One Minute Architectures

We recently had the pleasure of working with Lukas – an architecture student at Princeton. While hanging out with us this summer, Lukas continued his work exploring the idea of “One Minute Architectures.”

Since the invention of portable devices that are capable of accessing the Internet everywhere at any time, wireless access points and signal strength have become as necessary as heating and electricity. At the same time, the physical and material qualities of the built world remain ever important: warmth, light, and relative proximity to Points of Interest affect our everyday lives, our bodies, and how we operate in cities.

I am a Master of Architecture student interested in computation, and the “One Minute Architectures” project I worked on at Fathom maps sensorial and often intangible qualities in the built world through a digital tracking system and seeks to render them into 3D visualizations. The goal of the project, which is part of my thesis, is to develop a device that enables users to engage in a new reading of their immediate surroundings; they can seek out routes through points of interest, guided by unique sensorial qualities. The mapping of such sensorial qualities seeks to look beyond the material aspects of architecture, and instead find “one minute architectures.” Inspired by Erwin Wurm’s “one minute sculptures,” where body positions and poses composed by the artist are performed by the museum visitor to create an ephemeral sculpture, “one minute architectures” are spaces that appear and disappear within the public and semi-public realm of the city throughout the day and that are suggested to a user through a device. They are prerecorded by other users and may include good spots to lie on the lawn, sit in the sun, or find warm shelter for work, often blending public and private spaces and advocating for their alternative uses.

Before joining Fathom, I completed a project with Axel Kilian at Princeton that visualizes environmental change by mapping individual geolocations according to where people congregated most during the day. At Fathom I built on this study, focusing on human input data with factors that can be predicted and calculated in the environment.

Recording device for mapping areas of interests. Sensors included heartbeat sensor, ultrasonic sensor (ceiling height), photoresistor (brightness levels), GPS elevation, body temperature, environmental temperature, and pressure sensor (body position; sitting or standing).

Resulting data was drawn as an architectural section, displaying spatial features in conjunction with biometric records collected throughout the day.

I chose the shadow calculation as such a factor. The importance of shade and its ties to thermal comfort becomes particularly important in urban areas during the summer, where shadows cast by buildings create a temperature difference of up to 8°C. The effective prediction of where the shade will fall, in relation to the user’s standpoint, therefore becomes important to how that user will navigate through the environment. Spots that might be considered “too hot/cool” or “too bright/dark” become more attractive when sun conditions are changing. This is not only true for open areas such as streets, parks or walkways, but also for buildings; their openings, for example, play an important role in illuminating and venting inner spaces.

In the project the shadow calculations are represented as visual diagrams and should be understood as an approach to the computation of the sunlight. Processing has been used to develop the calculations, mainly in combination with the Toxiclibs library for calculating the intersections.

As site to test calculations, I chose a small area in Beacon Hill close to the office, due to its diverse topographical features and its manageable size. I gathered building geometries from various governmental websites online and I generated the topography using Grasshopper in conjunction with Google location API. The initial tests, however, were done with imaginative geometry. The problem of generating shadows in a digital environment is obvious, when we think about what a shadow is: the silhouette of an object defined by the position of a light source. In this example a source sun point (S) was set, from which rays (R) are drawn to each vertex of the geometry. If the ray intersects with the geometry only once, thus hitting the outermost vertex (V) relative to the position of the sun point, the ray is prolonged and intersects (I) with the geometry of the terrain below.

These new intersection vertices have to be sorted in clockwise order, to be drawn as a closed shape. The illustration below shows how the shape is being drawn on a 2D canvas in Processing. Points are randomly spread over the canvas, and the midpoint (MID) is determined by calculating the mean value of all the points x and y coordinates. A baseline (BL) is drawn, from which the angle of all the points (V1, V2, V3…) are determined and sorted from smallest to largest angle. Ultimately the points are sorted, connected, and concatenated to a closed shape.

Random points are sorted and connected to a shape. Determined from midpoint MID, a baseline (BL) is drawn from which all the angles from MID to points (V1, V2, V3) are matched and sorted.

To predict the shadows accurately, the geolocation (represented as latitude, longitude, and elevation, the local time, and the sun’s position (azimuth and zenith) are plugged into formulas. Knowing the shadows would enable users to find places and paths of climatic comfort during the day.

