Generating Personalities
Interactive designer Romain Cazier and Bright co-founder Abdel Bounane on the collaborative digital art experience, Davatars.
We were recently commissioned by ERDF to create an original collaborative experience for the unveiling of its new brand image. As France’s main energy service provider, the company wanted to use the occasion to further emphasize its position as a player in today’s digital ecosystems.
With that in mind, we thought up a collaborative experience that allows anyone to take part in ERDF’s logo reveal. By visiting davatars.fr and loging in with either their Facebook or Twitter account, a unique avatar is generated in the image of their social behaviour. To ensure the uniqueness of each avatar, we pinpointed specific activity indicators such as how often you might post, like, favorite or share content, to create an amusing mirror image of your social presence.
In the days leading up to the logo unveiling, all the creatures will accumulate on davatars.fr, hinting at the new logo (to be unveiled on May 31st.). Each participant’s online persona will ultimately have had a hand at creating the company’s new brand image, all while placing data-art and generative design front row and center.
One of Bright’s goals is to give digital artists a real platform to shine on. As interesting as the technology is — and it is of course quite fascinating — without curious minds to explore the creative opportunities it offers us, none of this would be possible.
For this project, we chose to work with Romain Cazier, an interactive designer and graduate from the prestigious ECAL design school.
The experience was a eye opening to say the least, so we thought it appropriate to share some of the journey. Below, he answers some questions about his process, the challenges of this type of project and what the collaborative process was like for him.
You’ve just finished working on Davatars. Can you tell us about some of the challenges you’ve encountered during the creation phase?
It was the first time I’ve had to deal with a fully generative project. So many parameters have to be taken into account! I had done some projects before involving generating creatures, but they were way more narrow in terms of their criterias needed to define their physicality. Having to deal with such a wide range of possibilities was a big challenge for me. How does a parameter get interpreted? How do I translate the parameters into something that’s pleasing to the eye? And most importantly, how to ensure that every single creature meets these requirements?
Another big thing for me was how to make each creature dinstinctly unique. It was important to me that the general look and feel of the creatures remain very simple. So I had to figure out how to properly juggle the different specifics in order to coax the best results out of my code. Each creature had to really reflect its creator.
Our client, ERDF, is the main energy provider and distributor in France. The company is not only changing it’s brand image but it’s transforming itself into a digital player with an awareness of the Smart City. The aim with this artwork was to really showcase this transformation through the use of data and networks. Tell us a little bit about how you achieved this using generative creatures.
The data part is really quite simple. Based on social data culled from each user, we generate a unique creature. This creature is then added to a network where it can live by itself. But it can also interact with the other generated creatures . They all form a swarm that doesn’t appear to be doing much at first, but when they all start to gather and team up, they form the logo.
We’ve been pretty impressed with the variety of different characters you were able to generate! Can you tell us more about the different options and features you’ve included to make them so unique?
They all share the same basic composition. That means a body, some arms and a pair of eyes. I then defined parameters that characterized each element’s width, height, distance, roundness,or color for example. All of these specifications give the creatures little idiosyncrasies that set them apart from each other, kind of like DNA!
How was it iterated with our studio department? On the project as a whole but especially when it came to developing the characters ?
I made a custom tool that generates and tests all sorts of combinations, giving the user the option of setting the value of each parameter. But I also included an option to randomize everything, which gave users the possibility of just sitting back and seeing what happens.
When I came to Bright to do a one-day workshop with the studio team used this tool to test out each of the values and see which combinations were producing the best results. Once everyone agreed, I refined how the data we’d grab from Facebook and Twitter would be interpreted. We stayed in pretty close contact the whole time, and I made a few trips to Paris to hash out the final look and feel of the creatures.
We reached out to you after discovering REC.ALL. Can you tell us more about this project and the context in which it was produced?
Rec All is basically a game that I produced during my diploma at ECAL (Ecole Cantonale d’Art de Lausanne). It’s a really simple puzzle game where you have to find all the collectible items in each level. But in order to get the items, you have to make these very strange creatures. Strange because you have to set their movements, which once defined, are basically set for life. They’ll follow your instructions indefinitely. So the goal in the end is to produce the right gestures needed to succeed.
For this project I worked a lot on both the physical and the behavioral aspect of the creatures. It was important to me that they be coherent. For example, using one finger you could have a creature that could only catch one collectible item at a time. But using two fingers might allow you to generate a creature that could catch two items at the same time… That sort of thing.
What do you think generative creation in art & design? Do you think this type of technique has a lot of potential?
Well, I can’t say that I’ve done much in the field yet, so I don’t really feel like I have a firm enough opinion to give. But I will say that generative creation is a really great way of appropriating and using the computer for what it really is at it’s core.
In terms of potential, generative creations represent a way to very quickly achieve some intriguing results. But what’s most exciting are the unexpected outcomes that can result from this type of approach… Things we could never have thought of or come across otherwise.
Generative creation implies feeding a set of predefined rules to the computer — constraints that basically define your artwork — which it will then use to compute every possible iteration on your set of rules. Just in terms of potential outcomes this kind of huge. Add to that how quickly you can see the results of your ideas and the technique really speaks for itself. And of course there’s always the charm of unpredictability inherent to algorithms. This really opens the door for endless hidden discoveries during the creative phase, things no one can really foresee or anticipate otherwise.
Out of all the creatures you made (and there were many,) can you pick your top three for us and explain why you like them so much ?
To be honest, that’s a tough call. They all look pretty cute and weird to me. But if I have to choose, these three are among my favorites.
Can you tell us more about the next projects you’re working on?
I have a few websites in the works right now, that I’m developing with my colleague Emilie Pillet. And this summer I also plan to keep working on Rec.All. I want to hopefully release it by the end of the year.
