1.
Understanding the Needs
of Creative Coders
Towards a Creativity Support Tool in Processing
Mark C Mitchell: mmit9210@uni.sydney.edu.au
!
Oliver Bown: oliver.bown@sydney.edu.au

2.
Understanding the Needs
of Creative Coders
A formative study in a line of studies whereby we
hope to gain a more detailed understanding of
how creative coders interact with their code.

3.
What is Creative Coding?
Philosophy of open code-sharing.
A highly interconnected, “new
kind of creativity”.
A discovery-based process using
code as primary medium.
Understanding gained through a
context of experience.
A developer with loads of ideas on
how to make unconventional
combinations of technology work
together.
One who is in the process of
making a specific type of audio/
visual artefact.
“If you are familiar with the open source
language Processing then you'll have a good
feel for what creative coding is about.”
— David Bolton,
http://cplus.about.com/b/2010/12/09/what-is-creative-coding.htm

4.
Creative Coding
The Creative Coder
Bricolage
Programming
Reflection-in-
action
Creative coding is a process of exploration,
incremental building and reflection.

5.
Bricolage Programming
“Bricolage programmers construct theories by
arranging and rearranging; by negotiating and
renegotiating with code” (Turkle & Papert, 1990).
“Bricolage programming is particularly applicable
to our theme of the artist-programmer.”
(McLean, 2011).
Turkle, S., & Papert, S. (1990). Epistemological Pluralism: Styles and Voices within the Computer Culture. Signs, 16(1), 128-157
McLean, A. (2011). Artist-Programmers and Programming Languages for the Arts (Doctoral dissertation).

6.
Bricolage Programming
Bricolage programming is particularly applicable to
our theme of the artist-programmer (McLean, 2011).

7.
The bug I encountered was so bizarre, I decided to do a render.
The behaviour is totally a surprise to me. All I was hoping for
was simple GPU based sphere-sphere collision as the result of
gravitational attraction.
— Robert Hodgin, Repulsion
(Eyeo 2012)

8.
What is Creative Coding?
We define creative coding as a discovery-based
process consisting of exploration, iteration, and
reflection, using code as a primary medium, towards
a media artefact designed for an artistic context.

9.
Jonathan McCabe
Frederik Vanhoutte
Jason Rampe
Kyle McDonald
Felix Woitzel
esimov (as3)
Jason Rampe

10.
Aims
To understand and explore how the engagement
between creative coder and coding environment
can be bolstered.
Now that we have defined creative coding, we can ask:
Is the process of a creative coder being adequately
supported by current coding environments?

11.
Environments for Creative Coding
Processing
OpenFrameworks
Cinder
NodeBox (soon: MAAK…)
Field
LightTable
C4
more!

12.
An observational study
Analyse the practice of creative coders and better
understand their needs, which can inform the
requirements of a creativity support tool.
A qualitative study was conducted with self-described
creative coders, using the Processing
environment.

13.
Methodology
The study was conducted using direct observation
and semi-structured interviews.
Nine participants were recruited.
A post-task interview was employed instead of
using the think-aloud method, so the flow of the
participant was not interrupted.

14.
The Task
The participants were
encouraged to utilise any
resources for the task they
wished, including code libraries,
code from the Internet, the
example creatures shown at the
beginning of the task, or previous
work.
Observations and interview
questions were tailored to extract
the why of the practice, rather
than the what.

15.
A Taxonomy of Needs
Build code on a modular foundation
Initiate the creative feedback loop quickly
Make the program state visible
Minimise the creative feedback gap
Assistance in exploring the space of possible designs
Maintain a robust mental representation of the software

16.
A Taxonomy of Needs
Build code on a modular foundation
Starting-methodology: 1st stage: “scaffolding”
“The scaffolding conducted by 8 of the 9 participants facilitated a
modular approach to design.
“If didn’t scaffold and wanted to implement even more things, it
would be even harder. So I decided to make it modular.”
— Participant 2
“If I had more time, I would have separated the creature behaviours
into modules” — Participant 3

17.
A Taxonomy of Needs
Initiate the creative feedback loop quickly
Starting methodology: 2nd-stage
Achieving a minimal representation of the creative goal
to the screen: in the case of this task, it was, in almost
all cases, an ellipse drawn to the screen which acted as
a placeholder for the creature. This initial visual artefact
acts as a springboard for ideations, initiating the
creative feedback loop.

