> human long-term memory. Hard disk is a much
> closer analogy to it.
>

I agree entirely. Indeed that is what I have posted since.

--
Thomas T. Veldhouse
Key Fingerprint: D281 77A5 63EE 82C5 5E68 00E4 7868 0ADC 4EFB 39F0

Date: 07 Dec 2006 15:30:43
From: David Richerby
Subject: Re: Human vs Machine Matches

Guy Macon <http://www.guymacon.com/ > wrote:
> Is it fair to allow the computer to access data on a hard drive?
> Perhaps you should restrict it to whatever it can keep in memory if
> that's the restriction you put on humans.

Well, the opening book will fit easily in memory. A full set of
five-man tablebases are about 7GB compressed, which isn't a stupendous
memory requirement. (OK, it's excessive for current machines but it's
not hard to imagine that in, say, five years, a standard desktop will
have, say, 16GB of RAM.)

Dave.

--
David Richerby Salted Radioactive Tool (TM): it's
www.chiark.greenend.org.uk/~davidr/ like a hammer but it'll make you glow
in the dark and it's covered in salt!

Date: 07 Dec 2006 16:51:18
From: Thomas T. Veldhouse
Subject: Re: Human vs Machine Matches

David Richerby <[email protected] > wrote:
> Guy Macon <http://www.guymacon.com/> wrote:
>> Is it fair to allow the computer to access data on a hard drive?
>> Perhaps you should restrict it to whatever it can keep in memory if
>> that's the restriction you put on humans.
>
> Well, the opening book will fit easily in memory. A full set of
> five-man tablebases are about 7GB compressed, which isn't a stupendous
> memory requirement. (OK, it's excessive for current machines but it's
> not hard to imagine that in, say, five years, a standard desktop will
> have, say, 16GB of RAM.)
>

You equate active RAM as if it means the same thing as human long-term memory.
It could just as easily be said that RAM is the equivalent to human short term
memory and the hard drive is the equivalent to long term memory. Take away
the tablebases ... then perhaps you better consider a lobotamy for human
player.

--
Thomas T. Veldhouse
Key Fingerprint: D281 77A5 63EE 82C5 5E68 00E4 7868 0ADC 4EFB 39F0

Date: 07 Dec 2006 19:26:30
From: David Richerby
Subject: Re: Human vs Machine Matches

Thomas T. Veldhouse <[email protected] > wrote:
> David Richerby <[email protected]> wrote:
>> Guy Macon <http://www.guymacon.com/> wrote:
>>> Is it fair to allow the computer to access data on a hard drive?
>>> Perhaps you should restrict it to whatever it can keep in memory
>>> if that's the restriction you put on humans.
>>
>> Well, the opening book will fit easily in memory. A full set of
>> five-man tablebases are about 7GB compressed, which isn't a stupendous
>> memory requirement. (OK, it's excessive for current machines but it's
>> not hard to imagine that in, say, five years, a standard desktop will
>> have, say, 16GB of RAM.)
>
> You equate active RAM as if it means the same thing as human
> long-term memory.

I do not! I just said that an opening book will fit easily into a
current computer's memory and 5-man tablebases will fit easily in five
or so years. I said absolutely nothing about how that might compare
with human memory because, frankly, the comparison is meaningless.

I'm rather baffled by your post, I have to say.

Dave.

--
David Richerby Confusing Postman (TM): it's like
www.chiark.greenend.org.uk/~davidr/ a man who delivers the mail but you
can't understand it!

Date: 07 Dec 2006 19:55:05
From: Thomas T. Veldhouse
Subject: Re: Human vs Machine Matches

David Richerby <[email protected] > wrote:
>
> I do not! I just said that an opening book will fit easily into a
> current computer's memory and 5-man tablebases will fit easily in five
> or so years. I said absolutely nothing about how that might compare
> with human memory because, frankly, the comparison is meaningless.
>
> I'm rather baffled by your post, I have to say.
>

What makes it unfair then? Because a computer chess opponent has a much
larger persistant storage than a human being? Or is it that the persistant
storage is accurate? Long term memory could be considered persistant storage.
Short term memory could roughly equate to the registers in a CPU ... or maybe
active RAM, depending upon how you like to do it. I see arguments here saying
it is not fair for a computer to beat a human because it can "remember" a lot
more than a human can. Well ... duh! A computer chess engine is DESIGNED to
play chess, a human being is not. It should not surprise anybody that the day
would come when a computer can routinely beat even the best human chess
players.

