[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
File computer-go log.199803
This digest for list has been created on Mon Mar 2 02:03:22 3898
------- THIS IS A RFC934 COMPILANT DIGEST, YOU CAN BURST IT -------
Message-Id: <199803020059.JAA53276@xxxxxxxxxxxxxxxxx>
X-Sender: hiroit@xxxxxxxxxxxxxxxxx
X-Mailer: Windows Eudora Pro Version 2.1.2J
Mime-Version: 1.0
Content-Type: text/plain; charset="ISO-2022-JP"
Date: Mon, 02 Mar 1998 09:59:53 +0900
To: computer-go@xxxxxxxxxxxxxxxxx
From: Hiroshi Ito <hiroit@xxxxxxxxxxxxxxxxx>
Subject: Re: Notation
X-Sequence: 41
Precedence: list
unsubscribe
伊藤 浩
Hiroshi Ito
Advanced Development, Mobile Systems, Yamato, Japan
Tel:(0462)-73-3516/Fax:(0462)-73-4773
- hiroit@xxxxxxxxxxxxxxxxx (from Internet)
- hiroit@xxxxxxxxxxxxxxxxx@internet (from IBM Notes users)
- hiroyon@xxxxxxxxxxxxxxxxx (Home)
------- CUT --- CUT --- CUT --- CUT --- CUT --- CUT --- CUT -------
X-Sender: mueller@xxxxxxxxxxxxxxxxx
Message-Id: <v02140b01b12268776479@[192.31.197.141]>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Date: Wed, 4 Mar 1998 11:11:31 +0900
To: computer-go@xxxxxxxxxxxxxxxxx
From: mueller@xxxxxxxxxxxxxxxxx (Martin Mueller)
Subject: CFP - Conference on Computers and Games (CG'98)
X-Sequence: 42
Precedence: list
Call For Papers
The First International Conference on Computers and Games
(CG'98)
Tsukuba, Japan
November 11-12, 1998
The International Conference on Computers and Games '98 will be held in
Tsukuba, Japan from November 11-12, 1998. The conference has been motivated
by the previous successes of computer games related events held in Japan,
such as the series of four Game Programming Workshops (GPW'94 to GPW'97) and
the IJCAI-97 Workshop on Computer Games.
This conference aims to provide an international forum for researchers to
gather and exchange ideas on all aspects of research related to computers and
games. Relevant topics include, but are not limited to: the current state of
game-playing programs, new theoretical developments in game-related research,
general scientific contributions produced by the study of games, social
aspects of computer games, cognitive research of how humans play games, and
issues related to networked games. This will be the first time that an
international conference has been dedicated to these issues.
Timetable for Paper Submission:
May 10, 1998 Deadline for Paper Submission
July 20, 1998 Acceptance notification
Paper Submission Requirements:
Each submission must include a cover and a copy of the complete manuscript.
The cover page should include :
Author Name(s), affiliation(s), complete address(es); telephone, fax,
E-mail for the principal author, and the title of the paper.
Area(s) that relate(s) to your paper with some key words.
Authors must submit 3 printed copies of paper or a postscript of the paper.
Other requirements are:
the papers must be in the English language, not exceeding 10 pages.
(The receipt of a paper will be acknowledged). Notice of acceptance of
papers will be sent by July 20, 1998 to the principal author.
It is encouraged for submitted papers to follow the IJCAI-97 LaTeX style,
available from IJCAI's home page: http://ijcai.org/ijcai-97 or by sending
E-mail to info@xxxxxxxxxxxxxxxxx
Review Process and Paper Submissions:
Papers will be refereed, and those selected will be scheduled for
presentation and printed in the proceedings of the conference. Authors of
accepted papers, or their representatives, are expected to present their
papers at the conference.
Publishing Policy of the Conference Proceedings
All papers accepted for presentation will initially appear in Proceedings
at the conference. Later, a book of the proceedings will be published by an
international publisher (under negotiation). The decision on which papers to
include in the book publication will be based on the recommendations of the
conference reviewers. For some papers, there may be a second reviewing
process in which authors will be given the chance to address shortcomings
(either technical or presentational) identified in the initial review.
Send papers to:
Prof. Jaap van den Herik, Program chair
Phone: +31-43-388 3485 or 3477
Fax: +31-43-3252392
E-mail: herik@xxxxxxxxxxxxxxxxx
Department of Computer Science
University of Maastricht
P.O. Box 616, 6200 MD Maastricht, The Netherlands
or
Dr. Hiroyuki Iida, Program co-chair
Phone: +81 534 781 456
Fax: +81 534 754 595
E-mail: iida@xxxxxxxxxxxxxxxxx
Computer Games Research Institute
Faculty of Information
Shizuoka University
3-5-1 Johoku, Hamamatsu, Shizuoka, 432 Japan.
Conference Registration Fee
Registration fee (including proceedings and drinks) is $US100 or 12,000yen.
