Art from code - Generator.x
Generator.x is a conference and exhibition examining the current role of software and generative strategies in art and design. [Read more...]
 
Archive for June, 2005
 
Art+Com: Terra:vision

Art+Com: Terra:vision

Google has announced a new service called Google Earth, providing a coherent interface to satellite imagery that allows you to “surf” the earth and look at information mapped directly on top of it. Microsoft is hot on Google’s heels with their MSN Virtual Earth, set to launch in July 2005 providing similar features. With any luck, there will be ways of hacking with these services.

Interestingly, electronic artists did this over 10 years ago. The German group Art+Com created a project called Terra:Vision way back in 1994, featuring a large adaptive-resolution satellite image database, streaming of data sets over the internet, fully interactive VR flyovers and mapping information to geographical locations. Back then it needed a Silicon Graphics Onyx to run, these days it could probably be ported to a high-end PC.

 
PixelRoller

rAndom: Pixelroller

rAndom international have developed a paint roller that rolls out pixels on just about any medium. Their first prototype was called the LightRoller, and used phosphorent inks.

rAndom are Flo Ortkrass, Stuart Wood and Hannes Koch, all graduates from the RCA in London. They have done a number of projects related to alternative displays, including PixelTape, Temporary Marker and Watch Paper.

One obvious reference is Jürg Lehni’s famous grafitti robot Hektor, or his more recent interactive drawing tool Rita (not much online about her yet, she premiered at Tensta Konsthall recently.) A more political alternative would be the GrafittiWriter from the Institute for Applied Autonomy.

 
Jun 30/05
19:26
TAGS:

Network deviousness is a poetic-sounding geographical term defined as follows:

In spatial analysis, network deviousness is the discrepancy between the lengths of the actual routes in a network and the straight-line distance between the places linked up. For any pair of places on the network it can be measured by the detour index. The detour index is a measure of how directly movement may be made on a network. It is calculated as the ratio of the shortest actual route distance between a given pair of nodes and the direct, straight-line or geodesic distance between the same two points.

From the Mapping Hacks book by Schuyler Erle, Rich Gibson, Jo Walsh. Description attributed to technical reviewer Edward Mac Gillavry.

 
Lisa Jevbratt: 1:1

Lisa Jevbratt: "1:1"

Martin Wattenberg: Shape of Song

Martin Wattenberg: Shape of Song

Databases, seismic data, Computed Axial Tomography scans, Mozart’s symphonies, the first 1000 prime numbers: All these are large data sets containing patterns hidden from view unless presented in a human-readable form. With the increasing power of personal computers it is now becoming possible to visualize data sets that previously would have been inaccessable to anyone but researchers with access to old-school supercomputers. As a result, information visualization has become a fruitful new field of aesthetic exploration.

The theorist Lev Manovich posits that mapping one data set into another is one of the principal operations of computing. He argues that art projects like Carnivore and Lisa Jevbratt's "1:1" produce profound emotional responses despite their being essentially data visualizations. Where the Romantic artists were concerned with the sublime and the un-representable, data art is concerned with making representations of phenomena previously invisible.

On the more pragmatic side of things, designers are employing computational techniques to escape traditional 2-dimensional representations. The results are dynamic software visualizations of complex data like genome structures, version history in documents with multiple authors or power structures in American corporations.

Information visualization theory references:

Some projects & people:

 

Google has just released a public API for their Google Maps service. There have been a number of unauthorized Google Maps hacks posted online (notably GCensus, HousingMaps and Geobloggers), but with the launch of an API developers are granted more stability and legitimacy to their efforts.

In the past Google has released an API for their search engine, allowing non-commercial developers up to 10000 requests per day. Even though Google’s reason for releasing them is its continued quest for world domination, these APIs allow independent hackers to play with some powerful technology. No doubt Yahoo, MapQuest and others will follow suit and provide some welcome competition. Update: Yahoo has released documentation for their Yahoo! Maps Web Service.

