Artificial life software

Citation

Adamatzky, A. (2003), "Artificial life software", Kybernetes, Vol. 32 No. 1/2. https://doi.org/10.1108/k.2003.06732aaa.001

Artificial life software

Artificial life software

This special double issue of Kybernetes is about software tools, environments and realities dealing with creation, imitation and analysis of artefactual and living forms. There are thousands of “talkers” but just handful of “doers” in the field of Artificial Life, the present volume is unique because it is written by “doers” – designers and makers of software, hardware and art installations in artificial life, simulation of complex systems and virtual reality. The papers of the volume are self-contained, each one includes an excellent prelude to the field and is itself a treatise of advanced results and discussions. All papers must be read in parallel, however, our conventional journal format has imposed the following serial arrangement: from simulations to art forms.

To teach decentralised thinking about a decentralised world and to give non-experts a tool for interactive exploration and a simulation understanding of complex systems, these, probably, were the primary objectives in developing StarLogo – a language of concurrent programming and simulation (Colella et al., 2001; Resnick, 1990, 1991, 1994). The aims are successfully achieved: kids and adults, zoologists and mathematicians, physicists and computer scientists, driven by curiosity, are building in StarLogo their models of real-life phenomena. In their paper “StarLogo under the hood and in the classroom” Eric Klopfer and Andrew Begel give an accessible introduction to StarLogo, they guide readers from basics of modelling, to architecture of StarLogo platform, interface and schemes of modelling; they also share with readers their experience on how to transform novice students to masters of modelling using StarLogo. Authors of the paper are renowned scientists, programmers and teachers. Eric Klopfer developed many biology and physiology animations, Java simulators of ionic exchange in cells and evolution of co-operation (Taumoda.com, 2002). Andrew Begel, besides StarLogo, also worked on YoYo, Java for kids (YoYo, 1996-2001), and LogoBlocks, graphical environment for programming tiny robots (Cricket, 2002).

“On center stage…was a quiet and intense young man who was writing equations that travelled the length of the stage and back again. Lt. Gen. van Riper understood what Dr. Andy Ilachinski was writing, although most in the audience did not.” (Johnson II, 2001). This happened in 1993, when Andrew Ilachinski has been (I am just guessing) presenting seeds of his complexity-based theory of non-linear war-gaming. In his present paper – “Exploring self-organized emergence in an agent-based synthetic warfare lab” – Andrew Ilachinski offers a comprehensive introduction to a theory of agent-based simulation of fundamental space-time phenomena of war and describes in details his EINSTEIN software package, designed to use autonomous minimalist agents in modelling individual behaviours and personalities of combatants. Andrew’s approaches to warfare simulations are rooted in his background in mathematical physics and his cellular automata works. In 1987, he invented structurally dynamic cellular automata (Ilachinski and Halpern, 1987) and recently published an immense handbook on cellular automaton theory and applications (Ilachinski, 2001).

In 1992 Andy Wuensche got world-wide respect for his treatise of global dynamics of cellular automata (Wuensche and Lesser, 1992), this book alone persuaded hundreds of scientists to study the fundamental of space-time behaviour of large-scale locally interacting systems. Wuensche’s Discrete Dynamics Lab (DDLab) is an interactive tool for designing and investigating the dynamics of discrete dynamical networks, including networks with random topology. This is a tool for the building of models and producing valuable results in complexity, emergent behaviours, neural networks, cognitive sciences and many aspects of theoretical biology. Recent examples of applying DDLab to research problems include approximation of Garden-of-Eden states of cellular automata (Adamatzky, 1999) and simulation of gene expression networks (Somogyi et al., 1997). DDLab quickly evolves almost every month and new and significantly enriched version of the program appear several times per year. In the present collection, Andy Wuensche – “Discrete dynamics lab: tools for investigating cellular automata and discrete dynamical networks” – shows how to simulate decision networks, build attraction basins of discrete systems, automatically search for mobile patterns, weakly coupled networks, static parameters, reconstruction of past dynamic of discrete systems. Stuart Kauffman’s words about the second edition of Andy Wuensche and Mike Lesser’s book (Wuensche and Lesser, 1992-2000) can be fully applied to DDLab software: “…their success, provides analytic insights that trained mathematicians working in the field missed left and right” (Kauffman, 2000).

Mirek Wójtowicz stormed the world of cellular automata with his powerful, precise, friendly and comfortable cellular automaton explorer – MCell, a free open-source application. This tool for studying cellular automata runs almost all known nontrivial cellular automaton rules and has a regularly updated database of old and newly discovered patterns. “Mirek’s Cellebration is a fantastic 32-bit CA program for Windows, far and away the best general-purpose CA engine on any platform. Get it!”, writes David Griffeath (Griffeath, 2001). Rudy Rucker gets even more emotional – “You’re a genius”, he exclaims (Rucker, 2001). Present volume includes first ever paper, written by Mirek Wójtowicz for an academic journal – “Exploring cellular automata with MCell”.

Jon McCormack and Alan Dorin, academicians and artists, and masters of graphics and sound, lead the Center for Electronic Media Art (Monash University, Australia), which research activities “…are driven not only by technical and scientific challenges, but also by creative and artistic aims” (CEMA, 2002). The centre concentrates on algorithmic approaches to media generation, sound synthesis and computer-assisted music composition, virtual reality, and philosophical analysis of generating techniques in electronic art (CEMA, 2002).

