## Monday, December 19, 2005

Will Wilkinson well reminds us that collective actions problems shouldn't always be solved by the state. (Along the way he invents some new technical jargon.)

### Actions speak louder than words

The popular Indonesian image of America improved after our aid following the tsunami last year; the same thing has happened in Pakistan since the October earthquake:

ACNielsen Pakistan shows that the number of Pakistanis with a favorable opinion of the U.S. doubled to more than 46% at the end of November from 23% in May 2005. Those with very unfavorable views declined to 28% from 48% over the same period.

## Saturday, December 17, 2005

### Multiplayer Sudoku

Arnold Kling suggests creating Sudoku for two.

Each player gets a sheet of paper with a completed Sudoku on it. Each player keeps his Sudoku hidden from his opponents. Using pencils (or numbered/colored tiles), the players take turns filling in cells of a common Sudoku grid, which is initially blank. Once the common Sudoku grid is full, each player gets a point for each cell of his private Sudoku that is in agreement with the common grid. The highest scorer(s) is the winner(s).

To make the game nontrivial, players are not allowed to fill in a cell arbitrarily. It is not legal to fill in a cell if it would result in a row, column, or block containing two of the same number. Thus, the common grid will be a valid Sudoku once it is completely filled in. If a player is unable or unwilling to fill in a cell, then he may pass. However, if all players pass in succession, then the next player must either fill in a cell or erase a cell.

Obviously, there is no guarantee the common Sudoku grid will be completely filled. Therefore, players should either agree to a time limit or a limit on the number of turns. Also, it would be a good idea to encourage players not to pass unnecessarily. If a player fills in a cell after the previous player passed, then he should get a scoring bonus. My initial guess is that this bonus should be one point, but a more desirable bonus value (not necessarily a whole number) might be found experimentally.

Instead of numbers, use colors. It looks nicer, and I think it would be more fun.I like this idea, except that it's not very interactive. It's essentially a race: who can solve a Sudoku puzzle faster? I can imagine a much more interactive Sudoku game for two or more players. Moreover, no electronics are required.

Have an electronic board that generates the puzzles and keeps track of each player's time. When the game starts, it's the first player's turn. His clock is running. He selects a color and places it on the board (by pushing on the spot--the electronic board will sense both the color he picked and the place he marked). If his placement is correct, it is the other player's turn. The other player's clock starts, and he selects a color and places it on the board. Play continues until one player runs out of time (loss) or makes an incorrect move. If the puzzle is solved with neither player losing, it is a draw.

Each player gets a sheet of paper with a completed Sudoku on it. Each player keeps his Sudoku hidden from his opponents. Using pencils (or numbered/colored tiles), the players take turns filling in cells of a common Sudoku grid, which is initially blank. Once the common Sudoku grid is full, each player gets a point for each cell of his private Sudoku that is in agreement with the common grid. The highest scorer(s) is the winner(s).

To make the game nontrivial, players are not allowed to fill in a cell arbitrarily. It is not legal to fill in a cell if it would result in a row, column, or block containing two of the same number. Thus, the common grid will be a valid Sudoku once it is completely filled in. If a player is unable or unwilling to fill in a cell, then he may pass. However, if all players pass in succession, then the next player must either fill in a cell or erase a cell.

Obviously, there is no guarantee the common Sudoku grid will be completely filled. Therefore, players should either agree to a time limit or a limit on the number of turns. Also, it would be a good idea to encourage players not to pass unnecessarily. If a player fills in a cell after the previous player passed, then he should get a scoring bonus. My initial guess is that this bonus should be one point, but a more desirable bonus value (not necessarily a whole number) might be found experimentally.

*Update*: The first player presumably has a small advantage, so the game should be played repeatedly such that each player goes first the same number of times.## Friday, December 16, 2005

House votes for border fence. It's about time.

## Sunday, December 11, 2005

You know you've been blessed when you have friends, games, Dr. Pepper, DVDs, a computer, and a deep fryer all in the same room.

## Thursday, December 08, 2005

I suppose one could argue that an increase in federal nondefense discretionary spending by a few tenths of a percent of GDP is "modest," but I just can't bring myself to apply that word to hundreds of billions of dollars. Anyhow, looking at the graph linked to above, the most depressing thing is the almost monotonic rise of Medicare & Medicaid since 1970.

Todd Zywicki on the difficulty of Anti-

*Kelo*reform:One advantage in California is that the reforms will be proposed as ballot propositions, rather than being generated by the state legislatures. This thus avoids the agenda-setting and related public choice problems otherwise associated with enacting reforms designed to tie the government's hands when it comes to taking property.A little technology could make a big difference here. Suppose the Secretary of State of California---or any other state---were to maintain a website listing all potential ballot propositions that have been submitted. Further suppose this website allows voters to "sign" petitions corresponding to propositions they want to be on the ballot. Then the cost of getting a proposition on the ballot would be almost all advertising. The number of digital signatures required would doubtless be higher than the number of physical signatures required precisely because digital signatures cost less. Thus, the barrier between a propositon and the ballot would be more popularity and less money.

Nonetheless, the public choice problems remain daunting, but perhaps not overwhelming. As Tim writes, "Now that this proposition has been submitted to the Secretary of State, it must get enough signatures to qualify for the ballot. That costs money, and that’s one of the big problems this initiative faces. If it gets on the ballot, polls suggest it would pass overwhelmingly. But getting it on the ballot requires money, and who’s going to pay it? There’s little money to be gained in eminent domain reform...."

A very resourceful maze algorithm page. (Hat tip: Geekpress.)

## Monday, December 05, 2005

### Kill two birds with one stone

Ever concerned about the growth of government, I used to think that when it comes to Social Security benefits, the important thing is that their growth be significantly slowed, and that how isn't particularly important, except that it be politically feasible. I've recently changed my mind. The goal should be raising the retirement age. No other approach can so seemlessly address both the increasing costs of Social Security and of Medicare. And the projected fiscal shortfalls are earlier and bigger for Medicare than they are for Social Security. Moreover, even trimming promised benefits of the distant future is politically painful, so it is easier if the cuts are agreed to all at once.