I posted this photo to Instagram on April 2, 2013, along with the caption, “Inklings of a thesis?” I actually remember thinking at the time, “I bet this ends up in my process book.” And so, over a year later, here we are.
The time since has been a roller coaster ride, to say the least. It's hard to believe that barely two months ago, I scrapped my existing prototype and started from scratch yet again. But looking at that list above, I'm comforted by the fact that I ended up more or less where I began. This site tells the story of how I got here, the thinking behind Slow Down, and the many fits and starts along the way.
Slow Down is an email digest that allows you to select up to 10 friends from Twitter and Instagram whose posts you don’t want to miss. You can choose to receive your digest either weekly or daily, at a specified or random time. Each email consists of four sections:
At the bottom of the email, there’s an appendix containing the original tweets and photos from everyone you’ve subscribed to using Slow Down. All pieces of content in the digest have their own URLs, allowing you to share them as links or post them back into your social media stream(s). You can also click through to view any Instagram photo or tweet in its original context on the web.
Slow Down serves two distinct purposes. On the one hand, it’s a useful tool for managing your social media feeds. It reduces both the volume and velocity of the stream, by constraining the amount of content you receive and delivering it at a predictable time. The constraint of ten people is deliberate, and important. So much of our digital stress is a problem of quantity, of not being able to hear the signal through the noise.
Slow Down also experiments with the nature of social media and the ways in which we consume it. The mashups, haikus, and podcasts all take familiar content as source material and transform it in unexpected, often delightful ways. I hope that when people interact with Slow Down, they feel something like, “I wonder what I might find here” rather than “I need to catch up.” In summary, Slow Down:
Notebook: February 4, 2014
To be honest, I'm not even sure anymore. I read a lot of blogs, and think a lot about technology, and I used to work at Facebook. I guess on some level I've been dissatisfied with The Way Things Are on the Web for a while.
In April of 2013, I made a “thesis almanac,” a speculative document containing a preface, table of contents, introduction, references and acknowledgements: my thesis without the actual thesis part. As Liz said at the time, it was a “kind of commitment to a thing.” In my introduction, I explored the relationship between efficiency and technology, and began to consider alternatives to our current trajectory.
Blog Post: April 2, 2013
One of the primary goals of technology throughout history has been to increase efficiency: the efficiency of our manufacturing processes, our communication networks, and (more recently) our daily lives. The word “efficiency” generally carries positive connotations; thesaurus.com lists among its synonyms the words “ability,” “productiveness,” and “skill.”
With my thesis, I intend to question the effects of efficiency on our interactions with the world. How has our experience of efficiency evolved over time? What do we sacrifice when we become more efficient?
Historically: The Assembly Line and the Effects of Mass Production
During the Industrial Revolution, new technologies such as the assembly line dramatically increased the efficiency of manufacturing processes. Whereas previously, a skilled craftsperson was responsible for creating and assembling an entire product, now the work was divided among a series of semi-skilled people, each responsible for repeating a specific task. By increasing the productivity and decreasing the cost of manufacturing, the assembly line fueled the growth of many industries at the beginning of the 20th century. At the same time, however, it separated workers from the objects they were helping to create. Efficiency, in this case, came at the cost of craftsmanship and creative engagement.
Currently: Smartphones and the Red Notification Dot Problem
Fast forward 100 years or so, and technology continues to promise progress via increased efficiency. We know the exact moment to leave our homes to catch the subway, then we read our email while on the train, then we use our GPS to optimize the route to wherever we’re going. If we’re running late, we call and let someone know. All of these services are, arguably, convenient, but I wonder if they also cause us to be less engaged with our surroundings. Our smartphones notify us whenever new information becomes available, and we struggle to resist the pull of so many red notification dots on our homescreens. Variable reinforcement keeps us constantly checking; the red dot looks the same regardless of its true importance. We are easily distracted, and less present.
Speculatively: Google Glass and Filtered Reality
What might a more efficient future look like? As we walk down the street, our Google Glasses show us only the storefronts it thinks we’ll be interested in entering. At a crowded bar, we see only the people with whom we share interests or friends. When we get home, we watch only the news that confirms our opinions instead of challenging us to think critically about them. This is a more efficient future, but is it a better one? It’s certainly not the future I want to design.
After a summer of training to be a fitness instructor (and doing thesis research), I came back to school excited about three potential problem spaces. I had a sense that they were all connected somehow, but wasn't sure how to articulate that yet.
Blog Post: September 23, 2013
Over the next few months, I refined my ideas, but there were still three of them, and they still all seemed different. I decided to focus on one idea at a time and see what would come out of that.