--
Thomas T. Veldhouse
Key Fingerprint: D281 77A5 63EE 82C5 5E68 00E4 7868 0ADC 4EFB 39F0

Date: 08 Dec 2006 14:40:36
From: Guy Macon
Subject: Re: Human vs Machine Matches

Thomas T. Veldhouse wrote:

>What makes it unfair then? Because a computer chess opponent has
>a much larger persistant storage than a human being? Or is it
>that the persistant storage is accurate?

I am a human, and my persistant storage consists of memories,
books, files on my PC, and things I have put on the net and
can later retrieve. I am not allowed to access much of my
persistant storage while playing chess, yet a chess-playing
computer has no such restrictions placed upon it.

If the computer is allowed to bring reference material to
the game and look things up during play, why am I not allowed
to do the same?

Guy Macon
<http://www.guymacon.com/ >

Long term memory could be considered persistant storage.
>Short term memory could roughly equate to the registers in a CPU ... or maybe
>active RAM, depending upon how you like to do it. I see arguments here saying
>it is not fair for a computer to beat a human because it can "remember" a lot
>more than a human can. Well ... duh! A computer chess engine is DESIGNED to
>play chess, a human being is not. It should not surprise anybody that the day
>would come when a computer can routinely beat even the best human chess
>players.
>
>--
>Thomas T. Veldhouse
>Key Fingerprint: D281 77A5 63EE 82C5 5E68 00E4 7868 0ADC 4EFB 39F0
>
>

Date: 09 Dec 2006 11:13:37
From: Guy Macon
Subject: Re: Human vs Machine Matches

Thomas T. Veldhouse wrote:

>I see arguments here saying it is not fair for a computer to beat
>a human because it can "remember" a lot more than a human can.

Nope. The argument is that it is not fair to allow the computer
to bring reference material to the match while not allowing the
human to bring reference material to the match.

Guy Macon
<http://www.guymacon.com/ >

Date: 07 Dec 2006 20:51:48
From: David Richerby
Subject: Re: Human vs Machine Matches

Thomas T. Veldhouse <[email protected] > wrote:
> David Richerby <[email protected]> wrote:
>> I do not! I just said that an opening book will fit easily into a
>> current computer's memory and 5-man tablebases will fit easily in
>> five or so years. I said absolutely nothing about how that might
>> compare with human memory because, frankly, the comparison is
>> meaningless.
>>
>> I'm rather baffled by your post, I have to say.
>
> What makes it unfair then?

When have I said that opening books or tablebases are unfair? What
does this have to do with my comment about the size of opening books
and tablebases? Are you actually reading what I'm writing?

Dave.

--
David Richerby Flammable Transparent Smokes (TM):
www.chiark.greenend.org.uk/~davidr/ it's like a pack of cigarettes but
you can see right through it and it
burns really easily!

Date: 07 Dec 2006 21:12:58
From: Thomas T. Veldhouse
Subject: Re: Human vs Machine Matches

David Richerby <[email protected] > wrote:
>
> When have I said that opening books or tablebases are unfair? What
> does this have to do with my comment about the size of opening books
> and tablebases? Are you actually reading what I'm writing?
>

Yes. As I have stated, just because the size of tablebases are big and take a
while to compute, does not make them HARD to compute. One simply has to be
patient. It's like PI ... it is not hard to get out as many digits as one
wants too ... as long as they wait long enough or apply more resources to get
it done faster.

It is easy ... we know how to do, we have done it, and rather than do it
again, we simply store the results; which is no different than memorization.

Since this approach is not feasible for a human being ... we expose a
limitation of the human brain perhaps that is an advantage for Fritz.

--
Thomas T. Veldhouse
Key Fingerprint: D281 77A5 63EE 82C5 5E68 00E4 7868 0ADC 4EFB 39F0

Date: 08 Dec 2006 11:32:21
From: David Richerby
Subject: Re: Human vs Machine Matches

Thomas T. Veldhouse <[email protected] > wrote:
> David Richerby <[email protected]> wrote:
>> When have I said that opening books or tablebases are unfair? What
>> does this have to do with my comment about the size of opening books
>> and tablebases? Are you actually reading what I'm writing?
>
> Yes.

If you're reading what I'm writing, why do you keep following up to my
comments in this subthread about the size of opening books and
tablebases with comments about how easy they are to compute and how
fair or unfair teir use is?