Conference Officials
Conference Organizing Committee
Conference co-chairs:
Hitoshi Matsubara (ETL, Japan)
Yoshiyuki Kotani (Tokyo Univ. of Agr. & Tech., Japan)
Takenobu Takizawa (Waseda Univ., Japan)
Atsushi Yoshikawa (NTT Basic Labolatory, Japan)
Secretary-Treasurer:
Morihiko Tajima (ETL, Japan)
Liaison Officer:
Reijer Grimbergen (ETL, Japan)
Ian Frank (ETL, Japan)
Martin Muller (ETL, Japan)
Advisory Committee
Elwyn Berlekamp (University of Calfornia, Berkeley, USA)
Hans Berliner (CMU, USA)
Jurg Nievergelt (ETH, SW)
Monty Newborn (McGill Univ., CA)
Tony Marsland (University of Alberta, CA)
Conference Program Committee Chairs:
Jaap van den Herik (Maastricht Univ., NL)
Hiroyuki Iida (Shizuoka Univ., Japan)
Programming Committee:
Jaap van den Herik (Maastricht University, chair)
Hiroyuki Iida (Shizuoka University, co-chair)
Don F. Beal (Queen Mary and Westfield College, UK)
Janet Wiles (Queenland Univeristy, AUS)
Jonathan Schaeffer (University of Alberta, CN)
Jos Uiterwijk (Maastricht Univ., NL)
Michael Buro (NEC Institute, USA)
Noriaki Sanechika (ETL, Japan)
Kenji Koyama (NTT Communication Science Lab, Japan)
Kohei Noshita (Univ. of Electro-Communications, Japan)
Ralph Gasser (Microsoft Inc., SW)
Richard Korf (University of Calfornia, Los Angels, USA)
Shaul Markovitch (Technion - Israel Institute of Technology, Israel)
Susan Epstein (Hunter College, USA)
Takao Uehara (Tokyo Engineering Univ., Japan)
Wim Pijls (Erasmus Univ. Rotterdam, NL)
Reijer Grimbergen (ETL, Japan)
Ian Frank (ETL, Japan)
Martin Muller (ETL, Japan)
The Conference will be sponsered by Electrotechnical Laboratory,
Shizuoka University, Japan Shogi Association, Japan Go Association and
Computer Shogi Association.
HOME PAGE http://www.cs.inf.shizuoka.ac.jp/‾iida/CG98-CFP.html
----------------------------------------------------------------------------
------- CUT --- CUT --- CUT --- CUT --- CUT --- CUT --- CUT -------
X-Sender: wetter@xxxxxxxxxxxxxxxxx
Message-Id: <v04003a06b125eb84a892@[192.55.95.8]>
In-Reply-To: <199803020059.JAA53276@xxxxxxxxxxxxxxxxx>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Date: Fri, 6 Mar 1998 11:13:11 -0700
To: computer-go@xxxxxxxxxxxxxxxxx
From: "Pierce T. Wetter III" <pierce@xxxxxxxxxxxxxxxxx>
Subject: New Direction, a computer Tutor
X-Sequence: 43
Precedence: list
Perhaps Computer Go has gone in the wrong direction...
What I would really like would be a program that could load my IGS games
and make rude comments about the stupid moves I've made!
In other words, not an opponent, but a teacher...
Perhaps this could be done without considering the "whole board
strategy", which would make it easier to write the program. Also, the
program could have "preset scenarios" to teach things like invasions, etc.
For instance, how about a joseki tutor that asked you questions at each
step of the path, like:
Do you want the corner, or a wall to the left?
As the old proverb goes: "When you study Joseki, you lose two stones".
That's because the sequences are taught, not the decisions that go into
those sequences.
Perhaps these could even be seperate programs:
Joseki Tutor
Invasion/Defend Tutor
Shape Tutor
Fuskei Tutor
Endgame Tutor
Life/Death Tutor (but up to dan-level problems!)
Thoughts?
Pierce
----------------------------------------------------------------
Pierce T. Wetter III, Director, Twin Forces, Inc.
e-mail: pierce@xxxxxxxxxxxxxxxxx
<http://tfe.infomagic.com/>
U.S. Mail: 1300 South Milton Rd, Suite 206, Flagstaff, AZ 86001
------- CUT --- CUT --- CUT --- CUT --- CUT --- CUT --- CUT -------
Message-Id: <3.0.1.16.19980306194131.383714d4@xxxxxxxxxxxxxxxxx>
X-Sender: m-15061@xxxxxxxxxxxxxxxxx
X-Mailer: Windows Eudora Light Version 3.0.1 (16)
Date: Fri, 06 Mar 1998 19:41:31
To: computer-go@xxxxxxxxxxxxxxxxx
From: Jens Yllman <jens.yllman@xxxxxxxxxxxxxxxxx>
Subject: Re: New Direction, a computer Tutor
Cc: "Pierce T. Wetter III" <pierce@xxxxxxxxxxxxxxxxx>
In-Reply-To: <v04003a06b125eb84a892@[192.55.95.8]>
References: <199803020059.JAA53276@xxxxxxxxxxxxxxxxx>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
X-Sequence: 44
Precedence: list
Actually I thinking about putting some kind of teachers in my program. One
easy solution I was thinking about is just problems. You can choose
category and/or difficulty. And after every problem you can get some text
about why you should play as in the problem and maybe links to related
topics. I think this is great. Especially for people that does not have
that much time. You start the program say I want 3 related problems for
5k-1k. And after solving them you read the text/lesson about the problems.
