Synynyms ties together the APIs of three important biodiversity informatics projects. The names and taxonomic synonyms themselves come from the Encyclopedia of Life, those names are parsed into their components by the Global Names Index, and most importantly of all, the publication occurrences come from the Biodiversity Heritage Library.

Because the datasets are large and take a long time to download, I wrote this tool in Node.js and Socket.io to facilitate realtime graphing in the client. As vast amounts of data stream in from the web services, the Node server parses and analyzes the data, then sends it to the client over a WebSocket, thus allowing the user to watch the graphs build themselves in realtime. The graphs act as their own progress bars. Up front, I used Backbone to keep the front end sane, and Raphael to draw the graphs themselves.

Breedlight is a series of sculptural lamps that I’ve been working on, whose design is dictated by a genome that I created. Breedlight lamps are capable of reproducing with each other, passing on their traits to future generations of interior lighting.

Each lamp has a unique genetic code—a DNA of sorts—that describes its size and shape. Although I make the lamps, I do not design them. I breed them from other Breedlight lamps!

Each lamp’s traits are contained in a single chromosome, shown above, comprised of 20 genes. These 20 genes encode the size and shape of the Breedlight lamp. Its height, width, and the prevalence and shape of its curves are all described by these genes. Breedlight lamps with two different shapes will have correspondingly different chromosomes to manifest those physical differences.

I chose the chromosome model to enable two lamps to reproduce with each other, passing on their their genetic material and physical traits to their offspring. This works very much like in real life: to reproduce, each parent lamp splits its chromosome, independently assorts the genes, and contributes half its genetic material to the offspring. The offspring receives a half-chromosome from each parent, combining them to form a full chromosome. Laws that I’ve encoded in the genome determine how the traits inherited from each parent combine and contribute to the offspring’s physical form.

Consider the example breedings above, showing the children of lamp A 💖 B and the children of lamp B 💖 C. You can see the children of both pairings have a tendency to bulge in the middle, a trait that they inherit from lamp B, which is common to both. However, the children of A 💖 B have a tendency to be short, squat, and round—traits they get from lamp A. On the other hand, the children of B 💖 C have a tendency to be tall and slender, just like lamp C. Also notice the wide variety amongst the children; although the children inherit all their traits from both parents, they don’t necessarily inherit the same ones in the same proportions.

I showed the first real Breedlights at TEDxWoodsHole. I made a series of three lamps—two parents and one offspring. All three lamps are capable of breeding with each other; yeah, kind of Oedipal, but hey they’re lamps. Breeding is made possible by QR Codes attached to each lamp that contain their genetic material, analogous to DNA.

The QR Code contains the all the genetic material that describes the lamp. When the QR Code is scanned, software that I wrote can decode the chromosome and genes from the information contained in the code. To breed two lamps, scan both of their QR Codes. The software can decode the chromosomes of both lamps, breed them repeatedly, and render out drawings of each child.

From there, a customer could select an offspring that they’d like to purchase, and off into the software its genome goes. Given a genetic code, the Breedlight software can produce full-scale vector drawings of the lamp’s forms, which I currently use to make jigs upon which I construct the lamps. For the series that I showed at TEDx, I made the lamps in the traditional style of a Japanese chochin, with mulberry paper on a raffia frame.

This construction method, while elegant, historic, and well suited to prototyping the Breedlight idea, does not fully do justice to the varied organic forms of Breedlights; not to mention its unscalability for meaningful production. Based on the overwhelmingly positive response I received at TEDx, I continue to experiment with more efficient materials and construction techniques. Because my software can render a Breedlight chromosome into a vector drawing, it is possible to output directly to a CNC machine, laser cutter, or rapid prototyping machine. Now that many of the genetics and technical hurdles are accomplished, it’s a matter of finding materials that are harmonious with scalable one-off production, and capable fully expressing the Breedlight idea. Thus Breedlight remains a constant work in progress.

The following graphics show the usage of all the scientific names of a particular organism through time. You can see how names grow and change.

The taxonomic synonym data is provided by the Encyclopedia of Life. For each organism, every taxonomic synonym is searched in the Biodiversity Heritage Library to retrieve a list of all the publications that mention that name. The publication dates of those works are used to build histograms of the number of publications-per-year, which are then used to generate the graphics.

