Generative Art & Interaction Design

Categories Interaction, Programming

I find generative art to be a strange concept.


I feel the concept is at best too broad in that it would essentially describe all Art and at worst too narrow in that it describes only certain artistic movements in which a generative approach is apparent at a surface level.


Marius Watz: Three stills from Electroplastique, 2005.


EvoEco, Jon McCormack, 2010.


Generative art is created by a seemingly autonomous system, usually a computer algorithm that has room for variance informed by some random element – though analog systems also apply. Some systems are completely chaotic, others enforce various over-arching rules in order to create an artistic focal point, perhaps in order to more faithfully represent existing artistic conventions; abstract paintings, ambient music and light-shows all lend well to this sort of contained generative methodology.


Some digital generative art is so complex – or perhaps, some forms of digital art have exhausted themselves into predictability – that we can produce a near infinite amount of “unique” pieces on command from a single source (see “The Infinite Music Machine”). In this way, the algorithm-system itself is the artwork and it’s offspring a kind of sub-art. It’s an interesting development that has parallels in consumer culture. The movement from the naturally imbued variance of artisan-produced objects, through to the precision and conformity of mass-produced objects and moving on now into an era of mass-produced bespoke objects. Though I believe that there will always be a territory of Art that is unable to be mass-produced simply by virtue of that fact that it isn’t – there may be less of a distinction between the lifeless-via-duplication and the lively-via-mutation in regards to new forms of generative Art.


We can create things that create things. But the things that we create can only create things that we essentially let them create.


If any of the above sounds radical or pointless or removed from reality, consider the humble wind chime.

The wind chime is a generative art object, it’s enforced over-arching rules are defined by the tonality that the relationships between each chime creates (as in, the musical scale that the wind chime adheres to) and it’s random element is the wind itself, an autonomous system (which to discuss further is a philosophical conversation of it’s own).


The wind chime is a great example of how simple constructs can lead to complex, “embedded” or “reactionary” behaviours. It’s beautiful, transitory, unpredictable and fairly non-invasive to it’s environment. Besides sporadic melodies, it’s behaviour also contains information about the wind patterns outside encoded within it’s output.


I’d like to explore what the wind chime is employing for it’s random element – that is, nature – in the context of a complex digital structure. The everyday world is a constant ever changing source of variables with intrinsic relationships between them. These related variables may yield beauty, simply as a result of having to conform to the physical laws of existence (leading to ubiquitous relationships such as the Golden Ratio and Pi). If we data-mine the world around us using computer systems and use that data to inform generative art objects, we could fabricate simulated synaesthetic experiences.


The exploration of this idea is what I am basing my major undergraduate project on throughout 2013. I will be posting about that project on this blog as it progresses.


Following from my earlier post Embracing the Digital Landscape, I think generative procedures will be invaluable for creating modern computer-human interfaces and technologies. As defined above, generative art need not be chaotic. It can be a way to condense, rearrange, reform and “perceptify” data. Rather than presenting data as numbers (or equivalent), with singular meanings, we could attempt to create shapes, colours and sounds that contain multiple threads of data, to be understood at a glance.


David Eagleman (PhD) talking about colour perception in relation to real-world information

Interaction design and UX design often strives to streamline data, present a “cleaner” interface and occasionally to reduce the “depth” of an interface, to bring all the information as close to the user as possible. Generative methods might be useful in this regard.


No Comments

Leave a Reply

Your email address will not be published. Required fields are marked *