You should also be able to just read the texts/lessons with out the problems.
The problem is to collect all this information. The program is simple to
make. Anyway I will put something like this in my program when I think it
is time to let others use it. Or maybe this will come when I can sell the
program.
Jens Yllman
At 11.13 1998-03-06 -0700, you wrote:
>
> Perhaps Computer Go has gone in the wrong direction...
>
> What I would really like would be a program that could load my IGS games
>and make rude comments about the stupid moves I've made!
>
> In other words, not an opponent, but a teacher...
>
> Perhaps this could be done without considering the "whole board
>strategy", which would make it easier to write the program. Also, the
>program could have "preset scenarios" to teach things like invasions, etc.
>
> For instance, how about a joseki tutor that asked you questions at each
>step of the path, like:
>
> Do you want the corner, or a wall to the left?
>
> As the old proverb goes: "When you study Joseki, you lose two stones".
>
> That's because the sequences are taught, not the decisions that go into
>those sequences.
>
> Perhaps these could even be seperate programs:
>
> Joseki Tutor
> Invasion/Defend Tutor
> Shape Tutor
> Fuskei Tutor
> Endgame Tutor
> Life/Death Tutor (but up to dan-level problems!)
>
>
> Thoughts?
>
> Pierce
>
>----------------------------------------------------------------
> Pierce T. Wetter III, Director, Twin Forces, Inc.
> e-mail: pierce@xxxxxxxxxxxxxxxxx
> <http://tfe.infomagic.com/>
>U.S. Mail: 1300 South Milton Rd, Suite 206, Flagstaff, AZ 86001
>
>
>
>
------- CUT --- CUT --- CUT --- CUT --- CUT --- CUT --- CUT -------
Message-Id: <35005ECF.B3C806ED@xxxxxxxxxxxxxxxxx>
Date: Fri, 06 Mar 1998 14:38:40 -0600
From: Carlisle <carlisle@xxxxxxxxxxxxxxxxx>
X-Mailer: Mozilla 4.02 [en] (Win95; I)
Mime-Version: 1.0
To: "Pierce T. Wetter III" <pierce@xxxxxxxxxxxxxxxxx>
Cc: computer-go@xxxxxxxxxxxxxxxxx
Subject: Re: New Direction, a computer Tutor
References: <v04003a06b125eb84a892@[192.55.95.8]>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Sequence: 45
Precedence: list
Pierce T. Wetter III wrote:
> Perhaps Computer Go has gone in the wrong direction...
>
> What I would really like would be a program that could load my IGS games
> and make rude comments about the stupid moves I've made!
>
> In other words, not an opponent, but a teacher...
> ...
Maybe not so much a new direction as another viewpoint. It occurs to me that
such a tutor could be 'connected' to a 'learning' computer-go program with
comments which could be converted into algorithmic modification.
.
BUT, when the tutor no longer identified any 'not-the-best' moves, the program
would be as strong as the tutor... unless the tutor was also learning from the
losses of it's symbiant program, etc.
.
Carlisle
------- CUT --- CUT --- CUT --- CUT --- CUT --- CUT --- CUT -------
Message-Id: <3500648E.F83B633A@xxxxxxxxxxxxxxxxx>
Date: Fri, 06 Mar 1998 13:03:10 -0800
From: zhang <zhang@xxxxxxxxxxxxxxxxx>
Reply-To: zhang@xxxxxxxxxxxxxxxxx
Organization: USC/ISI, 4676 Admiralty way, Suite 1001, Marina del Rey,
CA 90292
X-Mailer: Mozilla 4.04 [en] (WinNT; U)
Mime-Version: 1.0
To: computer-go@xxxxxxxxxxxxxxxxx
Subject: Java interface for a Go program
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Sequence: 46
Precedence: list
This may not be the right place to ask this question, but I couldn't
find a better place for help. I am looking for a Java program that
shows the board on screen, accepts user input (such as a mouse click),
and displays the input. If you have this kind Java program or know
where I can find one, please let me know. Thanks very much!!
-wayne zhang
University of Southern California
Information Sciences Institute &
Computer Science Department
------- CUT --- CUT --- CUT --- CUT --- CUT --- CUT --- CUT -------
Message-Id: <3500A677.E3A2D1E6@xxxxxxxxxxxxxxxxx>
Date: Fri, 06 Mar 1998 19:44:23 -0600
From: "David E. Hartman, Ph.D." <dart@xxxxxxxxxxxxxxxxx>
Reply-To: dart@xxxxxxxxxxxxxxxxx
X-Mailer: Mozilla 4.04 [en] (Win95; U)
Mime-Version: 1.0
To: computer-go@xxxxxxxxxxxxxxxxx
Subject: Bimonthly wish for WinCE go program...