Blog Post: November 20, 2013
I’ve spent the past few weeks mostly banging my head against a wall because every thesis-related idea I have seems pointless, redundant, or “would have been cool three years ago.” I’m still really interested in my topic area(s), and continue to read all sorts of relevant things, but when it comes to actually doing anything I feel stuck and frustrated (said every grad student ever). I’m also torn between these three possible problem spaces:
In my head these three problems are all related but I haven’t been able to articulate the connection in a meaningful way yet.
By the Spring 2014 semester, I had come up with a more specific list of criteria for what my thesis should be like. I still didn't know WHAT it was going to be, though...and wouldn't for a bit longer.
Blog Post: February 3, 2014
User Experience Criteria
Ideas in hand (head?), I set off to do some research.
Many more articles contributed to my thesis thinking (and are listed in References), but these four in particular stand out now as the ones that had the most influence on Slow Down.
“What is the Fast Web? It’s the out of control web. The oh my god there’s so much stuff and I can’t possibly keep up web. It’s the spend two dozen times a day checking web. The in one end out the other web. The web designed to appeal to the basest of our intellectual palettes, the salt, sugar and fat of online content web. It’s the scale hard and fast web. The create a destination for billions of people web. The you have two hundred twenty six new updates web. Keep up or be lost. Click me. Like me. Tweet me. Share me. The Fast Web demands that you do things and do them now. The Fast Web is a cruel wonderland of shiny shiny things.”
Jack Cheng’s Slow vs. Fast Web distinction served as a useful framing device for me, and Frank Chimero's eloquent words on the increasingly fast pace of our digital (and analog) lives resonated strongly.
“A mythology of speed is one of willful ignorance to the small details that hold the whole arrangement together. And, I think, if you’re building things for the internet, those small details matter, because they are repeated ten-fold, hundred-fold, million-fold, as they are replicated effortlessly through screens, across the globe, and into people’s consciousness for countless hours of exposure. Economies of scale make small decisions matter, but speed—both in making those small decisions and in interacting with them—makes both sides blind to what’s going on. We’re thoughtlessly writing things we can’t read, because we’re going too fast.”
I read some books, too.
“Here’s the thing about technology: for everything we gain, we lose something in return. We’re like the clown in the circus act, carrying an armload of bright rubber balls. We pick up one ball and drop another, and we go on, convinced we’ve made progress when all we have is a different set of balls. We’re nostalgic for the way things were, yet we crave the things that are not yet, and I have found no other way to come to terms with this than to acknowledge both what we have gained and lost, and at times look at the balls we’ve dropped and say, no, actually I’d rather keep that.”
“I refuse to accept that the only good response to an imperfect technology is to abandon it. We need more specific criticisms than the ever-present feeling that ‘something’s not right.’ What thing? Developing a political agenda to remake, improve, or forbid technologies requires some sort of rubric: how can I judge what I’m using? What are the deleterious impacts? How are they specific to these media and this time? Which effects are *caused by* the technologies and which are *enabled by* the technologies and which just happen to *occur through* the technologies? What are the ethics? What are the mechanics? What is the baseline?”
My research also included a survey of similar or like-minded projects and/or products. The following all served as inspiration.
As my research continued, I started to synthesize my thinking and clarify my ideas by posting weekly blog posts along with assignments for David Womack's Narratives and Interactivity class.
Blog Post: January 20, 2014
That technology is making our lives more efficient/optimized.
That something gets lost, or at least changes, as a result of this process.
That valuable and/or meaningful experiences often result from chance, accident, error, or some combination thereof.
Serendipity (vs. efficiency or predictability)
Presence (vs. distraction)
Quality (vs. quantity)
Slowing down (vs. speeding up)
Archival (vs. real-time)
We consume less of “the stream” and it consumes less of us. Our attention is no longer being commoditized. We are connected without being distracted and we are moving slowly enough to see our reflection and reflect back about it.
We might be less productive, or make less money, or have fewer objective measures of “success.” It may not always be possible to use the computers in our pockets to answer any question we have about anything at any time. We may feel like we have fewer connections to the things in our world, but perhaps those connections we do have will feel more meaningful.
Blog Post: February 10, 2014
After a productive conversation with Mark, my advisor, I realized that my fundamental goal is to connect people to things they didn’t realize they were connected to, by revealing paths that enable unanticipated discoveries and/or experiences. My method for surfacing these paths is to restructure all the “stuff” of our digital lives along alternative axes such as location, rather than reverse chronology.
Blog Post: February 11, 2014
Describe the setting in which your thesis takes place. What are the characteristics that define the setting? What impact does the setting have on the action?
