Code from crayons: new work from Douglas Bagnall at Te-Tuhi

For those put off by the gun-wielding heroes and photorealistic environments of videogames, this Saturday is your chance to create your own. Douglas Bagnall, who has previously shown works like A Film-Making Robot at Window, is exhibiting his latest project at Te-Tuhi, which transforms crayon drawings into game worlds via some clever coding.


Bagnall's Video Game Machine is part of a host of recent interactives which mix physical reality like shadows and drawings with screen-based additions. Philip Worthington's Shadow Monsters tracks users, adding spikes to their arms, medusa tendrils to hair, and the ability to throw flames and projectiles.



MIT's sketch engine takes simple physics diagrams drawn on a whiteboard and digitises them, translating crude down arrows into gravity, force, and inertia. The demo is dry, the graphics dull. The potential is not.



Flash-based physics toy Line-Rider became a smash success over the last year as users took its simple mechanic of drawing lines for a sled to an art form. Massive levels like the one shown below, or others with up to 22,000 lines are entered into competitions in communities like I Ride the Lines. Little Big Planet, an Xbox 360 title poised for release, hopes to capitalise on the same idea by letting users create and post levels. Players assemble game worlds from a variety of objects which are translated into game worlds with deep physics applied.


With any set of rules and scripts, from Bagnalls 'robots' to interactive narrative such as Façade and simple 'toys' like Line Rider, the fascination lies in 'gaming the system' - finding the edge cases, glitches, and grey areas. The algorithm behind Film Making Robot favoured oversaturated images, creating a very selective memory of images sourced from Wellingtons bus routes. Only hours after Façade was released, players were already trying to break it, ignoring any goals like reuniting the protagonists and instead posting scripts featuring serial killers in an effort to defeat the natural language scripting. The top rated movies on LineRider comps are not huge hills or triple backflips, they're 'microquirk tracks', hardcore users who can place a dot in a specific location, confusing the game code and causing havoc with gravity.

Te Tuhi Video Game Machine
Douglas Bagnall
17th November 2007 – 10th February 2008

Update: James McCarthy also performed using a site-specific wall work comprised of high-tension wires. He repeated the performance on Tuesday with some members of experimental music organisation Vitamin-S. Thanks to James for the pic.

Oliver wins Opensource award, progresses videogame

New Zealander Julian Oliver just won the Open Source for Creativity award for SelectParks, the information-rich website and blog focusing on art-based games we've mentioned in the past. The site is built on open-source software like PHP Nuke, and allows syndication through standards such as RSS.


Oliver's also progressed that Escheresque augmented reality game we blogged about a few months ago. Still in alpha stage, the latest version, titled levelHead is starting to take shape. A simple block with painted identifying marks on each side is placed in front of a camera, which reads these and projects the correct view of the game world. Tilting the block directs the tiny avatar up stairs, through doorways and eventually to an exit in a delightfully simple, practical use of AR.

Pushing (and pulling) the interface

A bespectacled, slightly overweight man stands in front of a massive flat screen. With a few ‘pull apart’ motions of his fingertips, he zooms from the continental US into downtown Boston, moving from satellite imagery to a detailed roadmap in a few seconds. He flicks into a photo application, flicking thumbnails around and drawing a selection with his index finger before copying and pasting it. With a throwing motion, he sends a couple shots over to a colleague working on another area of the board. It’s Minority Report, come to life (video).

The unlikely hero is Jefferson Han, a previously unknown researcher at Cornell/NYU before making a presentation on his multi-touch interface at a tech conference last year. Since then, he’s been featured in Wired, sold units to the CIA, and helped Apple implement some touch behaviours into the iPhone.

And while the CIA might find spying faster with Hans interface, many people have responded that it looks fun. Alternate interfaces have abounded in the last few years.

Locally HitLab (Human Interface Technology) New Zealand have been hard at work pushing alternate reality interfaces. Mixing the real world with the virtual, they're an attempt to provide a natural way to work with objects, while overlaying CG elements or feedback. In principle this is fantastic - urban developers move and rotate virtual skyscrapers the same way they'd normally use wooden or plastic blocks - only the work environment now allows them to scale, create roads, and view the whole scheme in the context of a geomapped overlay. In practice the interface is a little glitchy and slow - 3d models stutter around the stage or the interface hesitates when manipulating objects (is that a move or a scale?).

Just yesterday New Zealander Julian Oliver posted about an experimental AR interface he and Simone Jones had created at the Interactivos conference at Media Lab Madrid. "The idea was simple, augment a solid cube with 6 little rooms such that the cube becomes a tangible interface for navigating through an architecture: a mind-game - 'How are the rooms connected?'" The result is a tiny Escheresque cube that seems filled with possibilities. One idea to extend the concept is to develop a mini-game/interactive where the user would guide a tiny avatar through a world by flipping a series of cubes to expose exit points.