This project thus responds to the fact that, given ever more portable technological devices and the resulting ability to work everywhere, personal metadata in predictive ad hoc environments plays an increasingly important role in our lives. Nomadic tendencies – long term and short term — due to rising rents or changing work situations are already prevalent in metropolitan areas. Laura Forlano, an assistant professor of design at Illinois Institute of Design (IIT), provides an illustrative example about the relationship between the spread of Wi-Fi technology and urban nomadism in her article “Wifi Geographies: When Code Meets Place.” In the article, Victor, a young graphic designer living in New York, has shifted his workplace entirely to Starbucks, where he meets other professionals working from the coffee shop every day. Free Wi-Fi and a table is all he needs for conducting his work. Depending on which phase of a project he is working, he shifts in between three coffee shops. Social factors (other professionals with whom he is meeting) and environmental factors (the proximity to print shops or libraries) play an important role in which location he uses. This example shows how scarcities provoke activity outside of a predetermined context (like a permanent office). The importance of local knowledge and the predictability of needs become key factors when navigating through these spaces.

This antagonistic relationship between need and the shaping of an architectural envelope has existed since the beginnings of humanity. Changing societal hierarchies and relationships can be found in all epochs, but technologies and materials – the elements by which physical architectures are constructed – have changed very little: glass walls are still walls and automatic doors are still doors. Their materiality and technology might have changed over the years, but their typological assembly and sequence has not. In our age, dominated by digital resources that are permeating architecture and bleeding into other places, activities are no longer restricted to hard architectural shells, but can be carried out at places that fulfill a specific need at a particular time. This nomadism might not only change architecture in terms of appearance, but it might also shift the focus towards the notion of places, rather than spaces. People might own places privately, but share the ownership, or rent spaces by the hour. Existing infrastructure might be used as is or converted into something that enables a wider range of activities; new structures will be either highly specific, such as laboratories or hospitals, or very flexible, only providing basic amenities. “One Minute Architectures” is an attempt to illustrate this shift from architecture to an infrastructure of comfort, with changing properties and amenities. It examines how we can make ourselves at home not in spaces, but places.

I want to thank Fathom for the unique opportunity of working in such a productive and insightful environment.

Elaine’s summer of research

This summer I researched and analyzed data, learned how to code in p5.js, and participated in the iterative workflow of Fathom. Additionally, I took a trip to the aquarium on my second day of work, practiced building Ikea furniture for the new office, and (unsuccessfully) tried to convince everyone that crumpets are delicious.

Elaine's first day
Elaine’s first day


I spent my first month at Fathom diving into the world of wrongful convictions. I was so amazed, and shocked, when I read the draft of the ProPublica article, Busted. It was an engaging piece to read, but even more exciting to look at the data that supported the article. I found it incredible that all these variables existed—plea date, lab date, lab result—yet at each step a system failure still occurred.

Once I got used to the legal lingo, one of the most interesting parts was deciding which variables to use. Even while narrowing the scope to temporal variables there were numerous: filing to plea, filing to lab, filing to dismissal, plea to lab, plea to dismissal, plea to now, lab to dismissal, lab to now, etc.

Preliminary look at how long people were, and still are, engaged with the wrongful conviction.

Trying to organize and understand these variables in a way that made sense was challenging. It helped to identify the points in the article that were complicated. For example, the sequencing of events was crucial to the story, yet utterly confusing because at each stop the proper order was reshuffled.

Tracking the frequency of events in the first month after individuals were arrested.

In the short span of the ProPublica piece I got to experience and participate in the Fathom workflow. The article provided the background and research for the piece, but the data was everything—it revealed the complexities and supported the widespread nature of the narrative. The data was where the team started before jumping into portraying the narratives of time, demographics, and sequencing. Through the ProPublica immersion, I came to fully understand the importance of iteration, because it happened at every step: data cleaning, variable calculation, analysis, development, design, and text editing until we reached the final.

Women’s Equality Day

Equal Pay for Equal Work screenshot
Check out the final piece here.

I also worked with Olivia and Paul on a piece for Women’s Equality Day—August 26—not every day as we would like. Using national data on job earnings, we analyzed the idea of equal pay for equal work. People often throw around the phrase “women make 81 cents on the dollar” or “78 cents on the dollar,” but we wanted to nuance that figure in order to show what women really earn at the same jobs as men.

The data is very granular; job titles get as specific as “extruding, forming, pressing, and compacting machine setters, operators, and tenders.” However, despite the detailed 550+ occupations each year in the original data, only about 100-115 have annual reported data for both men and women. Even with this decreased data pool, we had some interesting findings. In 2015 alone, women in only 5 of the 119 jobs made more or equal to their male counterparts. Additionally, it’s right to question the generalization that women make 81 percent of what men make for the same job. In 2015, women made anywhere between 56 and 111 percent of what their male counterparts made.

2015 wage gap: percent away from equal pay that women earn.