References: <3500648E.F83B633A@xxxxxxxxxxxxxxxxx>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Sequence: 47
Precedence: list
Dear Group:
I keep hoping that someone (David Fotland are you listening?) will put out
a version of their computer go program for those little Windows CE
organizers. I am using a Sharp Mobilon with a little color screen and 16
mb of RAM. Surely there's a way to port someone's go program to such a
Windows CE device and have a good pocket go machine -
Continuing to hope...
David Hartman
------- CUT --- CUT --- CUT --- CUT --- CUT --- CUT --- CUT -------
Message-Id: <3.0.3.32.19980306174225.00851210@xxxxxxxxxxxxxxxxx>
X-Sender: rtayek@xxxxxxxxxxxxxxxxx
X-Mailer: QUALCOMM Windows Eudora Pro Version 3.0.3 (32)
Date: Fri, 06 Mar 1998 17:42:25 -0800
To: zhang@xxxxxxxxxxxxxxxxx, computer-go@xxxxxxxxxxxxxxxxx
From: Ray Tayek <rtayek@xxxxxxxxxxxxxxxxx>
Subject: Re: Java interface for a Go program
In-Reply-To: <3500648E.F83B633A@xxxxxxxxxxxxxxxxx>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
X-Sequence: 48
Precedence: list
At 01:03 PM 3/6/98 -0800, zhang wrote:
>This may not be the right place to ask this question, but I couldn't
>find a better place for help. I am looking for a Java program that
>shows the board on screen, accepts user input (such as a mouse click),
>and displays the input. ...
i have source for a very simple go program on my web page (SimpleGo). this
just allows you to place stones until the first capture. you might also
look at www.gamelan.com or rec.games.go
Ray (will hack java for food) http://home.pacbell.net/rtayek/
hate Spam? http://www.compulink.co.uk/‾net-services/spam/
------- CUT --- CUT --- CUT --- CUT --- CUT --- CUT --- CUT -------
Message-Id: <3.0.5.32.19980306213426.007e5d40@xxxxxxxxxxxxxxxxx>
X-Sender: fotland@xxxxxxxxxxxxxxxxx
X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.5 (32)
Date: Fri, 06 Mar 1998 21:34:26 -0800
To: "Pierce T. Wetter III" <pierce@xxxxxxxxxxxxxxxxx>
From: David Fotland <fotland@xxxxxxxxxxxxxxxxx>
Subject: Re: New Direction, a computer Tutor
Cc: computer-go@xxxxxxxxxxxxxxxxx
In-Reply-To: <v04003a06b125eb84a892@[192.55.95.8]>
References: <199803020059.JAA53276@xxxxxxxxxxxxxxxxx>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
X-Sequence: 49
Precedence: list
The next release of Many Faces will have an analysis command
which analyzes the last move played and gives reasons for it,
and how far off it is from MF's favorite move. This has been
a request for years, but I never thought the engine was good enough
to give reasonable advice before.
MF's joseki tutor has the bad moves as well as the good moves, so
that when you study at least you can see why some of the bad moves
don't work. Not as good as what you want I know...
And for life and death problems there are several good programs.
David
At 11:13 AM 3/6/98 -0700, you wrote:
>
> Perhaps Computer Go has gone in the wrong direction...
>
> What I would really like would be a program that could load my IGS games
>and make rude comments about the stupid moves I've made!
>
> In other words, not an opponent, but a teacher...
>
> Perhaps this could be done without considering the "whole board
>strategy", which would make it easier to write the program. Also, the
>program could have "preset scenarios" to teach things like invasions, etc.
>
> For instance, how about a joseki tutor that asked you questions at each
>step of the path, like:
>
> Do you want the corner, or a wall to the left?
>
> As the old proverb goes: "When you study Joseki, you lose two stones".
>
> That's because the sequences are taught, not the decisions that go into
>those sequences.
>
> Perhaps these could even be seperate programs:
>
> Joseki Tutor
> Invasion/Defend Tutor
> Shape Tutor
> Fuskei Tutor
> Endgame Tutor
> Life/Death Tutor (but up to dan-level problems!)
>
>
> Thoughts?
>
> Pierce
>
>----------------------------------------------------------------
> Pierce T. Wetter III, Director, Twin Forces, Inc.
> e-mail: pierce@xxxxxxxxxxxxxxxxx
> <http://tfe.infomagic.com/>
>U.S. Mail: 1300 South Milton Rd, Suite 206, Flagstaff, AZ 86001
>
>
>
>
------- CUT --- CUT --- CUT --- CUT --- CUT --- CUT --- CUT -------
Message-Id: <3.0.5.32.19980306214521.007ecdd0@xxxxxxxxxxxxxxxxx>
X-Sender: fotland@xxxxxxxxxxxxxxxxx
X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.5 (32)
Date: Fri, 06 Mar 1998 21:45:21 -0800
To: dart@xxxxxxxxxxxxxxxxx
From: David Fotland <fotland@xxxxxxxxxxxxxxxxx>
Subject: Re: Bimonthly wish for WinCE go program...