My thesis takes place in two parallel settings: the digital world and the physical one. In the digital world, it exists as a counterpoint to the prevailing setting, “the stream,” in which content flows by us at an increasingly rapid pace and is becoming harder and harder to keep up with. Much like its physical incarnation, the stream is linear, roaring, fast. My thesis, on the other hand, is multidimensional, quieter, slower. It provides a space for reflection and thoughtfulness in both the digital and physical worlds, and preserves the nonlinear, unpredictable paths that otherwise would have been eroded by the stream.
While I had a sense of what I wanted my thesis to be about pretty early, I had no idea what I wanted it to actually be. For a really long time. My “process,” if you can call it that, went something like: Have an idea. Get extremely excited about the new idea and decide it is the best idea ever. Prototype the idea. Decide the idea is terrible. Repeat. This went on for about a year.
I didn’t necessary think of this as a thesis prototype at the time, but looking back I think it led to everything that happened next. As a side project last spring, I made it a goal to complete a photographic random walk at every subway station in New York City. Each walk assumed the following structure:
My ultimate aim with this project was to observe and document visual information – people, moments, interactions, encounters, perspectives – that might otherwise have gone unnoticed or ignored. I wanted to encourage people to pay more attention, to notice more of the infinite tiny details that, together, make life fascinating.
This prototype, which I developed during the first half of the fall 2013 semester, attempted to bring a little serendipity back into our lives. It was called NYC Random Walks, a mobile app concept that forces people out of their daily routines, encourages them to travel to parts of the city they wouldn’t otherwise have visited, and once there, to pay more attention to what’s in front of them.
To use the app, you first shake your phone and then are given a randomly selected subway station somewhere in New York. Upon arriving at this destination, you put away your phone and use a small crystal-ball-like object to guide you around the neighborhood.
I prototyped the crystal ball using Arduino and a ping pong ball. Once you have arrived at the random destination, it blinks one of four colors between one and four times. Each color corresponds to a cardinal direction (north south east or west) and the number indicates how many blocks you’re to walk in that direction. The purpose of the guide is to allow you to put your phone away for the duration of the walk.
At the end, the app prompts to record your experience by jotting down some notes and taking a photo, and then you can view all of your previous walks on a map.
After completing this prototype, I began to feel less and less excited about the idea and it has been hard for me to articulate exactly why. I think it remains unclear who the audience is and what would incentivize people to use this. I feel like the people who would use it would already be adventurous types with a lot of free time on their hands, who are already inclined to do things that challenge their comfort zone. It was difficult for me to figure out how to reach those people that need challenging, or even to decide if that’s really what I want to do at all.
For the remainder of the fall 2013 semester, I decided to switch gears and focus on filter bubble as a problem space. Not Recommended For Me is an experimental anti-recommendation engine that provides suggestions for things you probably won’t want to read, watch, or do. It is an attempt to mitigate the effects of filter bubble by appropriating an existing and familiar web technology in an unexpected way.
I built the site by first using the Facebook API to compile lists of all of the books, movies, TV shows, and general topics of interest that are “liked” by myself and my friends. Each of these lists had over 1000 unique items.
To generate anti-recommendations, I used Amazon’s Mechanical Turk service, which allows you to pay anonymous workers small amounts of money in exchange for completing simple tasks such as tagging a photo or summarizing text. I selected a sample of 250 items from each of my lists of likes, and created tasks such as the one shown here that instructed workers to provide anti-recommendations for each item. The workers were presented with the prompt: “someone who likes the above thing will probably dislike this other thing” which they could fill in below.
When you log in to Not Recommended, the site dynamically checks your Facebook likes to see if it contains any of the items that were sent to Mechanical Turk, and if so, displays the anti-recommendations. If your Facebook data is too sparse, you are prompted to select from a list of six items for your favorite, and then you are given the anti-recommendation for that particular item.
Inspired in part by Somewhere and Instaphotobooth, a project for another fall 2013 class, my idea here was to create the digital equivalent of looking into lit windows across the street at night, an experience most New Yorkers are familiar with. Even though most of the photos you see are taken by complete strangers, the process of viewing them is strangely fascinating, as the locations themselves have personal meaning. Searching for photos taken at old addresses is particularly compelling, and quite nostalgic.
After building this and playing around with it, I realized I hit on a bigger idea that I wanted to pursue much further: a shift or reframing of the way we consume digital content, with less focus on the “newness” of things and more on other attributes that enable more serendipitous discovery. So far I had explored organizing photos by location, but there were many other axes with potential for interesting results, as well as other types of public content (e.g. text-based, such as tweets).
While struggling to define the audience and use cases for Cityscapes, I had a new idea: use art and art history as a means of framing or packaging the experience. This led to a quite a departure from the Instagram search app, though the underlying goal remained similar. Look has three basic premises:
Ultimately I decided to abandon this idea for a few reasons: I had trouble finding an open-source database of art to use, I couldn't decide if I wanted it to be more about creative expression or educational, and I wanted to build something that was fully functional rather than a simulation of the experience.