The purple silk represents a timeline from 1750 to 2000. It displays a histogram of the number of new species discovered per year in any particular grouping of organisms (called “taxa”). In the image above, we can see the number of new species of octopus discovered every year. There are two major spikes in the early Twentieth Century, which correspond to two major published volumes on mollusks. In the video below, we can see the number of new turtles, birds, and rodents discovered per year since Linnaeus first described species in 1750. The silk moves beautifully as it transitions from one species to another.

Taxatron Prototype from Ryan Schenk on Vimeo.

The chiffon creates an entirely new setting for the scholarly data that Taxatron displays. Taxatron recontextualizes “hard science” data into something soft, translucent, and easily mutable. Taken out of the safety of a hard glass computer screen, or the comfort of a poster session, the data is vulnerable, suspended for viewers to inspect from any angle, and even to be buffeted by a strong wind.

Lately, there has been much ado about “transparency” in regards to data. Taxatron takes this charge literally. Light easily passes through the cloth, allowing viewers to visually interact with each other, and allowing the projected metadata to radiate through the cloth, wholly becoming part of the medium. The material selection also plays a pivotal role in the movement of the graphs, effortlessly transforming from one shape to another, and raising important questions about the permanence or fleetingness of scientific discovery and the organisms that inhabit our planet.

Taxatron was displayed at the TDWG 2010 conference in Woods Hole, and again at TEDxWoodsHole. Receptions at both events allowed scientists and nonscientist alike to view, discuss, and even touch scholarly data. I think my favorite part of the event was hearing the librarians and archivists from the MBLWHOI Library correlate spikes in the histograms with the influential monographs that described those species.

From a technical standpoint, the chiffon is suspended from an array of 15 stepper motors, which are mounted to a housing hung from the ceiling. With the help of an electrical engineer friend, I created custom circuit boards that contain the driver circuit, stepper motor mount, and a photo switch used to calibrate the system. The chiffon is suspended from a strand of thread, attached to a small pulley mounted on the stepper motor.

Inside the housing, I fabricated mounting brackets from sheet metal to hold the stepper motor boards in place. All the steppers are wired to an Arduino Mega, which runs the show. I wrote firmware in the Arduino to drive the motors given commands from the serial port. Every time the sculpture powers on, it calibrates itself with the photo switches on each motor board; this allows me to use any length of string to suspend the cloth; the sculpture will scale the measurements accordingly.

The stepper motors can move the silk into a wide variety of shapes by spooling the thread around a pulley, thus shortening or lengthening the thread suspending each segment of cloth.

Taxatron hung at TDWG 2010 and TEDxWoodsHole 2010; unfortunately, both events were extremely short, and I was not able to push the limits of what Taxatron can do because of its temporary location. As you can tell, I barely had time to snap a few photos. I am looking for a home for Taxatron that would allow it to hang for an extended period of time, and allow me to fully develop the projection software to transform the cloth into a unique and truly memorable experience.

I created a batch of Settlers of Catan tiles for you! You can download print-resolution versions at my Flickr page.

If you would like to edit or remix the tiles, you can also download the original, fully editable Illustrator artwork. If you make any improvements, please send them my way!

The first piece is a cascade of laser cut felt feathers, called Feather Vane.

The second piece is a slightly more conceptual mobile. It’s a large wood and paper sail that holds a piece of charcoal on the end.

It is very delicately balanced so the charcoal just barely skitters across a paper-covered table. As the wind blows the sail around, it traces light marks on the paper.

As the exhibition progresses, the charcoal drawing will build up layer-by-layer. This is the progress after one week. Looking great!

You can see more photos in the Flickr set.

There are several driver circuits available for this motor, which were very helpful in figuring out the strange wiring inside this motor. However, none of the above drivers played nicely with the Arduino Stepper Library, which uses Tom Igoe’s stepper driver circuit. It took an evening to figure out how to connect this stepper to Tom’s driver and the Arduino stepper library, so I thought I would post it here.

The first step is to cut the white wire, which turns the stepper from a weird hybrid into a normal bipolar stepper. The links above will explain this in more detail if you’re curious.

The next step is to correlate the four wires on the motor to Tom Igoe’s circuit and the Arduino stepper library. These use the numbers 1-4 to describe the wires coming out of the motor, and order does matter. With the 21-02485-03, the wires numbered 1-4 correlate to the Red, Blue, Yellow, and Black wires respectively.

