Final Proposal  So for my data rep final I’m not totally set on this idea for a couple reasons, but the basic idea is a visualization of space and environmental inputs over time.

I’m currently interning at Base Design designing their permanent interactive installation for the NY headquarters. They want it projected in the five windows facing the street off Lafayette and Grand in Soho. It’s the top floor corner windows.

Coincidentally they have 5 international offices and what inspired me was that at the top of their site they have the current times and whether or not each office is “open” or “closed.”

I love that they show some current contextual information about their other offices and it feels like it gives a little bit of a story with just a tiny bit of information. So what I’m thinking about taking video feeds of the different offices and transforming them into a slit scan in the shape of a circle and have it animate based on some environmental input such as sound. Then each day at midnight it will animate through each segment and show the feed from each hour.

1) What is your data set?- The data set will be video feed from different offices and surrounding sound levels.
2) What is the medium?- The medium is an projected installation on windows.
3) What is the central question?- What do these different spaces look like over time?

 

Assignment We were asked to choose a data set from the Guardian and create two different visualizations from it. The first was to make the information as clear as possible and the second to represent it in a way that addresses the specific character set. I chose a set US debt: how big is it and who owns it? Amazingly, half is owned by China and Japan alone. I think it is more apparent in the first visualization than the second one.

 

 Why connect computers to the Rest of You? In short, it can be used to identify, give context to and serve as a jumping off point for possible solutions to problems. It can provide insight to issues not usually given much attention or it can be used as leverage and create new ways to explore old problems. The rest of the post I explain my thoughts through “thesis glasses” as I am using ROY to explore ideas pertaining to my thesis.

The Problem: We suck at communicating (and it’s only getting worse) In the last 20 years, with the advent of personal computers and smart phones our eyes are spent mostly gazing at screens- not at each other. In a relatively short amount of time, we have become increasingly inexperienced and unpracticed at the skill of nonverbal communication. Essentially we are bad at it. The average adult spends 8.5 hours staring at a screen. This leaves little time to engage in face to face communication and to exercise our nonverbal communication muscles. Our typical mode of communication now is mostly passive, mediated through screens- computers, cellphones and tablets.

Context: Screens are undoing 200,000 years of evolution “Modern” humans as we know it are around 200,000 years old and although verbal communication is superior for conveying detail and analysis, nonverbal communication is the primary way we convey emotion and is the foundation upon which we build our interpersonal relationships. Nonverbal communication is a multi-channeled form; in essence there are many small cues of nonverbal communication that make up the whole of the message we send out.

One of the most noted forms is oculsesics, the study of the messages sent by the eye. It is nuanced and slight, registered by eye contact, gaze, pupil dilation, conjugate lateral eye movements and blinks. To others, it gives strong and authentic cues to another’s personality, true emotions and intent. It is a crucial communication channel in close relationships and in some cases perceptions of intimacy were a function of increased eye contact.

To communicate mostly through screens is to create a situation where the complex messages of communication (verbal and nonverbal) are largely filtered and reduced into simplfied signifiers (i.e. haha, omfg, lol). These signifiers are reductive and ambiguous leaving us confused, anxious and unsure. No wonder we take solace in curating and consuming- it’s a clear, one way relationship that doesn’t leave us feeling like a mixed bag of taxed emotions.

 Solution: Create an intervention With the problem identified, the intent of my thesis is not to judge, berate or guilt others in a process that I find myself an often a complicit and willing participant in. It is also not a pie in the sky, fix the world solution. Instead, my goal is to identify a problem, give context to it and provide a solution- albeit a temporary one. It is a designed experience that uses playful procedures (inspired by Sureralist parlor games) to challenge and be a counterpoint the conventions of our typical modes of communication- one that is uncomfortable and unfamiliar with direct intimacy.

Now that the semester is over, but the project is still fresh in my mind I wanted to do a post-mort on Plinko Poetry. I wanted to give a little context on how Plinko Poetry came around.

Context: Deqing and I were partners in Spatial Media (Jared Schiffman of Potion‘s class) this past spring 2012. We had been partners for the mid-term and before spring break started, we talked about doing a projection on a building or outside. After the break we brainstormed and thought it would be interesting to do a projection on the Astor Cube near ITP. We came up with a combination touch / projection map of a Rubik’s cube on top of The Alamo.

The Alamo has been pranked many times, including the group All Too Flat doing a cardboard Rubik’s cube surface taped on top of the exterior. We hadn’t known this when we came up with the idea of projecting a Rubik’s cube, but when we found out we figured it would still be an interesting idea to elaborate on since it has never been made interactive. When we presented the idea to Spatial Media, the feedback was brutally harsh. There were many concerns about feasibility and also how to actually make the surface a touch surfaceAlso, randomly enough another group also proposed a project using the Alamo and received similarly critical feedback.

After class, Deqing and I spoke about just doing a proof of concept model that was scaled down. So instead of trying to demo a real projection on the Alamo, to make a scaled model with the approximate dimensions and make it a touch surface. A lot of projects at ITP fall to the proof of concept example when you have to present your project, usually due to size, time, scalability or technological constraints, but it just seemed like a wasted opportunity to put so much time into a project that increasingly seemed like a idea that could never be executed.

