Data & Humans: A Love Story


>>From the Library of
Congress in Washington DC.>>So let’s begin. Our opening keynote speaker just got
back from the Okavango River Basin, Africa’s largest wetland, where
he and his team of scientists, filmmakers, and journalists
collected environmental data to document the river’s ecosystem. This National Geographic
project is making all of the data open-source
for us to explore. Participating in this expedition
and almost getting eaten by a lion, I’m telling you, you should
really read the journal article, is one of many ways in which
Jer Thorp is using data to improve our world. His portfolio reflects the dynamic
ways in which data can be applied, from creating the algorithm for the
9/11 memorial that placed the names of the deceased closest to those
they knew, to developing a tool that constructs a detailed
picture of how information moves through social media, to leading
a theater group to interpret and perform 120,000 objects
in MoMA’s collection. These projects demonstrate how Jer
is changing the public’s perception of data from cold, remote, and sterile to interesting,
relevant, and vibrant. Even though many of us
haven’t been able to trek through Sub-Saharan Africa, I believe that we have the same
mission as Jer in many ways, to make information in all of
its forms relevant to the public in the hopes that it
will improve our Earth, ourselves, and our future. Please welcome award-winning data
artist, cofounder of the Office for Creative Research, 2013 National
Geographic Emerging Explorer, and adjunct professor at New
York University’s Interactive Telecommunications
program, Jer Thorp. [ Applause ]>>Jer Thorp: Thank you. Good morning. I think it goes without saying that I feel tremendously
honored to be here today. I’m a Canadian and, nonetheless,
I’m really honored to be here. I came up with this entire
talk to give this morning, I run a studio called The Office for
Creative Research and we do a lot of work surrounding collections,
and so I put together a talk which was specifically about
our work around collections, although as was just mentioned, we
do a lot of very different things. I was almost finished with the
talk when I read the program for the event when I found out that
the talk that I was actually going to give was called “Data
and Humans: A Love Story,” which is a title that
I didn’t make up. And so, I thought, okay,
I’ll do this talk instead, and what will this talk be? Well, obviously, this talk is
going to be about data and humans, and to start to talk
about data and humans, we must begin, I think, with data. There’s an artifact of our
understanding of data which many of you have seen before
which is this pyramid, and it describes a kind
of progression from data at the bottom upwards
towards wisdom. And I think the shape of
that pyramid defines a lot of what we’ve been trying to
do with data over the last five to ten years, and that
is create as much — collect as much of it as possible
with the hope that, at scale, we can also get wisdom at scale. But what I’m going to
argue during this talk is that we’re missing something
at the bottom of that pyramid, and I’m not going to
give it away yet. But I am going to start with a
little bit of wisdom from one of my favorite people,
Ursula Franklin, who actually just recently
passed away. She’s a metallurgist
and a thinker and writer and in her Massey Lectures, she
talks about the idea that most of the world of technology that
we live in wasn’t really made with our well-being in mind. And I think, as we are
constructing data systems and as we’re constructing
a world of data to live in, this has never been more true. And so the central question that
I want to bring to us today is, what is it like to live in data? Because, at one time, we might have
been satisfied with collecting data. At one point, we might have been
satisfied with visualizing it, but more and more and more,
we’re actually creating systems and worlds that we must reside in. This is a diagram by a
sociologist named Jacob Moreno — made this in the beginning
of the 20th century and it’s called a sociogram, and you
can think of it as a social network. This diagram is of a seventh grade
classroom, girls on the right, boys on the left, two brave vectors
of puberty crossing that gulf. I like this diagram
though because I think about how it would have been made. The scientist or the
researcher would have had to sit in the classroom and they would
have had to watch the students. The students would
have to watch them. They would have had to
smell that teenage smell and learn people’s names
and understand that, while these objects are being
drawn an circles and triangles, that they are in fact humans. How would we do something
like this today? Well, we might do it
like a university or, a research group rather, in
Cambridge, Massachusetts, did where they used the Twitter API. They pulled 600,000 tweets from the
New York area, and they ran them through something called
sentiment analysis, and sentiment analysis attempts to
decide which tweets are sad or angry or happy, so on, and so on. They found out that the
saddest place in New York City, according to their study, was a high
school called Hunter College High School, which is up
above the reservoir. It shows up as a purple dot
above the reservoir in this map. And I think a lot about what it
have been like to be those students. This was a September day, and the day before you were not
the saddest school in Manhattan, but the day of, you were the
saddest school in Manhattan. And I often think about what it
was like to be in high school, and I can only imagine how weird
of a day that must have been for these students
to arrive in class. I don’t think the researchers
were thinking about that, because they didn’t, I don’t
think, understand these data points as humans, but instead
as returns from an API. And this story — I
become obsessed with partly because they made a big mistake — their algorithm that they used
to locate the tweets was wrong, and so it wasn’t actually
this high school, but it was somewhere else
nearby that they had determined to be the saddest point