In 2015, the 5 jobs where women made more than their male counterparts were:

  • bookkeeping, accounting, and auditing clerks
  • police and sheriff patrol officers
  • office clerks, general
  • data entry keyers
  • wholesale and retail buyers, except farm products

These jobs are spread across all industries, and differ by year. Similarly, the pay gap exists in the vast majority of jobs, regardless of industry.

Some additional trends I spent time exploring were the relationship between the percent of women in a job and the wage gap, and the male salary and the wage gap. Neither of these have an absolute, linear correlation, but both have significant relationships. There is a negative relationship between the percent of women in an occupation and the wage gap, meaning that jobs with a larger percent of women may have a smaller pay gap. On the contrary, there is a positive relationship between the male weekly earnings and the pay gap, suggesting that positions with higher pay for men have a greater gap. Neither of these relationships show causation however they do reveal correlation of a few the many factors regarding gender inequality in the workplace.

Percent of women in a job vs. the pay gap. All years 2006-2015.

Male wages vs. the pay gap. All years 2006-2015.

Sea Level Rise

The final project I worked on this summer was independent research on sea level rise. I first became interested in this topic within the field of international law with the idea that there are not existing structures to handle climate refugees and disappearance of parts of, or complete, nations.

I started my research by scoping the existing data, and settling on data from the NASA Socioeconomic Data and Applications Center (SEDAC) at the Center for International Earth Science Information Network at Columbia University. This source provided country-level data for the variable “percent of country population living under a certain elevation.” I used this variable as a prediction for the people who would be impacted by sea level rise even though not everyone would be equally affected due to the uneven nature of sea level rise and different mitigation capabilities. I selected this variable for 1 meter, 3 meters and 5 meters given the incremental and unknown aspects of sea level rise.

Once I started looking into the relationships between countries and regions given the percentages, one of the first calculations I made was the conversion from the percent of the country population to the raw number of people. This started to shape the direction of my research because the countries with the highest percent of the population affected were not those with the most people impacted.

I think the metric we use to quantify the effects of sea level rise can alter the perceived danger, and change the narrative regarding the mitigation and adaptation response. Focusing on the percentage of the country population values the nation, culture, community, language, history, and heritage. On the other hand, quantifying the population that would be impacted recognizes the individuals, families, and lives affected. Both are important, and I will carry this question of how the metric changes the response into future research.

Another aspect that interested me from the data was which regions or countries would be the most impacted. In conjunction with this I wanted to incorporate the variable of development, which impacts a country or region’s ability to mitigate or adapt to the impacts of sea level rise. In my final visualization I decided to use GDP per capita as this indicator.

Only after I’d identified these themes from the data did I start the design process. With the goal of a communication, jumping off piece for future studies, I went through many iterations of how to portray the ideas I extracted from the data: metric of quantification, comparisons across countries and regions, regional trends of impact and development.

I cycled through many versions and considerations: how to show the difference between the country’s percent and raw population, how to ground this idea in geography, how to incorporate geography and regional trends by abstracting away from a traditional map, how to preserve comparisons across countries and across regions, and more.

The final design I landed on was something that (I hope):

  • grounds the viewer in some sort of geographic reality
  • shows the relationship and differences between the percent of the population impacted and the number of individuals impacted
  • provides an alternate view of the same information that incorporates the regional trends in GDP per capita
  • shows the different sea levels: 1, 3 or 5 meters

This piece is still a work in progress, but it combines the research, data analysis, design, and coding that I have worked on for the past few months.

This summer has been an amazing, enlightening experience working with the Fathom team to study, create, and ponder representations about complex topics in our world.

Parima’s two week sprint

Welcome Parima! It’s not often we have high school interns, but Parima was an exception. She learned about us through Girls Who Code after a visit last summer. Over the past two weeks she has been on turbo learning a lot about visualization, information design, and web programming. She even made a pull request fixing a bug in the No Ceilings dataset. It was a pleasure having her. Read more about Parima in her own words.

For my internship, I worked on compiling data from the No Ceilings dataset and presenting it in new ways.

During the first week, I focused on using D3 for the data visualization. I explored this throughout the week along with the female entrepreneurs indicator. The reason I chose this indicator is because I have a personal interest in entrepreneurship, and have participated in some entrepreneurship programs where I worked on developing two ventures. My most current venture is called School Munch: a food delivery service for high school students. There have been several successful pilots, and our official launch will be this October.

Below is an interactive world map built with D3. While working on this mini project, I noticed a trend that African countries have the highest amount of female entrepreneurs. I found that many African women need to support themselves, and if they are unable to find jobs, they start small businesses of their own.

View this post on a larger screen for the interactive version of the map above.