Back to the drawing board: We decided to spend the weekend brainstorming and re-thinking our concept. Around that time I was increasingly interested in my RWET (Reading and Writing Electronic Text class taught by ITP alum Adam Parrish). I fell in love with the playfulness of munging text and aleatoric procedures. I have always loved poetry and liked the opportunity to add an element of absurdity and user interaction into an expressive form that can sometimes be intimidating. Also, by chance I had recently watched a documentary on Netflix called “Special When Lit“- a film about the cultish nature of Pinball machines. So I had a loose idea to make a Pinball Poetry machine. I thought it would be interesting to project on top of a pinball machine interface and to create poems a person could “play” the machine. Eventually outputting a munged poem.

I suggested this idea as well as a few others to Deqing and he liked the Pinball Poetry idea. However, he brought up a good point that you run into the problem of length of use. Basically- how long does one play, is there a limit, will it take too long, should we impose a time limit, etc. So he suggested a box, but he didn’t know the name of it. After a little research we found out it was the Galton Box or Bean Machine. Designed by Sir Francis Galton, it is a device to demonstrate the probability theory on the bell curve. The Galton Box has been adapted many times into games including: Pachinko (Japan), Bagatelle (France) and Plinko (made popular by The Price is Right). We decided on Plinko Poetry as our concept- funnily enough since neither Deqing and I grew up outside of the U.S. neither of us were that familiar with how well it was known as a pop culture reference.

Design + Implementation: We both liked the idea and re-pitched our concept. In our presentation our biggest concern for point of failure was that programming the software to fit the interface of a plinko machine could be more challenging than anticipated. Either way we knew the only way to figure that out would be to rapid prototype the Plinko frame to start testing as soon as possible. So we quickly put together a frame CNCing and laser cutting the parts. We had some errors- like a hole drilled in the wrong place and we used scrap materials so it looked pretty terrible, but we figured it was more important to get something physical made and we could easily re-cut the materials for the final. A simple Logitech webcam to do the color tracking. We concealed it in a laser cut box and mounted it above the frame with an auto-pole and articulated arm. We also mounted a simple polarizing camera filter in front of the camera to bring out the red of the color chip.

Tech run down: We used openFrameworks color tracking to distinguish the red plinko chip versus the green pegs.
The edge of the screen was calibrated using four discretely mounted LED’s at each corner so that it could be easily re-calibrated.
The text display and animation was handled in Processing. We actually spent a bit of time figuring out what font to use. Most fonts did not display well especially when scrolling. We ended up using Rockwell, a simple legible serif font.
Tweets from @NYtimes and @FoxNews were scraped using a Python script. We hadn’t set up a cron job so we had to physically run this in the mornings, but it was fine for our purposes. It pulled about 200 tweets from each account and then parsed it into 8 different text files (one file per line).

I’m working on the code for scraping the twitter api for both @nytimes and @FoxNews for the Plinko Poetry Machine. So far the code does this:

Scrapes 200 through the @nytimes tweets
Prints tweets and puts them into 4 different .txt files
I stripped out any tweets with an “@” symbol
And also stripped out any hyper links – so basically go though the tweets, and if you find “http” remove the text starting at “http” and ending at the break. This retains any text after the hyperlink.
Then it runs through an if else statement to sort the files.

import urllib
import json
import re

num_tweets = 200

resp = urllib.urlopen(“https://api.twitter.com/1/statuses/user_timeline.json?screen_name=nytimes&count=” + str(num_tweets))
tweets = json.loads(resp.read())

print tweets

file1 = open(“nyt1.txt”, “w”)
file2 = open(“nyt2.txt”, “w”)
file3 = open(“nyt3.txt”, “w”)
file4 = open(“nyt4.txt”, “w”)

i = 0
for tweet in tweets:
output_str = tweet['text']
if ‘@’ in tweet['text']:
continue
if ‘http’ in tweet['text']:
output_str = re.sub(r’\bhttp\S+\b’, ”, tweet['text'])

output_str = output_str + “\n”

if i < 50:
file1.write(output_str.encode(‘ascii’, ‘ignore’))
elif i < 100:
file2.write(output_str.encode(‘ascii’, ‘ignore’))
elif i < 150:
file3.write(output_str.encode(‘ascii’, ‘ignore’))
else:
file4.write(output_str.encode(‘ascii’, ‘ignore’))

i += 1

resp = urllib.urlopen(“https://api.twitter.com/1/statuses/user_timeline.json?screen_name=foxnews&count=” + str(num_tweets))
tweets = json.loads(resp.read())

print tweets

file1 = open(“fox1.txt”, “w”)
file2 = open(“fox2.txt”, “w”)
file3 = open(“fox3.txt”, “w”)
file4 = open(“fox4.txt”, “w”)

i = 0
for tweet in tweets:
output_str = tweet['text']
if ‘@’ in tweet['text']:
continue
if ‘http’ in tweet['text']:
output_str = re.sub(r’\bhttp\S+\b’, ”, tweet['text'])

output_str = output_str + “\n”

if i < 50:
file1.write(output_str.encode(‘ascii’, ‘ignore’))
elif i < 100:
file2.write(output_str.encode(‘ascii’, ‘ignore’))
elif i < 150:
file3.write(output_str.encode(‘ascii’, ‘ignore’))
else:
file4.write(output_str.encode(‘ascii’, ‘ignore’))

i += 1