Dave.

--
David Richerby Fluorescent Dangerous Watch (TM):
www.chiark.greenend.org.uk/~davidr/ it's like a precision chronometer but
it could explode at any minute and
it'll hurt your eyes!

Date: 07 Dec 2006 14:55:49
From: Thomas T. Veldhouse
Subject: Re: Human vs Machine Matches

In rec.games.chess.computer Guy Macon <http://www.guymacon.com/ > wrote:
>
> Is it fair to allow the computer to access data on a hard drive?
> Perhaps you should restrict it to whatever it can keep in memory
> if that's the restriction you put on humans.
>

Absolutely! What limits a computer doesn't limit a human and what limits a
human doesn't limit a computer ... to each its own. If a human uses a
computer to beat a computer, now that is a different story. If a human uses a
book to beat a computer, now that is a different story. If a computer uses a
human to beat a human, now that is a different story.

We have two opponents here. Deep Fritz and Kramnik. They set the two down
and let them play. Fritz one. End of story ... the human was defeated
[soundly]. It was fair. Kramnik, a grand master and the loser, did not
complain about the fairness of the match.

--
Thomas T. Veldhouse
Key Fingerprint: D281 77A5 63EE 82C5 5E68 00E4 7868 0ADC 4EFB 39F0

Date: 06 Dec 2006 12:50:47
From: Kenneth Sloan
Subject: Re: Human vs Machine Matches

Thomas T. Veldhouse wrote:

> What I don't get is why, when humans lose, they try to find away to invalidate
> and justify their loss such that it hasn't really occurred. I guess humans
> are fundamentally lawyers at heart ... all of us.
>

This is a very old, and long recognized, part of the cycle of Artificial
Intelligence research.

a) AI folk ask: what would it take to convince you that a machine is
intelligent?

b) skeptics reply: I'll believe a computer is intelligent if it can do
X. (where X changes over time - for the moment, X = win at GM level chess)

c) AI folk produce an artifact that does X

d) skeptics say: how does it work?

e) AI folk explain the mechanisms used by the artifact.

f) skeptics say: oh, well then, I was wrong. Obviously, it does not
require intelligence to do X. It only requires these cheap tricks, that
are *obviously* not related to intelligence.

g) AI folk ask: what would it take...

repeat until done

--
Kenneth Sloan [email protected]
Computer and Information Sciences +1-205-932-2213
University of Alabama at Birmingham FAX +1-205-934-5473
Birmingham, AL 35294-1170 http://www.cis.uab.edu/sloan/

Date: 08 Dec 2006 15:48:55
From: Claudio Grondi
Subject: Re: Human vs Machine Matches

Kenneth Sloan wrote:
> Thomas T. Veldhouse wrote:
>
>> What I don't get is why, when humans lose, they try to find away to
>> invalidate
>> and justify their loss such that it hasn't really occurred. I guess
>> humans
>> are fundamentally lawyers at heart ... all of us.
>>
>
> This is a very old, and long recognized, part of the cycle of Artificial
> Intelligence research.
>
> a) AI folk ask: what would it take to convince you that a machine is
> intelligent?
>
> b) skeptics reply: I'll believe a computer is intelligent if it can do
> X. (where X changes over time - for the moment, X = win at GM level chess)
>
> c) AI folk produce an artifact that does X
>
> d) skeptics say: how does it work?
>
> e) AI folk explain the mechanisms used by the artifact.
>
> f) skeptics say: oh, well then, I was wrong. Obviously, it does not
> require intelligence to do X. It only requires these cheap tricks, that
> are *obviously* not related to intelligence.
>
> g) AI folk ask: what would it take...
>
> repeat until done
>
This reminds me about a funny story I witnessed myself one time in the
far past:

Two children taking part in an family event wanted to impress the adults
by their skills. One of them was good at playing an instrument and the
another could sing.
After the one playing an instrument played one strophe of a song to
demonstrate his skill, the one singing asked hoping its 'opponent' will
maybe disgrace himself (what would help better to expose its advantage
as a singer):

"What about the second and third strophe? Can you play it, too?"

Claudio Grondi

Date: 06 Dec 2006 19:17:32
From: Chess One
Subject: Re: Human vs Machine Matches

This is a cheap understanding of a Turing test - which NB! does not require
a machine, and as far as AI goes is utter nonsense. What Dr. Sloan is
addressing is 'apprehension'.