Cc: computer-go@xxxxxxxxxxxxxxxxx
In-Reply-To: <3500A677.E3A2D1E6@xxxxxxxxxxxxxxxxx>
References: <3500648E.F83B633A@xxxxxxxxxxxxxxxxx>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
X-Sequence: 50
Precedence: list
Well, that's two requests for windows CE in the last month :)
I'd like to do a winCE version of Many Faces, but I'm putting it
off until the winCE API gets a little closer to the win32 API. I'm
hoping to have a single GUI source with a few ifdefs.
It's also stacked up behind a chess engine, a weak shogi engine, a
9x9 free windows version, strength improvements for this year (a complete
rewrite of the static life/death evaluator and a new territory evaluator),
entering the moves from the 6 volume Japanese Joseki dictionary I bought,
improving the problem presentation part of the gui and adding a few thousand
problems, adding more pro games, and releasing the 11.0 CD-ROM.
If I could find someone who lives near me, reads Japanese, plays go, and is
interested in entering Joseki from a dictionary for minimum wage, at least
one of the tasks would get done :)
David
At 07:44 PM 3/6/98 -0600, you wrote:
>Dear Group:
>
>I keep hoping that someone (David Fotland are you listening?) will put out
>a version of their computer go program for those little Windows CE
>organizers. I am using a Sharp Mobilon with a little color screen and 16
>mb of RAM. Surely there's a way to port someone's go program to such a
>Windows CE device and have a good pocket go machine -
>
>Continuing to hope...
>
>David Hartman
>
>
>
------- CUT --- CUT --- CUT --- CUT --- CUT --- CUT --- CUT -------
Message-Id: <3501706E.67F4653F@xxxxxxxxxxxxxxxxx>
Date: Sat, 07 Mar 1998 10:06:06 -0600
From: "David E. Hartman, Ph.D." <dart@xxxxxxxxxxxxxxxxx>
Reply-To: dart@xxxxxxxxxxxxxxxxx
X-Mailer: Mozilla 4.04 [en] (Win95; U)
Mime-Version: 1.0
To: David Fotland <fotland@xxxxxxxxxxxxxxxxx>, computer-go@xxxxxxxxxxxxxxxxx
Subject: Re: Bimonthly wish for WinCE go program...
References: <3500648E.F83B633A@xxxxxxxxxxxxxxxxx>
<3.0.5.32.19980306214521.007ecdd0@xxxxxxxxxxxxxxxxx>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Sequence: 51
Precedence: list
David:
Well, I guess I'll be waiting a while then...hope not too long. I thought
though that many CE programs could simply be recompiled from Windows Code with
only minor modifications, using a CE compiler. If that's true, maybe it
wouldn't be as difficult as entering 10,000 new Joseki : -) dh
David Fotland wrote:
> Well, that's two requests for windows CE in the last month :)
>
> I'd like to do a winCE version of Many Faces, but I'm putting it
> off until the winCE API gets a little closer to the win32 API. I'm
> hoping to have a single GUI source with a few ifdefs.
>
> It's also stacked up behind a chess engine, a weak shogi engine, a
> 9x9 free windows version, strength improvements for this year (a complete
> rewrite of the static life/death evaluator and a new territory evaluator),
> entering the moves from the 6 volume Japanese Joseki dictionary I bought,
> improving the problem presentation part of the gui and adding a few thousand
> problems, adding more pro games, and releasing the 11.0 CD-ROM.
>
>
------- CUT --- CUT --- CUT --- CUT --- CUT --- CUT --- CUT -------
Message-Id: <3501768B.85BB89A@xxxxxxxxxxxxxxxxx>
Date: Sat, 07 Mar 1998 10:32:11 -0600
From: "David E. Hartman, Ph.D." <dart@xxxxxxxxxxxxxxxxx>
Reply-To: dart@xxxxxxxxxxxxxxxxx
X-Mailer: Mozilla 4.04 [en] (Win95; U)
Mime-Version: 1.0
To: Defee Pawel <pawel@xxxxxxxxxxxxxxxxx>, computer-go@xxxxxxxxxxxxxxxxx
Subject: Re: Bimonthly wish for WinCE go program...
References: <Pine.OSF.3.95.980307181139.11545A-100000@xxxxxxxxxxxxxxxxx>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Sequence: 52
Precedence: list
I disagree, especially since many of the organizers have a full compliment of RAM
and powerful graphics processing that is already a part of WinCE. As far as
demand, that depends if a decent go program could be marketed in Asian markets,
where these organizers are made anyway. I would think that there's a good chance
at a world wide market, as well as the possibility of having the game promoted in
the U.S. as an alternative to those mindless microsoft games already in the unit
(minesweeper, reversi, etc.)
