I’ve registered a SourceForge project for it as well (link from the page above).
Initial design is just about finished (I know how it will work, again see the above link for details) and I’ve coded some tests.
Next up is planning the milestones and starting on the code proper. Who said “vapourware” at the back? 🙂 It’s past that point, just…
Visually the images draw on Dr. Ahmed Moustafa’s art and the simple photocopied flyers that advertise bands playing in small venues. Conceptually they draw on Dean Motter’s “Mister X” and whoever used the term “postmodern hyperspace” to describe the non-linear space of the mall. The source images came from sketches of The Bentall Centre, Kingston-on-Thames, photographs of Canary Wharf, London, and video footage of a multi-storey car park in Canterbury.
I made these works in 1994. The 386 PCs running Corel Draw let me have a tea break each time I edited a single piece of text, and the more complex images took a couple of weeks each to make. The images were laser printed A4 at 300DPI and digitally photocopied up to A0. They project well, too. You need to see them big to get the full impact.
Minara is “Emacs for graphics”, a programmable vector graphics program editor written using libArt, libSDL and librep. Read that again slowly. 🙂 This is that “tinker under the hood, write your own tools” environment you’ve always heard about.
Minara is a program that can render graphics descriptions, and allows the user to modify those descriptions using scripts in the *same* language. The descriptions and the modifications are all written in a very simple data format / programming language which interfaces with the graphics and UI engine. This enables artists, researchers, reprographics operators, brand designers and many other groups to create generative, dynamic, generic and/or shareable artwork in a format that can easily be converted to standards such as EPS, SVG or Flash.
Most graphics formats are already programming languages or contain programmable elements (PostScript, SVG, Flash). The only major non-programmable graphics format is PDF, which was intentionally designed as a static subset of PostScript. Most graphics programs are scriptable. Editors such as Illustrator and iDraw have scripting languages. But these are in a different language from the graphics description language, and with varying access to the underlying data structures. There’s nothing that is code all the way down or that can be freely extended or modified by user scripting.
An environment that allows the user to write graphics code, generative code, analysis code and processing code all in the same language using the same tools within an illustration program with open code, open APIs and no barriers between content and process will enable greater creativity (and productivity).
A common data format that allows dynamic, generic or richly described content and that doesn’t suck will allow archive-quality (“bitfast”) graphics to be exchanged in open content projects and corporate workflows.
Who and When?
Me to start with, then anyone else who’s interested. At some point in the future.
The choice algorithm is random, but the choices are structured hierarchically so there is structure to the generated descriptions. I’m going to add complexity to it and start working on non-random choices.