It is very common [I read it 5 times here today] to hear that the human side
of things is being engaged, while from people's writing, it is clear that
this is not true.

If the term intelligence has to be used, then let it at least contain the
sense that the agency of the intelligence is capable of learning from
experience, otherwise the agency is what is called in psychology, dumber'n a
brick.

This is not a general thread to correct programmer's understanding of
themselves! lol
It proposes various agencies of human machine interaction which resolve
something of what we have not yet established, and such diversions into
rhetoric and assertion are hardly scientific, from either the prospective of
psychology or machine behavior, and in fact avoid the tests which would, if
not resolve, then clarify certain factors.

Phil Innes

"Kenneth Sloan" <[email protected] > wrote in message
news:[email protected]...
> Thomas T. Veldhouse wrote:
>
>> What I don't get is why, when humans lose, they try to find away to
>> invalidate
>> and justify their loss such that it hasn't really occurred. I guess
>> humans
>> are fundamentally lawyers at heart ... all of us.
>>
>
> This is a very old, and long recognized, part of the cycle of Artificial
> Intelligence research.
>
> a) AI folk ask: what would it take to convince you that a machine is
> intelligent?
>
> b) skeptics reply: I'll believe a computer is intelligent if it can do X.
> (where X changes over time - for the moment, X = win at GM level chess)
>
> c) AI folk produce an artifact that does X
>
> d) skeptics say: how does it work?
>
> e) AI folk explain the mechanisms used by the artifact.
>
> f) skeptics say: oh, well then, I was wrong. Obviously, it does not
> require intelligence to do X. It only requires these cheap tricks, that
> are *obviously* not related to intelligence.
>
> g) AI folk ask: what would it take...
>
> repeat until done
>
> --
> Kenneth Sloan [email protected]
> Computer and Information Sciences +1-205-932-2213
> University of Alabama at Birmingham FAX +1-205-934-5473
> Birmingham, AL 35294-1170 http://www.cis.uab.edu/sloan/

Date: 06 Dec 2006 16:30:10
From: Dave (from the UK)
Subject: Re: Human vs Machine Matches

Thomas T. Veldhouse wrote:

> Computer only gets opening book and tablebases. Both of which humans are
> capable of learning/memorizing.
>
> Humans once had an advantage that we could prune the options down in a
> reliable way. That is proving to no longer be the advantage. Had we still
> have it ... I would think somebody should be arguing that humans not be able
> to do this ... wait, you can't take that away from a human. Same thing with a
> computer. A computer has photographic memory and thinks REALLY REALLY FAST.
> That is ITS competitive advanatage. It has finally proven to be overwhelming.
>
> What I don't get is why, when humans lose, they try to find away to invalidate
> and justify their loss such that it hasn't really occurred. I guess humans
> are fundamentally lawyers at heart ... all of us.
>

It can really be no surprise the computer has won - it was expected to
happen one day. There is no way I could do maths as fast as a computer
or run as fast as a car.

My computer, with a high end maths package and a mere 450 MHz CPU can
compute exactly using just integers 10000! in 32 ms (FWIW. a floating
point approximation is about 2.846259680917055 x 10^35659).

It would take me a very long time to compute 10000! and write down all
35660 digits by hand and the chances of an error would be extreamly
high. (Actually, it is quite possible it would take me longer than my
lifetime to do it.)

But I should not feel threatened. For a long time man has built tools to
do jobs faster / better than he can do. Chess can't be expected to be
any different really.

--
Dave (from the UK)

Please note my email address changes periodically to avoid spam.
It is always of the form: [email protected]
Hitting reply will work for a few months only - later set it manually.

http://witm.sourceforge.net/ (Web based Mathematica front end)

Date: 06 Dec 2006 17:20:33
From: David Richerby
Subject: Re: Human vs Machine Matches

Dave (from the UK) <[email protected] > wrote:
> My computer, with a high end maths package and a mere 450 MHz CPU can
> compute exactly using just integers 10000! in 32 ms (FWIW. a floating
> point approximation is about 2.846259680917055 x 10^35659).
>
> It would take me a very long time to compute 10000! and write down
> all 35660 digits by hand and the chances of an error would be
> extreamly high.

We should ban computers from using fast multiplication algorithms. It
gives them an unfair advantage.

Dave.

--
David Richerby Electronic Homicidal Clock (TM): it's
www.chiark.greenend.org.uk/~davidr/ like a clock but it wants to kill you
and it uses electricity!