3D view of gallery installation, showing 4 projectors. To view, copy link location:
http://marek.someprojects.info/gallery/gallery2.wrl
and open location in InstantPlayer, or other vrml/X3D player.

“Menard (perhaps without wishing to) has enriched, by means of a new technique, the hesitant and rudimentary art of reading: the technique is one of deliberate anachronism and erroneous attributions. This technique, with its infinite applications, urges us to run through the Odyssey as if it were written after the Aeneid, and to read Le jardin du Centaure by Madame Henri Bachelier as if it were by Madame Henri Bachelier. This technique would fill the dullest books with adventure. Would not the attributing of The Imitation of Christ to Louis Ferdinand Celine or James Joyce be a sufficient renovation of its tenous spiritual counsels?”
“Pierre Menard, Author of Don Quixote” (Borges)

Today we made some significant progress on the development of both the architecture and the implementation of the client/server infrastructure.

Architecture
On the client side, we will have a small “client manager”, written in Java, which will communicate with both the VRML player and the server, coordinating the two.

Implementation
A super basic version of the Client Manager is now working. All it does is connect to an InstantPlayer instance running on the same machine, and instantiate protos.

The next steps of the implementation process will be:

1. a basic version of the Server with a simple database
2. an improved Client Manager that can determine how the user has moved, and communicate this motion to the server
3. logic on the Server to share this motion information with all other Client Managers
4. logic on the Client Manager to receive motion/position information for other users and update InstantPlayer instance accordingly

The first workshop on Sunday May 10th is to establish elements of a city. These elements will become the specification for the generated objects in the project [here][now].

part 1:
List all the elements of a city, each on a separate piece of paper. If there is not enough paper, cut up
more.

part 2:
Create about 4 groups, divide the pieces of paper between the 4 groups.
Create a single diagram that structures all the elements in a relationship to each other.
You can only use 12 words in the diagram. You can’t use the pieces of paper. You can draw anything
you want.

part 3:
As a large group, look at the 4 structures and create visual criteria, forms and behaviors, that characterize the pieces. For example, politics has an abstract quality, whereas an assembly hall is a concrete part of the city. Criteria could be complexity, color, scattered/orderly, tall, skinny, dense/sparse, etc.

These spatial and temporal attributes become the specification for the generated objects. So a fast
red thing could be a fire-engine and a tall, fat, etherial structure could be a bureaucracy.

From May 17th to 29th Rory and Marek are going to Montalvo Art Center for a residency. The bulk of the code and the avatar generation will take place at that time.

When people enter the site, they enter adjacent to the last person who was there. Near the entry point City is most congested, full of structures left behind, those that did not participate. Structures that are visited often are full of ‘energy’, their colors and textures are dependent on people visiting them. There are also structures beyond the visible, the farthest out that people have ventured, the farthest out that people have congregated. We are wondering how to create a series of breadcrumbs, dots, lines or signposts that show the way to these hidden places.

Each avatar has a proximity sensor. When a person crosses this sensor, the sensor keeps track of how long a person has been there. When they leave the avatar or simply exit, the time they were there is sent to the server. When the server loads the main file again, it sends a list of criteria for each element in the file.

Avatars, which have no relationship to the human body at all, are generated afresh for each person entering the space. On which basis should these avatars be created? Two criteria emerged:

The avatars be ‘topical’ based on the date that they are made. This includes news feeds, recent flickr images, design/culture/political blogs etc.

The avatars be made with all the elements of a city:
cloudy shape, big city block, fractured bits of glass,
suspension bridge, skyscraper, cantilevered house, picket fence,
politician, minister, businessman, teacher, cop, thief,

We start by listing the above, and ask a question: What are the elements of a city? If their ‘element’ is not available, we ask them to put it into a form. This creates a list of the elements of the city.

We then look at the list and create a series of ‘types’. These are forms that do not necessarily have a one to one relation to the ‘elements’. Using a set of transforms, color/texture/scale/speed, we create a set of criteria by which to create any of the elements of a city.

Finally, when people first enter the City online, they can choose a range of elements, say ‘fireman skyscraper’ or ‘bridge tycoon’, and an avatar is created for them out of the transformed types.

People can also suggest and upload alternative forms for a particular element of the city, these are then added to the pool of choices.

Currently we are testing InstantReality as a possible engine for the project.
An x3d player, it reads vrml!

Test Avatar

For this avatar I did the following:
- Took a headline from Google News… about AIG.
- Put every word into a Flickr search, selected first image from each search, with creative commons.
- Attached each texture to a part of a random model.

We will create a workshop to evaluate what criteria to use for avatar creation.

Top left
Plan of gallery space showing real people and virtual space together. The green rectangle is the floor plan. In the center are the location of the 4 virtual cameras, their views are shown in the bottom row (below).

Person A is a fractured sphere
Person B is a series of tiled cubes
Person C is a star shape
Person D is a particle system

Top two middle and right
Plan and elevations of the virtual avatars with fragments of real people (for context only)
The dark green rectangle is the floor plan of the room.

Above
Cameras 1 to 4 showing the views as they would be projected on the gallery wall.

So, a person walks into the room. An avatar is attributed to them. Others enter, get avatars.
There are 4 ‘virtual’ cameras in the space. Each one is rendered to one of the 4 walls of the room.
The cameras can ‘move’ in the virtual space based on location of people.
If people are all crowded to the left, the cameras move left, etc.