Historically, access to good maps has equated to the possession of power, and the geoinformation industry has been a top-heavy field dominated by governments and military organizations. With Google Maps the growing locative media community have access to a large map database to do great things with, even if it’s just to play Pac Man on the streets of Manhattan.

For some in-depth reading, check out the books Google Hacks (by Tara Calishain, Rael Dornfest) and the newly released Mapping Hacks (by Schuyler Erle, Rich Gibson, Jo Walsh).

 

Of no relation to the Generator.x project, the Generator Blog is a gathering place for software that generates content according to simple rules. Or in their words:

This blog is not about those machines used to change mechanical energy into electrical energy. It’s about software that creates software. Software to play around and have fun with.

Some sample generators: Word Generator, Dummy Text Generator, Cat Name Generator. A definite classic, the Blog-o-Matic Generator generates blog entries of the angsty rant type popular with personal weblogs.

Most of these generators utilize simple text grammars applied to a problem of limited scope (i.e. creating a plausible cat name). A more ambitious project of the same type is the Postmodern Essay Generator, powered by the Dada engine for generating random text from grammars.

 
Jun 28/05
23:06
biot(h)ing - swells

biot(h)ing: swells

biot(h)ing is the research-design laboratory of alisa andrasek – creating and studying algorithmically derived structures in virtual and physical environments.
Computational design processes are used to dynamically manipulate the conditions of various surface systems. The presentation of the project “swells” for example shows the different steps how a network of genetically scripted cells is transfered from the screen to the surface of a high-rise building.

 

VVVV is a patch-based visual programming environment for real-time graphics, video processing and installation control. It was created by the Meso group in Frankfurt, who originally designed it as a tool to develop their own projects.

Similar in operation to Max/MSP but focused on visuals and show control, VVVV is Windows only, sacrificing portability for hardware acceleration using DirectX 9. It’s free software, but not Open Source. It’s visual programming style makes it a unique alternative to text-based programming systems like C++, Java, Flash or Processing.

One of the most exciting uses of VVVV is for live visuals. Unlike compiled programming languages, patching systems like VVVV are constantly running, so it’s even possible to “program live”. In performance use the tool feels very responsive and allows live improvisation. Using the AudioIn and FFT nodes, sound-responsive graphics are easy to create. VVVV also interfaces easily with a wide range of input and output protocols such as RS232, MIDI and DMX-512 for show control or installation use.

The most unique feature of VVVV is the “boygrouping” feature. Boygrouping is a way of setting up a piece to render on a cluster of separate PCs using a master-slave distribution setup, allowing easy multiple-projection output.

For examples of the tool in use, see the Screenshot of the day, published by VVVV users directly from the software itself.

 
Lionel Theodore Dean: FutureFactories

Lionel Theodore Dean: FutureFactories

From FutureFeeder: Computer-controlled Rapid Prototyping is old news in the industrial design world, but recently designers have been realizing the full potential of the technology by using it in new ways.

The FutureFactories research project at the University of Huddersfield has designer Lionel Theodore Dean and colleagues using laser sintering techniques to create parametically designed objects that mutate within a set of constraints. Since the production technique always produces one-offs anyway, why not make every object unique?

The Autonomatic symposium at the University of Falmouth in January 2005 discussed the use of such techniques, as well as the implications of computer-controlled techniques on craft traditions like jewelry design. Is craft defined by the involvement of human hands in the production of a piece, or can it include computer-aided dynamic generative designs executed by machines? The papers from the symposium are available for reading in PDF format.

Other designers using Rapid Prototyping in new ways include materialize.MGX. One of their contributing designers, Batsheba Grossman, has long been producing sculptures based on mathematical formulae using the same techniques.

 

The excellent media art theory archive Media Art Net has a section called “Generative tools”. Here different writers deal with issues (sometimes complementary, sometimes opposing) surrounding the fields of generative art, software art and game art. Interesting reading: Generative tools.