Best Regards, David Hartman
Defee Pawel wrote:
> On Fri, 6 Mar 1998, David E. Hartman, Ph.D. wrote:
>
> > I keep hoping that someone (David Fotland are you listening?) will put out
> > a version of their computer go program for those little Windows CE
> > organizers. I am using a Sharp Mobilon with a little color screen and 16
> > mb of RAM. Surely there's a way to port someone's go program to such a
> > Windows CE device and have a good pocket go machine -
>
> Hi,
>
> I do not think so. Those oranizers tend to be built around a good
> microcontroller or a simple microprocessor (80186 level). So, unless you
> are satisfied with playing against a 20-30 kyu program, there is no hope
> on getting anything to run on those. Besides, there are simply not enough
> users (potential buyers) for any CE Go program.
>
> Just my thoughts,
> pawel
>
> *******************************************************************************
> Pawel Defee
> pawel.defee@xxxxxxxxxxxxxxxxx
> *******************************************************************************
------- CUT --- CUT --- CUT --- CUT --- CUT --- CUT --- CUT -------
Date: Sat, 7 Mar 1998 18:03:36 +0100
From: Heikki Levanto <heikki@xxxxxxxxxxxxxxxxx>
Message-Id: <199803071703.SAA24438@xxxxxxxxxxxxxxxxx>
To: computer-go@xxxxxxxxxxxxxxxxx
X-Sequence: 53
Precedence: list
To: computer-go@xxxxxxxxxxxxxxxxx
Subject: Re: New Direction, a computer Tutor
Newsgroups: lsd.compgo
Organization: LSD - Levanto Software Development
X-Newsreader: TIN [version 1.2 PL2]
On Fri, 6 Mar 1998 11:13:11 -0700 Pierce T. Wetter III (pierce@xxxxxxxxxxxxxxxxx)
wrote in lsd.compgo:
> For instance, how about a joseki tutor that asked you questions at each
> step of the path
I think this is pretty hard thing for a program. I feel that translating the
program's reasons to human language is not that simple, as the program is
likely to "think" in some rather abstract terms, with lots of numerical
values and boollean flags. Sometimes the reason for a bad move is buried in
the bottom of a large search tree, so even identifying "the" reason may be
difficult.
The obvious alternative is to collect fixed problems, with fixed,
human-written explanations for each alternative move. This will take a lot
of time.
But why should one "guru" sit down and produce such a collection, at great
expense and effort? One of the most profitable teaching I have received has
been discussing a problem with other players at my level or little over...
So, why not let all the users of such a program collect their comments and
reasons? The obvious place would be a web page. Start up with a problem,
and let all readers propose moves and/or comment on existing proposals.
For ecample, when you see the problem, you can click on a point, and get a
summary ( Black 1 at B3: 17 comments say this move can kill, 85 says it can
not), the position with the move played, where you can go on clicking, an
option to see summaries of the reasons peolle have given, and a button to
add your own comments ( a simple evaluation of the move on a short scale
(optimal, works, ineffective, fails, horrible), and a box to explain
yourself in.)
I think coding such a web page would be a much smaller project than writing
a GO program, so there ought to be capable people on this list. If anyone is
willing, is of course another question - there would be quite some
maintenance to purge off irrelevant comments and flames "D4: Kilroy was
here" - "No he wasn't, you #%&)!". Maybe this could be dona automatically by
the readers as well, just by voting against bad advice, or by requesting
someone to second a comment within a certain time lest the comment be purged
(and possibly shown on another list for those who really want to see
everything...)
Any comments on that??
- Heikki Levanto
--
Heikki Levanto LSD - Levanto Software Development <heikki@xxxxxxxxxxxxxxxxx>
------- CUT --- CUT --- CUT --- CUT --- CUT --- CUT --- CUT -------
Message-Id: <6297CD447F92D11199F5006097BA9D1E65F2@xxxxxxxxxxxxxxxxx>
From: Scott Dossey <sdossey@xxxxxxxxxxxxxxxxx>
To: computer-go@xxxxxxxxxxxxxxxxx
Subject: Computer Go Tutor
Date: Mon, 9 Mar 1998 11:39:24 -0800
X-Priority: 3
Mime-Version: 1.0
X-Mailer: Internet Mail Service (5.0.1458.49)
Content-Type: text/plain
X-Sequence: 54
Precedence: list
There is one major advantage to the concept of having a "computer go
tutor" integrated with a go playing program--and that is having exposure
to how the program thinks....IE--why and how it makes mistakes.
I think many of the best programs could benefit from having a "go tutor"
add-on module, just so the programmer has some degree of visibility as
to what the program is thinking rather than having to walk through more
abstract data structures to figure out what's going on when the AI makes
a bad move.
Furthermore, graphical debugging displays of what the program "thinks"
it owns and to what degree could be very helpful.
I'm sure some of you are already using such techniques.