18.
A Taxonomy of Needs
Make the program state visible
Opaqueness of the system forced participants to make
assumptions about the cause of unintended behaviour,
selecting variables to print to console without any support
from the IDE.
Making the program state visible in a contextually appropriate
way could alleviate these cumbersome debugging events

19.
NodeBox 3

20.
A Taxonomy of Needs
Minimise the creative feedback gap
The gap that exists between tweaking and observing creates a
temporal delay that slows the creative process (McLean and Wiggins,
2011). Beyond slowing the process, the gap can also cripple it.
The act of perception is integral to the creative process, and a lag
between modification and the perception, judgment and analysis of
the outputted result can stifle the creative feedback loop as causal
mappings between modification and result are disrupted.
Minimising the gap in this creative feedback loop will alleviate these
problems, as well as improve flow.
McLean, A. and G. Wiggins, Bricolage programming in the creative arts. http://yaxu.org/writing/ppig.pdf

21.
A Taxonomy of Needs
Minimise the creative feedback gap
"creators need an
immediate connection
with what they're
creating … when
you're making
something, when you
make a change, or a
decision, you need to
see the effect of that
immediately. There
can't be a delay; there
can't be anything
hidden”
Victor, B. Inventing on principle. in Invited Talk at Canadian University Software Engineering Conference (CUSEC). 2012

22.
A Taxonomy of Needs
Minimise the creative feedback gap

23.
A Taxonomy of Needs
Allowing the feedback loop to begin uninterrupted at the beginning of a
creative task and optimising the tweak/run/observe cycle would increase the
role of creative ideation and flow.

24.
A Taxonomy of Needs
Assistance in exploring the space of possible designs
All participants engaged in rounds of parametric design
tweaking — modifying a parameter of interest, then
compiling and running the program to visually interpret
the results.
Tweaking of parameters can be considered an exploratory
type of creativity, which involves exploring, navigating and
testing the potential and boundaries of a conceptual
space.

25.
A Taxonomy of Needs
Assistance in exploring the space of possible designs
Boden defines three types of creativity:
Combinatorial creativity
Exploratory creativity
Transformational creativity
Creativity is a matter of mapping and exploring
structured conceptual spaces (Boden, 1994).
!
A map is used to search the space (exploration).
Exploratory creativity can offer surprises
comparable to the surprises provided by
transformational creativity (Boden, 1995)
M. A. Boden. The Creative Mind: Myths and Mechanisms. Weidenfield and Nicholson, London, 1990.
M. A. Boden. Modelling creativity: Reply to reviewers. Artificial Intelligence Journal, 79:161–182, 1995.

26.
A visual representation of search in a conceptual space (Veale, 2012)
Veale, T. (2012). Exploding the Creativity Myth: The computational foundations of linguistic creativity. A&C Black.

27.
A grid of possible design avenues to select from, evolving the parameters to an optimum state.
courtesy of Adam Younis (Design Lab)

28.
NodeBox 3

29.
A Taxonomy of Needs
Maintain a robust mental representation of the software
Contrary to our expectations, ready-made code was utilised by only one participant. This is
an unexpected finding, as we hypothesised that ready-made code would be a common
starting point for the task across all participants, regardless of expertise level.
Perhaps even more surprising was the lack of any third-party library use by any of the
participants.
8 out of 9 participants expressed hesitation with utilising third-party libraries.
The time it takes to learn an external library for use in a project requires not only an
understanding of the system the library represents, but also the coding style of the library
author.
Allowing a creative coder to keep a robust, mental representation of their software at all
times affords the integration of external code that lends extensibility to their program,
removes the lack of expertise barrier for specialised domains, and affords creative coders
to more easily implement their creative ideas.

30.
Future Work
Turning these needs into requirements, a prototype of a
creativity support tool can then be evaluated.

31.
Thank you!
Mark C Mitchell: mmit9210@uni.sydney.edu.au
!
Oliver Bown: oliver.bown@sydney.edu.au

32.
Creativity is often seen as some combination of
quality and novelty/innovation.
• Open Questions
• What questions should the judges be asked?
• Are we interested in a holistic judgement of “creative”?
• Is there a risk of the word “creative” having different meanings for judges?
• Do we have to be more careful about who the judges are?
• Summary:
• There are still open questions about measurement of “novelty” or “creativity”.
— Graeme Richie, Evaluating Quality in Creative Systems