21 December 2005

Hello Kitty USB key from iodata

You can now get a Hello Kitty USB key from IOData.
I think I know what inspired them -- Han Solo encased in carbonite.

Hello Kitty USB key

12,000 yen
IOData Corporation

(Tip to Akihabara news for the USB key info)

Digg this story.

15 December 2005

Web 1.0 Summit: My Garage, San Francisco

Merlin Mann takes back the buzz. It's not about Web 2.0, it's about bare HTML, baby!

09 December 2005

Will Wright on success and Failure in the Sims

Success and Failure in the Sims
Originally uploaded by ptufts.
See my earlier post on part one of the talk.

In the second half of his talk at When 2.0, Will Wright talked about success and failure in the game the Sims. Players don't mind failure, as long as they understand why they failed.

In the Sims, as in real life, there are lots of ways to fail. The slide shows what can happen on the road to success.

The game designers spent a lot of effort illustrating the failures, and players enjoy exploring them. I thought back to SimCity and how you could have traffic jams, crime, fire, earthquakes, and even Godzilla stomping on buildings, and I remembered how it was fun to actually push the limits of the game to explore how to make things break, and then how to make them work.

Good games allow people to explore and think about things more broadly than they might have otherwise. Failure in a game world is OK, and by failing, you can make a better model of how things should work.

07 December 2005

San Francisco street sign

Originally uploaded by Jef Poskanzer.
San Francisco is political, and it extends to the Department of Public Works.

Will Wright's presentation at When 2.0

Will Wright, creator of SimCity and the Sims, gave a great talk yesterday at the When 2.0 at Stanford.

He described how he designs games, and how a building a model of something can help in understanding the relationships between the objects.

To illustrate this, he described the following problem. Let's say I want to invite you to my home. I could show you a satellite map of the area and let you find your way.

But this isn't very useful. There's much more detail in the image than you need to find my place, and it gets in the way.

Instead, I could give you a map like this.

This isn't as realistic, but as a model it captures the important information and presents the relationships clearly.

When he was designing the Sims, Will started with a physical model of the environment. Little model houses on a green hilly surface with paper streets and blue water. The physical design allowed him to think about the sort of interactions that could occur in the game, before he started writing algorithms. He said that Robert Louis Stevenson took this approach with Treasure Island. First, he drew a map of the island, with places and currents charted, and then he imagined himself at each point on the island, to see what it would be like to live there. Stevenson spent three weeks doing this before he began writing.

Will then talked about game design, and how many games (linear board games, chess, Myst, Doom, and the Sims) have topologies that a gamer quickly recognizes.

For instance, a board game like Life has a linear path, with the game state explicitly displayed on the board.

Chess, however, has a tree-structure, where every move you make, and every counter-move your opponent makes, leads to a complex branching topology that eventually leads to an end.

Myst too has a linear sequence. You must solve a sequence of puzzles in order to proceed. Doom and other level based games allow more freedom, but this collapses at the end of the level (you can do whatever you want, but you have to kill the Boss, or get the key, and the new level starts cleanly from that point).

In the Sims, he wanted players to pursue a balanced approach to the game. If a player pursues a solely social or material path, they eventually reach a local maxima where they can rise no further. In order to get on to a balanced strategy, the player actually takes a hit (the yellow area in the diagram represents a lower score than the light blue) before they again progress.

It was a great talk. I'll post more from it tomorrow.

(Digg this story and read the followup)

06 December 2005

Coke, soda, pop.

One of my contacts on Flickr posted this great map. It's a county-by-county breakdown of the preferred term for "soft drink."

Click through on the image for a larger version.

When 2.0 conference

When 2.0 conference
Originally uploaded by ptufts.
I'm at the When 2.0 conference today. In this photo, Esther Dyson is hosting a panel discussion on Time and Functionality. The speakers are Munjal Shah (Riya), Tantek Celik (Technorati), John Arenas (Worktopia), and Ben Cruz (Demand ID Systems).

Riya is doing some interesting work in applying face recognition to auto-tagging photos. Even with inexact algorithms (the best face-recognition software is far from perfect), the problem becomes much easier when you only have to deal with the hundred faces that a particular user might have photos of.

Tantek Celik of Technorati spoke about applying microformats to online organization of information. This is one of those places where a simple solution might work better than the best-designed semantic web system. Microformats are simple enough that you can use them on a web site without having to adopt XML, new publishing software, or in fact, anything beyond what you're already using.

Worktopia has a really interesting business model, providing on-demand meeting and workspace to groups that need it. Their premise is that workforces are increasingly mobile and distributed, and so traditional (fixed) office space does not serve the needs of these new workers. Instead, they occasionally need space (for meetings, kickoffs, and such) and Worktopia is aiming at the heart of this niche.

Finally, Ben Cruz of Demand ID talked about better ways of reaching people who would like to see live events. In the current reality, for many events, including live music, most tickets are not sold, but are given away in the week before the show as the promoter realizes the show isn't going to fill the venue. Promoters do this because news of a half-full venue, if it made it down the line to later tour stops, would be a self-reinforcing failure.