in Manhattan. And I actually went and visited the
vice principal of this high school because I was so interested in
this and I asked her, I said, what was the mood in the students like that first day
that this was published? And then she said, nobody cared. I was like, wow, that’s interesting. Why did nobody care? And they said, well,
we knew it was wrong because high school
students don’t use Twitter. We know that they’re not
allowed to use social media. We know they are using it, but
Twitter is for old people and nobody in our high school uses it,
which is kind of amazing, right? Like these guys had gone and
published the study based on a premise which was just
false from the beginning, and I think this example
highlights a lot of our problems. This amazing ability we have to
harvest and collect more data has to be balanced with the idea
that the mechanisms we use to do so are bringing us further
away from the people that we were engaged
with in the beginning. So what is like to live in data? It sucks. And I don’t mean
that only in the pejorative, I also mean that it’s like
a gigantic vacuum cleaner. We are the subject
of extraction, right? The word collection is in
the title of today’s event, and then word collection or the word
gathering are these kind of neat and polite ways to
describe something, which I think is a little
more violent, we’re scraping and abstracting and mining
data from individuals. A few years ago it — actually in
2009, I became a little bit obsessed with these tweets that people would
write when they landed at airports, so they weren’t allowed
to do it at the time, but people still what they would. They would tweet as soon as they
landed, and I just pulled a few from the feed this morning these. These are quotidian statements. There’s nothing particularly
exciting here, although they are little
glimpses into somebody’s life, but what they also are, are
data points, because I know where they’ve just landed, and I know from their
profile where they live. We could reconstruct a system from
just those tweets, which is a kind of simulation of air travel,
and this actually came out of a discussion with a friend
of mine who is an epidemiologist, and epidemiologists are
always looking for ways to model human travel, and we
thought maybe could model them based on these weird, kind of
thinly veiled, show-off tweets that people were making as
they landed on the tarmac. This is the project of this
which I made a week later, it’s called “Good Morning.” This is everybody on Twitter
in 2009 saying good morning. And you can see the waves
travel around the world. Green dots are people
who get up early, red dots are people who get up late. I’ll let you see the East
Coast and the West Coast, and you can make your decision as to
which one is lazier than the other. There’s nothing exciting
happening here, but there was something really
captivating for me about being able to construct something from
nothing and to do it at scale. And I wasn’t the only one, you
know, advertisers were very excited about this potential that we
could scrape up data from people and put it together into stories. In 2011, a couple of
researchers discovered that Apple was storing
all of your location data, and it was storing it
unsecurely on your device. What that meant is that anybody who
knew where to look could find a file that would say exactly where you
had been since you owned your phone. I don’t really have to think
about that for very long to understand the privacy
concerns here. But a group of people
that I was working with at the New York Times lab at
the time, Matt Boggie, Jake Porway, Brian House, and I decided,
what an opportunity! There’s all this data. Let’s get as much of it as we can,
and so we started a project called “OpenPaths” and you can
still find it openpaths.cc, and the way that it works today is that you install an
app on your phone. That app collects data about your
location, but it stores it securely and only you have access to it. There’s an additional thing
that you can do as well, which is that you can share
this data with researchers, for whom this kind of data
is kind of gold, right? Location data is really, really
exciting for epidemiologists to study disaster response for
urban planners, but almost all of this data is held by a
couple of private entities, and to get that data
is very difficult. So our idea was, can we
make a collective tool so that people could
visualize this data? They could understand what’s
going on in their world, but they could also share
this with researchers. So this is what the
interface looks like. You can go and look at your date
and kind of see where you’ve been and the patterns of your
travel, and so on, and so on. It’s pretty simple. But, for us, it was an
experiment in something, it was an experiment
in data ownership. What happens when we give
people ownership of their data? And, in this case, they
really do have ownership. If I wanted to look up your data
on OpenPaths, I run the server, I did for a while, I can’t do it. It’s encoded on our server. It’s encrypted on our server. Only you can find it. An idea was that maybe this idea — this experience of data
ownership would foster a kind of personal relationship with data, which I think we’re pretty far
away from right now, right? I think when I say this phrase
to people, they’re like, what are you talking about? But there is a lot of personal
information in this data. This is a graphic by Brian House and it shows what he calls
his meaningful locations, and this is just derived from
the data, and the data knows where he lives, where his
girlfriend lived at the time, where he practiced with his band. These pieces of meaning
are encoded in this data, just by the nature of it. I did something a little
bit different with my data, these are little arrows
and they’re the direction that I was traveling in, and this
reads like a clock on its so over on the left here is the
beginning of the day. That’s me going to work. I go, at the time, I
went northwest to work. And then that’s lunch up at the
top, and then down at the bottom, I just moved to New York. I never went home, so that’s kind
of me going all over New York, and there might be some
