flash supercomputing

the name says it all.

Link via Smartmobs

random connection network

example 1 - visualization

interactively Storytelling - more updates

maybe I should wiki-ify this, so I can stop having to write/export to pdf/repost

additions: lots of stuff.

Download .pdf

It's a little long, I know. But if you get the chance to read it, can someone post comments? This is probably too much to ask, I realize...How about this. If someone even reads this post, let alone the paper, can you post a comment. I think my comment system is err...broken.

eMurge (590 pdf)

download .pdf

interactively Storytelling - update

revision of the paper I posted a few days ago. this version talks more about data-mining, and it's place in "what I like to call" the "story-blogsphere." How can we create community and collaborative spaces for telling stories over a network? The paper aims to be general enough to provide an outline for how we can create these spaces, and what problems we may encounter. It is also narrow enough to propose a specific environment for these narrative-building spaces that I'm planning to construct. Have a read.

Interactivelly Storytelling: Download pdf

Interactively Storytelling

A rough and 1st draft of a short paper outlining a possible web publishing system for online community narrative and world building.

link to pdf: Interactively Storytelling

'who needs musicians when computers can think like bees?'

so asks a Discover Magazine article by the usual suspect, mr. Stephen Johnson. the article lives here. as johnson notes, making computers autonomously compose music has always been a big thing in computer music. a quick check at the 2003 international computer music conference's papers and demos yields a hefty number of these keywords: detection, pattern extraction, algorithmic approaches to composing, etc. etc. the real big thing now is computer improvisation. the guy from the article, Tim Blackwell (sound clips on this page) is having bots improvise with other bots. others, including my thesis advisor at brown (_Todd Winkler). this is a hard problem, and has been researched heavily. I think that Blackwell's got the right idea here, though. Music and sound creation via a whole-is-greater-than-the-sum-of-its-parts system is natural I think. music is so essentially modular to begin with. this is an area where I'm going. the key, and something I think blackwell is ignoring, is adequately placing the composer in this emergent space. they need to be something more than someone who initializes variables.

random connectors

demonstration of granovetter's theory on the strength of weak ties. nodes are added, and similar nodes--nodes within a small radius of one another when each is created--are linked. however, each node also has a 1/12 probability of becoming a connector node. connector nodes not only are part of the smaller local networks, but also branch out to other nodes outside of the given neighborhood radius. comparing the diameter of the network with the presence of these connectors versus the simplier local networks yields drastically lower numbers for the connector network. in gladwell's terms, these are your paul revere's or your baltimore pimps. they dramatically lower the average degrees of separation for the network, and therefore facilitate any type of epidemics. to do: assign random 'infectious-ness' vars to nodes. allow user to infect network with pieces of sound or text, and see how the information spreads through the network.

simplistic smallworld networks

diffusion-limited aggregation

aristocratic network using simple rules; start with one node (0,0). Another node wanders in, and if it is close to (10 pixels) or hitting another node, then it sticks in place. Due to time and processing restraints, the process here is only repeated 300 times and on a small (200X200) stage. The results (more so if you do this over a few hundred thousand times) yield amazingly complex forms, as the nodes self-organize. as the process continues, nodes begin to form into branched arms which act as filters. as a result, new nodes tend to stick to these outside branches, rather than wander deeper into the cluster.

check out an animation of the process.

military swarm robots

From New Scientist:

A battalion of 120 military robots is to be fitted with swarm intelligence software to enable them to mimic the organised behaviour of insects. The project, which received funding this week from the US Defense Advanced Research Projects Agency (DARPA), is aimed at developing ways to perform missions such as minesweeping and search and rescue with minimum intervention from human operators...

Read More

The question is how these swarming robots would be used in a military situation; or much more interesting, how swarming robots could be useful in a more social or cultural capacity.

dynamics for designers

GDC 2003 Video: Will Wright's 'Dynamics for Designers'
"Somewhere between high-level game concepts and low-level coding lies a region of design that's really at the core of the interactive medium. It's here that causal relationships, feedback cycles, information propagation and emergence mechanisms reign supreme. This is what Wright calls "dynamics"; the rules and principles that govern the way in which structures change through time. The design and use of early prototypes is covered as a means to explore and sculpt a variety of dynamic systems."

http://www.gamasutra.com/features/20030403/wright_01.shtml

free registration required. Video is about an hour.

self-organization_1

A Flash Experiment that demonstrates emergent, self-organizational behavior. Throw some balls into a space, apply a few rules, and see what happens - the result it pretty cool. I'm looking forward to expand this experiments into something more - see if I can't tweak some instructions a little more to see if they'll organize in a different manner. A lot of this relates to stuff we were talking about in 564 earlier today - I'm very interested by the role of authorship in these systems - or more precisely, how linear forms (I'm calling linear anything that is centrally controlled) can be merged with non-linear (I'm calling non-linear a system that is made up of collective actions of individual actors) in an engaging way that makes use of the strengths of both.

more links

Some more interesting emergence links on the web:

Emergence author Stephen Johnson has a pretty interesting blog talking about a lot of these issues.

This is a pretty cool "Prototype for an Emergent Behavior Visualization Tool, based on the principles presented in the reading 'Design of Cities', by Edmund N. Bacon."

Also I would really recommend everyone, if you don't bother with the above, check out levitated.net which is run by Jared Tarbell, who is a researcher who uses actionscript to create both highly appealing and emergent peices. Very compelling stuff.

complexity | narrative

I've been trying to write a proposal for a directed research project next semester - ctin 590. The major thing that is interesting me right now is the idea of emergence, and especially the applicabilty of these ideas to a really diverse range of interactivity. So in 564, we always talk about interactivity in a pretty circumscribed way - that is, we always talk about it with reference to the idea of the database narrative. Fundamental to this strand of thinking is navigation; the user makes their way around the world, exploring things. Based on what they are investigating / fiddling with, they are presented with some piece of information, be it textual (maybe some bit of a poem), auditory (a voiceover or a particular soundscape) or visual (an animation), that has been randomly, or in some prescribed way, pulled from a database. Then, as we've been talking about, the user assigns cause and effect relationships between these database elements, and therefore creates their own little narrative.

While I think this method can be rather successful for specific works (I haven't loved all of the Labyrinth Project stuff, but Tracing the Decay of Fiction is a pretty good implementation of a successful db narrative, I think), I've really been thinking that the database narrative is a bad way to go about achieving complexity in interactive projects.

After Kurt and I saw Will Wright at GDC, the more I've been thinking that emergent systems are the best way to feasibly achieve the level of complexity that many of us are working towards in our research. These systems give us information that evolves rather than information that some poor writer has plugged into a vast database. Since these systems are typically based on the idea of simple nodes interacting with each other and deciding how to behave based on a simple set of rules that have been applied to them (e.g. a sim), narrative elements could easy be integrated into an emergent system that gets closer to achieving the kind of complex behavior that simply cannot be bourne from database systems.

Anyway, I think that these ideas are something that we all should be thinking about when we are struggling with our research.

Smart Mobs deals with these issues on a societal / anthropological level, and is a pretty good place to start. The book certainly has been a good place for me to further my research, as it connects emergent structures with issues of networked collaboration. yes!