Alan Dorin authored well-known Bauhaus (simulator of three-dimensional evolving creatures made of cones, spheres and cylinders), Breese (simulator of evolving artefacts made of suspended particles) and Collide (imitator of a physical environment where virtual creatures interact) and highly original Liquiprism, a cellular automaton which generates sensible sounds(Dorin, 2002). Alan Dorin’s “On wonder and betrayal: creating artificial life” highlights somewhat a phenomenological understanding of what makes digital life appealing to scientists and artists, and why do people become attracted to certain forms of creative computer art.

Virtual and real worlds interact one with another in Eden, a “reactive”, as John McCormack calls it, artwork, which combines cellular automaton models of a discrete world interacting with outside physical reality via graphics, sounds and infrared sensors. The cellular-automaton world interacts with those who observe it and even evolves depending on its observers behaviour. This highly original and extremely attractive installation is discussed in John McCormack’s paper “Evolving sonic ecosystems”. Other McCormack’s art projects include Universal Zoologies (McCormack, 1996-2001), Turbulence, The Beauty to Be and many more (McCormack, 2002).

“For me this is simply cool! Did not expect to see so interesting work…”, – commented one of the referees about “Aesthetic selection of morphogenetic art forms” by Dale Thomas. In this volume Dale Thomas gives an insight into his Aesthetic Selection program, which employs genetic algorithms, reaction-diffusion, morphogenesis and selection to artificially grown artistic “living” forms. The project has been successfully tested at Science et Cité fair (Science et Citó, 2001), where dozens of participants evolved their artistic forms “on flight”.

Framsticks made one of their first public appearances in the European Conference on Artificial Life held in Lausanne in, 1999, where participants have been stunned by movies of these artificial creatures, which are made of sticks, muscles, neurons and receptors (Komosiński and Ulatowski, 1999). A demonstration room was rocked by participants’ ecstatic screams when a jumping spider attacked a handicapped walker. On that occasion Chris Langton personally awarded Maciej Komosiński for his excellent presentation at the conference. In last years, Framsticks are developed to a unique representation of artificial worlds, where you could evolve your own creatures, design new would-be animals from scratch, and virtually test prototypes of bio-inspired robots. There is already a huge network of Framsticks users, developers and collaborators, the software is used to evolve and design physical robots with unconventional architecture (Kennedy et al., 2001; Pollack et al., 2001). In his present paper – “The Framsticks system: versatile simulator of 3D agents and their evolution” – Maciej Komosiń ski offers first-hand introduction to his software, describes Framsticks anatomy, physiology and genetics, and gives accounts of his experimental research.

Bruce Damer, Karen Marcelo and Frank Revi present us with “Nerve garden: germinating biological metaphors in net-based virtual worlds” – excursion inside a biological three-dimensional virtual world explored by thousands of users on the Internet. Nerve Garden aims to develop a networked collaborative laboratory for artificial life, where users create, explore and study virtual ecosystems. Bruce Damer is a guru of virtual worlds (Damer, 1996-2002) and author of Avatars! (Damer, 1997), the book which became best-seller at once after its publication. In 1980 Bruce Damer developed a clone of Elixir products, including renowned Elixir Desktop, ElixiFont, and ElixiGraphics (DigiBarn Computer Museum, 1999-2002). He founded several companies, including the Contact Consortium, “…the first global organization focused on inhabited virtual spaces… shared in real time by thousands of users and represent a new frontier in the experience of cyberspace.” (Contact Consortium, 1995-2002).

“Simulated breeding: a framework of breeding artifacts on the computer”, by Tatsuo Unemi, is about how to breed images and music. A genotype of a musical species consists of three two-dimensional arrays which represent melody, rhythm and velocity; genotypes of graphical species are seen as mathematical expressions, which determine a colour of each pixel of a two-dimensional image. The paper discusses SBEAT (graphics) and SBART (music) computer programs that are designed on a generate-and-test framework, implement interactive genetic algorithms, island model of breeding, multi-field breeding and protection of individuals, partial breeding and direct genome operations. To develop these unique breeding tools Tatsuo Unemi applied his experience in evolutionary robotics, collective behaviour of reinforcement learning agents, simulation of plant development and evolutionary differentiation of learning abilities.

The authors of the issue are outstanding scientists, artists and academicians in artificial life, complex software development, virtual reality, multi-agent systems, and computer creativity. They are internationally recognised authorities. I am grateful beyond words to everyone who kindly agreed to participate in this issue.

I heartily thank reviewers, who spent so many days browsing through the submissions. Their emotional responses, which were mainly complimentary, showed that this issue is destined to touch its audience and the papers of the issue will excite and inspire everybody – from undergraduates to established academicians. Please let us therefore acknowledge with gratitude the contributions of: Larry Bull (School of Computing, University of the West of England, Bristol, United Kingdom), Christopher Melhuish (Intelligent Autonomous Systems Lab, University of the West of England, Bristol, United Kingdom), Pawel Siwak (Poznań University of Technology, Poznań, Poland), Alexander Tarakanov (Institute of Informatics and Automation, St. Petersburg, Russia) and Hiroshi Yokoi (Complex System Engineering, Hokkaido University, Sapporo, Japan).

Andrew AdamatzkyFaculty of Computing, Engineering and Mathematical Sciences, University of the West of England, BristolE-mail: andrew.adamatzky@uwe.ac.uk