With this knowledge we can easily hook up our stepper to Tom’s circuit and start using the Arduino Stepper Library immediately. Here’s a diagram of Tom Igoe’s 4-wire circuit, with the Arduino code after it. The IC chip used is a L293D H-Bridge and note that the notch in the IC faces to the right (a bit hard to tell).

Image developed using Fritzing.

/* 
 
  Drives the 21-02485-03 Stepper Motor attached
  to Tom Igoe's four-wire driver circuit
 
*/
 
#include <Stepper.h>
 
 
// Change these to the pin numbers of each color wire 
#define YELLOW  8
#define RED     9
#define BLACK   10
#define BLUE    11
 
#define SPEED 200 // RPM
 
 
#define LED_PIN 13
 
// create an instance of the stepper class, specifying
// the number of steps of the motor and the pins it's
// attached to
//
// Order matters here when giving it the pins
Stepper stepper(20, RED, BLUE, YELLOW, BLACK);
 
 
void setup()
{
  // set the speed of the motor to 30 RPMs
  stepper.setSpeed(SPEED);
 
  pinMode(LED_PIN, OUTPUT);
}
 
void loop()
{
 
  stepper.step(100);
  delay(300);
 
  stepper.step(-100);
  delay(300);
}

It is a Ruby on Rails application that I wrote when I was at the Marine Biological Laboratory. I designed and implemented the entire application stack from the ground up. The data is split between a MySQL database and a Redis cluster; the Redis cluster, which stores hundreds of millions values, was the largest in the world at the time it was built. To compute queries on demand, it has a large, horizontally scalable processing cluster, which pull tasks from an AMQP work queue and process them in parallel. Using this architecture, I processed each of the 1.9 million species in the Encyclopedia of Life though LigerCat, analyzing the metadata of tens of millions of scholarly articles, in a matter of days.

The basic idea behind LigerCat is that tagging is nothing new. The Web 2.0 folks got the idea from librarians, who have been tagging literature for many years. Librarians tag things using a “controlled vocabulary,” which is a set of tags that are curated and maintained by some authoratative body. For instance, scientific articles indexed by the National Library of Medicine are tagged with a controlled vocabulary called Medical Subject Headings (MeSH), which has over 20,000 tags in the set.

The Articles search, which is selected by default, allows the user to query the PubMed article database. Ligercat will download all the results, and build a MeSH tag cloud from the articles returned by your search. You can search for a topic, a person, or an organism, and LigerCat will build you a MeSH cloud based on the results.

In addition to designing the databases and writing the application code, I also designed and coded the UI. LigerCat has extensive client-side interaction that I developed with MooTools. The page below allows users to visually construct queries to the PubMed repository by clicking on interface elements. As the user clicks, LigerCat queries PubMed in realtime, providing instant on-screen feedback. The publication timeline at the bottom is fully interactive and is implemented entirely in CSS, no JavaScript (admittedly pedantic). I have been contacted by Dryad, Agricola, and Elsevier, who have wanted to use LigerCat’s interface to display their metadata. Because I designed LigerCat’s processing system based on the Strategy and Command design patterns, placing LigerCat on top of another literature corpus is as easy as writing two Strategy classes.

LigerCat can be cited as,

LigerCat: using “MeSH Clouds” from journal, article, or gene citations to facilitate the identification of relevant biomedical literature. Sarkar IN, Schenk R, Miller H, Norton CN. AMIA Annu Symp Proc. 2009 Nov 14;2009:563-7.

When you laser cut felt, it smells very strongly like burning hair (go figure), and it leaves a charred halo around the edges. Some people recommend dry cleaning to remove the smell and char, but that was too expensive and environmentally unfriendly for me. I experimented with a couple ways of cleaning the felt, and this is what worked best for me.

1. Make a mixture of cold water and dish soap
2. Soak the felt for 15-30 minutes
3. While still submerged, go around the edges with an electric toothbrush
4. Rinse with cold water
5. Pat dry on a tea towel, then leave on the counter to air dry

This method is pretty quick, easy, and works. Just remember to use an old toothbrush head.

Photos after the jump

Here’s some before and after photos!

As you can see below, the felt will shrink, but it’s not too bad.