• 検索結果がありません。

Groupware Support for Kansousen of Shogi-香川大学学術情報リポジトリ

N/A
N/A
Protected

Academic year: 2021

シェア "Groupware Support for Kansousen of Shogi-香川大学学術情報リポジトリ"

Copied!
2
0
0

読み込み中.... (全文を見る)

全文

(1)

Groupware Support Design for Kansousen of Shogi

Hiroyuki Tarumi Yuki Hiraga Toshihiro Hayashi

Faculty of Engineering, Kagawa University, Japan

tarumi@eng.kagawa-u.ac.jp

Abstract

Figure 2 depicts the moves of one game and some of its variations. In case of shogi, one game consists of about

In this paper, we will give a requirement analysis for Kansousen of shogi – Japanese chess. Kansousen is a reviewing discussion process of a Shogi game after the game is over. In case of Internet games, there have been no convenient tools for Kansousen. We analyzed the Kansousen process and introduced four functions to support Kansousen via the Internet.

Chat Window Game Board

Chat Window Game Board

Figure 1. An Example of Internet Shogi

76歩 34歩 66歩 33歩 52銀打 72龍 55馬 59角 79玉 投了 [Game Start] [Game End] 51龍 59角 79玉 78歩 同龍 同歩 32銀 31銀 positions to discuss variation variation variation

record of the moves 76歩 34歩 66歩 33歩 52銀打 72龍 55馬 59角 79玉 投了 [Game Start] [Game End] 51龍 59角 79玉 78歩 同龍 同歩 32銀 31銀 positions to discuss variation variation variation

record of the moves

Figure 2. Variations of the Moves in a Game

1. Introduction

Shogi (Japanese chess) is one of the good research targets for computer scientists. However, most of the research activities have been dedicated to develop strong AI programs for the game. Current AI programs are enough stronger than usual human players. Most players no longer need stronger programs.

The next thing for researchers to do is to provide more enjoyable environments for shogi games, whether the game is between human players or between human and AI. We are focusing on kansousen between players for shogi games on the Internet.

Kansousen is more popular with shogi games than

those with chess or Go games. It is a review and discussion process between players after the game. They discuss to seek better moves at particular game positions. In case of internet games, it is obviously a networked remote discussion on a set of structured data and their state transitions, to which the concept of groupware can be applied.

In this paper, we describe the requirements analysis for groupware to support shogi’s kansousen.

2. Shogi’s Kansousen

Figure 1 is an example of interface of internet shogi game. Below the shogi board, there is a small chat window for both players to communicate. Kansousen is usually given in this chat window, but it is not so active because only chatting is the communication method.

58

(2)

100-120 moves in average. In kansousen, two players (and in some cases some observers) have discussions about the moves and its variations at particular positions. However, as we described above, kansousen in internet shogi games are less active than face-to-face games, when they can use physical pieces for the discussion.

3. Requirements Analysis

The second author of this paper is one of the shogi players in our university’s team. He is also active as an internet shogi player and has been played more than 10 thousand games. According to his experiences, we have analyzed the requirements. The following four functions are considered to be high in priority for kansousen on internet shogi games.

3.1. Position Search

To start the discussion, players have to decide one position of the moves to discuss. In current internet shogi systems, the search for positions have to be done by clicking “forward” and “back” buttons, or directly pick up one of the positions in the moves record.

However, it would be more convenient for players if they can search a position by chatting in natural languages. For example, if a player inputs “When I put a rook (hisha) ...” the system would be able to understand the sentence and search candidate positions. It would not be so difficult because words to describe positions are limited.

In some cases, the search function needs more sophisticated algorithm. For example, players need to search a position where the battle started --- i.e., where a pawn first contacted an enemy’s piece.

3.2. Input Support

People of wide generation are users of internet shogi sites. There are many players younger than 10 years old or older than 70. Chatting with a keyboard is not an easy task for them. For such people to enjoy kansousen, it is necessary to provide some input support functions. Typically interfaces with voice recognition are required. Voice recognition with the position search function above would help such players in kansousen. Here, frequent words used in kansousen are limited so that the development of input support functions would be not so difficult.

3.3. Graphical User Interface

Discussion in kansousen is mostly on variations from a position in the game. In such discussions, it is often the case that players compare two situations, one that appeared in the played game, and one that might appear in a variation. However, current software for internet shogi has only one board on the screen.

The reason why graphical interfaces of internet shogi software is poor is it is just designed to play games, not to have discussions. It is possible to re-design a totally changed interface as an integrated environment for playing shogi games, discussing it and letting people including observers communicate each other.

3.4. Supports by AI

As described in section 1, AI shogi playing programs are enough strong comparing with usual amateur players. When human players are in kansousen, they discuss variations of the moves. But they would possibly miss some important variations. It is possible that an AI program give suggestions on important variations that are missed in the discussion. Such instruction from AI program would help players’ development.

It should be noted that the discussion must be controlled by human players and AI programs must give minimum instructions. Most human players take many mistakes in shogi games. To keep their motivations good, AI program should point out only a few important mistakes by humans.

4. Conclusions

In this paper, we pointed out that kansousen of internet shogi games is a good research target for groupware. We have analyzed requirements in kansousen, and showed four high-priority functions that should be supported.

Kansousen is a important communication process that

improve players’ shogi skills and also their communication skills. We hope our research will help expansion of shogi communities.

We are now developing some of the proposed functions.

5. Acknowledgements

We would like to thank Prof. Takeshi Ito and Prof. Hitoshi Matsubara for their advice. This research is supported by Hayao Nakayama Foundation for Science & Technology and Culture.

59

Figure 1. An Example of Internet Shogi

参照

関連したドキュメント

that of the majority rule, both as functions of the distribution parameter and the

If f (x, y) satisfies the Euler-Lagrange equation (5.3) in a domain D, then the local potential functions directed by any number of additional critical isosystolic classes

Aydi, “Common fixed point results for mappings satisfying ψ, φ-weak contractions in ordered partial metric spaces,” International Journal of Mathematics and Statistics, vol..

The case n = 3, where we considered Cayley’s hyperdeterminant and the Lagrangian Grass- mannian LG(3, 6), and the case n = 6, where we considered the spinor variety S 6 ⊂ P

[r]

This is applied to the obstacle problem, partial balayage, quadrature domains and Hele-Shaw flow moving boundary problems, and we obtain sharp estimates of the curvature of

These results are motivated by the bounds for real subspaces recently found by Bachoc, Bannai, Coulangeon and Nebe, and the bounds generalize those of Delsarte, Goethals and Seidel

The aim of the present paper is to establish some new linear and nonlinear discrete inequalities in two independent variables.. We give some examples in difference equations and we