For the second week of my internship, I focused on working with P5*js. Below is an interactive line graph representing data on the slum population in urban areas. The reason I chose this indicator is because my parents are originally from India, and as a result I go there frequently to visit my extended family. When I am there, I always see poor people on the sides of the street begging for money. Sometimes there are even entire families, and it is really striking to see people living in such conditions. Realizing how big of a problem poverty is, especially in India where more than half of the population is below the poverty line, I have taken an interest in this issue and wondered what could be done to help them and improve their living standards.

View this post on a larger screen for the interactive version of the chart above.

What I have learned here at Fathom could be applied to help in these situations. I could use the map above to show how female entrepreneurs are growing in numbers across the world, and help inspire other women to pursue entrepreneurship. In addition, I could use the line graph of the slum population to raise awareness and encourage people to help!

Let’s Hear it for the Girls!

Alas, we’ve released another episode of our podcast, Especially Big Data. The episode, Let’s Hear it for the Girls, dives into the many factors contributing to the dearth of women in tech– most of which are not captured in numbers.

There is tons of data on the small distribution of women pursuing computer science degrees or programming careers, but we wanted information on the experiences that are more difficult to quantify; like the number of cases where a team of primarily male developers goes out for brews and forgets to invite their sole female teammate; or the percentage of workplaces where restroom accommodations simply aren’t made for women (this is real!); or the instances where an individual was too afraid to speak up because of the glaring awareness that she (or he) was different from everyone else in the room. Our latest episode dives into the subtle and nearly invisible gestures, mindsets, and structures that make participating in tech more difficult for those who are under-represented.


For this episode we spoke with guests like p5.js creator, Lauren McCarthy, Harvey Mudd president, Maria Klawe, a handful of girls from Akamai’s Girls Who Code chapter in Boston, and our own Leslie Watkins. Olivia also shared tons of insight behind the scenes.

Lauren shared some fascinating stories on how gender influences the ways her peers approach professional opportunities. She also discussed its unexpected impact on the reception of one of her projects.

Next, we untangled two accounts of a single moment that nearly broke the internet. Back at the 2014 Grace Hopper Conference, Maria Klawe refuted comments made by Microsoft CEO, Satya Nadella, where he essentially said that women who worked hard could trust that the system would reward them. Both the audience and the press went wild at his statement. We’ll hear from Maria, and Leslie– who was sitting in the audience at the time– on what really happened.

To hear more unquantified experiences from badass women in tech, tune in!



Related posts
Especially Big Data
Oh, the places we go
Oh, the places we go

We’re excited to announce the release of the second episode of Especially Big Data, our new podcast. The episode, Oh, the places we go, explores the great lengths people travel to collect a single data point, and the many issues they encounter along the way. From the door-to-door surveys of the U.S. Census, to the mountain treks of community health workers, and then to NASA satellites hovering 650 km above the earth, tune in to hear some exciting tales from the trails.

Data collectors go to great lengths to gather information. For some, those distances extend beyond earth’s atmosphere. Oceanographer Gene Feldman gave us a sneak peak at how NASA uses satellites to measure the livelihood of microscopic plants in the ocean. The information that their satellite captures in single minute would require an entire decade for oceanographers to collect by boat.

We also spoke with Steve Klement of the U.S. Census Bureau. Steve told us about the complications of surveying and serving the same population– as the census captures information about the public, for the public. While the bureau goes to great lengths to make information accurate and accessible for a general audience, there are also occasions where they need to suppress, or hide information to protect the privacy of businesses and individuals.

Meryn Robinson of Dimagi also spoke with us about privacy and the sensitivity of information– particularly when it comes to metrics on health. Meryn is a senior research coordinator, and she helps train community health workers around the world to use Dimagi’s data collection software, CommCare, on phones and tablets. While Meryn spends most of her days at an office in Cambridge, MA, she has co-workers who climb mountains for their daily collection efforts.

In addition to the full audio piece, we put together a few teasers to highlight moments from the episode. We enlisted Rachel, our in-house animator, to bring a few of our favorite audio clips to life.

Gene Feldman, an oceanographer at NASA, explaining how data collection instruments change once launched into space.

Meryn Robinson, a senior research coordinator at Dimagi, describing how her commute is different from that of a co-worker, who was training data clerks in Guatemala.

Tune in to Especially Big Data here or check out the links below:

Related posts
Especially Big Data
Let’s Hear it for the Girls!

Founded in 2010 by Ben Fry, Fathom Information Design works with clients to understand complex data through interactive tools and software for mobile devices, the web, and large format installations. Out of its studio in Boston, Fathom partners with Fortune 500s and non-profit organizations across sectors, including health care, education, financial services, media, technology, and consumer products.

How can we help? hello@fathom.info.