In the middle of March, I still didn’t know what I was going to make for my thesis. This was terrifying and horrible and extremely stressful. Then, one day, in the middle of Paul Ford’s Content Strategy class, I jotted down a new idea that would eventually turn into Slow Down. In retrospect it’s hard to explain how this happened, other than the fact that there were a lot of ideas stewing around in my brain for a long time and they finally jelled into something that had potential.
Notebook: March 20, 2014
A few months earlier, I had brainstormed a similar concept, though it was more about filter bubble than social media overload. I do think that the form of this idea (an email digest) directly impacted the genesis of Slow Down.
Notebook: January 22, 2014
Step one: come up with a name. Most of these were pretty bad. Then one was good!
Notebook: March 20, 2014
Step two: sketch! Here is the sign up form. It looks kind of like the sign up form on the site.
Notebook: March 20, 2014
Step three: profit! Just kidding. Step three turned into steps who even knows how many, aka coding the thing. Refer to Development for all of the gory details.
I spent the rest of March and most of April programming. There was a lot to do! The work alternated between satisfying and frustrating, in more or less equal doses (with the tiebreaker going to frustrating).
Notebook: March 27, 2014
In moments of frustration, I vented my code-rage on Twitter. These tweets would eventually appear in my Slow Down digests, once I finally got them working.
Occasionally I took breaks to focus on the design and branding.
Notebook: April 3, 2014
Illustrator: April 3, 2014
And also to watch baseball.
By mid-April, I had the sign-up form and email content mostly working. The next step was to figure out how to actually send the emails at the right time.
Notebook: April 10, 2014
Turns out HTML email is a throwback to 90's web design. Yay, tables! Inline CSS everywhere!
The First Email! April 12, 2014
I don't even want to talk about how much time I spent trying to get the emails to send at the right time in different time zones.
I'm storing all the Social Podcasts as .mp3 files on my server, so I made this joke referencing Silicon Valley. It may not turn out to be a joke, though, if I end up having a lot of users.
I sent an email to my classmates on April 25, asking for help beta testing Slow Down. Many of them signed up and quickly started to find bugs. I spent the next week or so fixing these bugs, testing more edge cases, and incorporating feedback on the interface and interactions.
On May 1, I sent another email to a wider circle of friends. These new users uncovered yet more bugs, and continued to send me valuable feedback.
The biggest pain point for almost everyone seemed to be having to select 10 friends to subscribe to. Many users expressed anxiety at this point, or concern that the people they didn't pick would somehow find out. In response to this feedback, I added a “Choose 10 People For Me” button to both the sign up and edit settings pages, which selects randomly from everyone you follow on each service. I also clarified and refined much of the instructional copy, based on users' comments.
On May 7, I was finally ready to launch Slow Down to the wider internet. I tweeted about it and posted a link to Facebook, and enlisted the help of some SVA faculty members to get the word out. People started to sign up!
As of May 13, 2014, Slow Down has 221 active users. I plan to continue developing and maintaining the site indefinitely.
Here are the various Slow Down screens, as they appeared at the beginning of May. Slow Down is fully responsive and feature-complete on mobile devices as well.
You can also view a PDF of an example email digest.
Building Slow Down would not have been possible with the support, guidance, and input of so many people. Thank you to everyone who contributed!
Mark Shepard Thesis Advisor
Eric Forman and Megan Fath Thesis III Faculty
Liz Danzico Department Chair
Christine Aaron Assistant to the Chair (2012–2013)
Gwendolyn Kurtz Assistant to the Chair (2013–2014)
Frank Bonomo Systems Administrator
Matt Pokrzywa Technical and Emotional Support
Stephen Nickson Public Speaking Coach
Frank Chimero, Anil Dash, Paul Ford, David Womack, and Jeffrey Zeldman, for wisdom and insight regarding the internet and life
Ashley Quinn and Beth Wernet, for logo design input
Alex Ainslie, Emily Barlow, Frank Bonomo, Matt Brigante, Olivia Coffey, Liz Cunningham, Tyler Davidson, Jennings Hanna, Tom Harman, Sarah Henry, Avi Hoffman, Tammy Hwang, Meghana Khandekar, Thomas Kho, John Kim, Rachelle Milne, Shelly Ni, Sneha Pai, Matt Pokrzywa, Ashley Quinn, Melody Quintana, Nathaniel Roman, Nirav Sanghani, Dan Schafer, Nicole Sylianteng, Arun Vijayvergiya, Sam Wander, and Beth Wernet, for beta testing and user feedback