-Scott Dossey
------- CUT --- CUT --- CUT --- CUT --- CUT --- CUT --- CUT -------
Date: Mon, 9 Mar 1998 13:05:39 -0600 (CST)
From: Daniel Hallmark <dgh@xxxxxxxxxxxxxxxxx>
To: "David E. Hartman, Ph.D." <dart@xxxxxxxxxxxxxxxxx>
Cc: David Fotland <fotland@xxxxxxxxxxxxxxxxx>, computer-go@xxxxxxxxxxxxxxxxx
Subject: Re: Bimonthly wish for WinCE go program...
In-Reply-To: <3501706E.67F4653F@xxxxxxxxxxxxxxxxx>
Message-Id: <Pine.LNX.3.91.980309125132.30723B-100000@xxxxxxxxxxxxxxxxx>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
X-Sequence: 55
Precedence: list
Windows CE offers only a subset of the Win32 API functionality so while
it may be a simple task to recompile a CE program to run under the full
Win32, the converse is not true.
For those who might be interested, there is an article in the April 1998
Dr. Dobbs Journal which discusses some of the portability issues between
the CE and Win32 APIs as well as a sidebar containing information on
features of CE version 2 which have been upgraded form the 1.0 featureset.
Daniel Hallmark
On Sat, 7 Mar 1998, David E. Hartman, Ph.D. wrote:
> Well, I guess I'll be waiting a while then...hope not too long. I thought
> though that many CE programs could simply be recompiled from Windows Code with
> only minor modifications, using a CE compiler. If that's true, maybe it
> wouldn't be as difficult as entering 10,000 new Joseki : -) dh
>
> David Fotland wrote:
>
> > I'd like to do a winCE version of Many Faces, but I'm putting it
> > off until the winCE API gets a little closer to the win32 API. I'm
> > hoping to have a single GUI source with a few ifdefs
------- CUT --- CUT --- CUT --- CUT --- CUT --- CUT --- CUT -------
X-Sender: mueller@xxxxxxxxxxxxxxxxx
Message-Id: <v02140b02b12a70185c37@[192.31.197.141]>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Date: Tue, 10 Mar 1998 13:50:35 +0900
To: computer-go@xxxxxxxxxxxxxxxxx
From: mueller@xxxxxxxxxxxxxxxxx (Martin Mueller)
Subject: Update of AGA Computer Go pages
Cc: shaevel@xxxxxxxxxxxxxxxxx
X-Sequence: 56
Precedence: list
Dear Computer Go fans,
my January 20 update of the AGA Computer Go pages has finally gone online.
http://www.usgo.org/computer/index.html
There is lots of interesting new and updated information. See
http://www.usgo.org/computer/updates.html#update2
for a summary.
The html pages should be OK.
Unfortunately, the AGA's internet provider recently moved from UNIX to
something else, presumably Windows, which is not working too well. E.g. the
types for the plain text and sgf files are not sent correctly, and some
graphics are broken. AGA webmaster Jeff Shaevel <shaevel@xxxxxxxxxxxxxxxxx> is aware
of these problems and working to resolve them. Please bear with us in the
meantime.
Note that the AGA Computer Go pages are updated with low frequency. I have
created a local page here at
http://www.etl.go.jp/etl/suiron/‾mueller/cgonews.html
for more recent information. (Unfortunately, some big commercial sites such
as netcom and aol are currently blocked from accessing ETL. (Why??? Don't
ask me...))
Thank you, and sorry again for the glitches
Martin
------- CUT --- CUT --- CUT --- CUT --- CUT --- CUT --- CUT -------
Sender: pjl@xxxxxxxxxxxxxxxxx
Message-Id: <35050D73.5E65@xxxxxxxxxxxxxxxxx>
Date: Tue, 10 Mar 1998 09:52:51 +0000
From: "P.J.Leonard" <P.J.Leonard@xxxxxxxxxxxxxxxxx>
Organization: Applied Electromagnetic Research Centre
X-Mailer: Mozilla 3.0 (X11; I; OSF1 V2.0 alpha)
Mime-Version: 1.0
To: computer-go@xxxxxxxxxxxxxxxxx
Subject: Re: Computer Go Tutor
References: <6297CD447F92D11199F5006097BA9D1E65F2@xxxxxxxxxxxxxxxxx>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Sequence: 57
Precedence: list
Scott Dossey wrote:
> There is one major advantage to the concept of having a "computer go
> tutor" integrated with a go playing program--and that is having exposure
> to how the program thinks....IE--why and how it makes mistakes.
>
> I think many of the best programs could benefit from having a "go tutor"
> add-on module, just so the programmer has some degree of visibility as
> to what the program is thinking rather than having to walk through more
> abstract data structures to figure out what's going on when the AI makes
> a bad move.
>
> Furthermore, graphical debugging displays of what the program "thinks"
> it owns and to what degree could be very helpful.
>
> I'm sure some of you are already using such techniques.
In pubgo I have an option to search in step mode (this
takes an long time). You can also explore the search tree
after a search. This is useful to see if the computer
has looked at a sequence. Each node has got a goal which
is reported as you navigate the tree.
One interesting problem is the reuse of
the search tree for a different goal. For example an initial
search could be driven by string tactics the goal is to
keep the string alive. After this search you may wish
to switch goal. Now the goal might be to see
which forcing sequence builds the best territorial frame
work. I have only just started working on this switching of
focus within (the same branch of) my search.
In some cases a seqeunce which leads to a tactical failure (loss
of a string) can be the best sequence in terms of territory :-(
I would be interested to hearwhat other people do about this problem.
cheers Paul.
------- CUT --- CUT --- CUT --- CUT --- CUT --- CUT --- CUT -------
Date: Tue, 10 Mar 1998 11:48:33 +0100 (MET)
Message-Id: <199803101048.LAA13717@xxxxxxxxxxxxxxxxx>
X-Sender: darren@xxxxxxxxxxxxxxxxx
X-Mailer: Windows Eudora Light Version 1.5.2
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
To: computer-go@xxxxxxxxxxxxxxxxx
From: Darren Cook <darren@xxxxxxxxxxxxxxxxx>
Subject: Analyzing The Programs Thoughts
X-Sequence: 58
Precedence: list
>> Furthermore, graphical debugging displays of what the program "thinks"
>> it owns and to what degree could be very helpful.
>
> In pubgo I have an option to search in step mode (this
>takes an long time). You can also explore the search tree
>after a search. This is useful to see if the computer
>has looked at a sequence. Each node has got a goal which
>is reported as you navigate the tree.
I have an option to output log() statements to a log file. Sometimes these
can grow very large, but I can use a text search to only look at the
sections I'm interested in, which is much easier than using a debugger to
step through a recursive function. It also puts a time stamp on each log
entry so I can do crude profiling. I track recursion depth and have it
indent in the log file based on the depth.
When I'm just interested in one particular thing I'll comment out some of
the log() functions, or add "***" to the front of the text it outputs so
they are easy to find with a text search.
I also find _ASSERTE() (as its called in Visual C++) to be very useful, and
in recent code about 20-30% of the lines are now asserts. This is not so
much for catching NULL pointers (which the OS catches) but to find
parameters and variables with strange values.
Sometimes this discovers bugs that are not appearing in any visible way
except they make the program play (even more) stupidly (eg. a value is being
passed for white instead of black in one function, which causes it to think
the group is alive rather than unsettled, which causes it to tenuki rather
than defend). Other times it catches inefficiency bugs (eg. analyzing an eye
point it has already analyzed).
Darren
------- CUT --- CUT --- CUT --- CUT --- CUT --- CUT --- CUT -------
X-Sender: siabod@xxxxxxxxxxxxxxxxx
X-Mailer: Windows Eudora Version 1.4.4
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
To: computer-go@xxxxxxxxxxxxxxxxx
From: siabod@xxxxxxxxxxxxxxxxx (siabod)
Subject: hello
Message-Id: <E0yDTzZ-000046-00@xxxxxxxxxxxxxxxxx>
Date: Fri, 13 Mar 1998 12:55:18 +0000
X-Sequence: 59
Precedence: list
(?) hmm.. is this list still alive, i havn't recieved mail
for a week now
Jeroen Brilleman
------- CUT --- CUT --- CUT --- CUT --- CUT --- CUT --- CUT -------
Date: Fri, 13 Mar 1998 12:04:54 +0100 (MET)
From: Jan Ramon <Jan.Ramon@xxxxxxxxxxxxxxxxx>
To: siabod <siabod@xxxxxxxxxxxxxxxxx>
Cc: computer-go@xxxxxxxxxxxxxxxxx
Subject: Re: hello
In-Reply-To: <E0yDTzZ-000046-00@xxxxxxxxxxxxxxxxx>
Message-Id: <Pine.GSO.3.95.980313120442.1255v-100000@xxxxxxxxxxxxxxxxx>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
X-Sequence: 60
Precedence: list
On Fri, 13 Mar 1998, siabod wrote:
> (?) hmm.. is this list still alive, i havn't recieved mail
> for a week now
>
> Jeroen Brilleman
it still is.
jan
------- CUT --- CUT --- CUT --- CUT --- CUT --- CUT --- CUT -------
X-Sender: siabod@xxxxxxxxxxxxxxxxx
X-Mailer: Windows Eudora Version 1.4.4
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
To: computer-go@xxxxxxxxxxxxxxxxx
From: siabod@xxxxxxxxxxxxxxxxx (siabod)
Subject: Thanks Jan
Message-Id: <E0yDUIq-00009f-00@xxxxxxxxxxxxxxxxx>
Date: Fri, 13 Mar 1998 13:15:12 +0000
X-Sequence: 61
Precedence: list
Thanks Jan :-)
Jeroen Brilleman
------- CUT --- CUT --- CUT --- CUT --- CUT --- CUT --- CUT -------