stumbling bits there in the end in the very late evening. But the idea of this was,
could we make an artifact that people would put
up on their wall? I had this dream that someone
would put this up on their wall above their bed, and before
they went to bed at night, they would look up at it
and say, good night data. It’s weird, right? And it’s weird because this
is what this data looks like. This is one data point, another
data point, another data point. I moved to New York City from,
after living in Vancouver, Canada, for my whole life, and —
or for most of my life, and this is the moment I stepped
off the plane at LaGuardia Airport. It’s kind of the beginning
of a new chapter in my life. The second data point is, that night
I had this crisp glass of Riesling up at this Thai restaurant
on Amsterdam Avenue and I would have forgotten entirely about if it weren’t
for this data point. And then this data point, five and
a half years ago, I opened the door at the end of my hallway
of my apartment building and met my partner Nora, who’s
here, and instantly fell in love, and this point to me, I have this
idea that I’ll sit my grandchild down on my lap and I’ll
show them this data point. And I’ll say, this
is how it started, and we made some progress there. This is my favorite data
point, our son who was born, oh, just over a year ago. And, I actually, when I saw
that, when I wrote this out, I remembered something
very interesting, which is that we are all
kind of born as data. This is what we share with
our family and friends, we share these numbers and these
statistics to prove something, to prove that this child was born. And similarly, these numbers,
they’re evidence, they’re evidence of our existence of our, of our,
of our, our time on the planet. But anyways, what is it
like to live in data? I think that there’s this
sensation for most people that we’re being used, and
we are being used, and again, primarily by advertisers. So most of the machinery around
data goes towards placing ads in your web browser,
you know, we might argue that maybe the machinery
behind the NSA and surveillance is slightly more
sophisticated, but probably not. Most of the effort is being done
to place ads in your web browser. I teach a class at
New York University, and one of my students had this
great project called “Cookie Jar.” And she took collections
of people’s advertisements and then she sent them to
strangers and had them write stories about people, based only on their
ads, so it was kind of a way to reverse the — advertisers are
trying to find things out about you and give you ads, and could we turn
that around and could we take ads and try to find something
out about ourselves? And, at the time, I was working
on a project which allowed me to collect all of my ads, so this is
a month worth of web ads that I saw, and I sent this to ten strangers
and paid them ten dollars and asked them to write about me. And it was interesting [laughter]. I learned some things about
myself, actually, you know, I learned nothing about myself,
but I learned some things about what advertisers believe about
me, which is that I’m in my 30’s, I live alone, and I
play video games. None of those things are true. This is my favorite
one, I love this one. I want to get it framed [laughter]. But there’s something
interesting happening here, right? Advertisers are collecting
data about us, a then they’re building
these profiles, and they’re making decisions based
on that data, and those decisions, more and more, are
affecting our lives. They might affect the level
to which we can be insured. They might affect the type of health
insurance we’re being presented. They might affect the
type of credit card that we’re being offered, right? And so, this tool that we’ve built in my studio called Floodwatch
allows you to keep track of the types of ads that
are being shown to you, and like the last project I showed,
it allows you to donate this data to researchers, so researchers,
in particular, who are interested in discriminatory advertising
practices, so places in which advertisers
are discriminating based on the things you’re not
allowed to discriminate on under the Constitution, race and
age and gender and religious belief and political affiliation. Those are all things that
we’re trying to investigate and actually there’s no real good
way to investigate them right now, and so we’re building this tool. We’re hoping to get tens
of thousands of users so that they can collect the
largest database of advertising data and then just don’t
give it to advertisers. So you can find out more about
the tool at floodwatch.ocr.nyc, and we’re releasing a new version
in the fall, so if you’re interested in this, I would say hold off, because our next version’s
going to be really amazing. What we’ve done, I
think, in the creation of our data systems is not
dissimilar to what we started to do, or we continue to do,
with our cities. And Jane Jacobs, who of course,
the famous urbanism critic and an amazing writer and
thinker, talked about one of the biggest problems with
cities at the time and I think one of the continued problems in
cities, was the lack of feedback for the people who live in cities. And similarly, tools like
OpenPaths and Floodwatch are about giving people who live in
data mechanisms for feedback. What we want to do is be
able to give them agency, because I think one of the other
things that it’s like to be living in data today is to
be without agency. So what does it mean
to give people agency? This is a project that my
studio has been working on. We’ve been working with a doctor
in Manchester named Will Dixon and Will is a rheumatologist. He studies chronic pain, mostly, and
he had found from all of his work over the years with pain sufferers
this thing that we kind of know, which was that pain
symptoms can seem to be correlated with
the weather, right? And this is confusing,
we don’t really know why. And in fact, we don’t
really know if it exists. There had been no really good
scientific research about this. So we started this project called
“Cloudy with a chance of pain,” and we recruited, now almost
10,000 UK chronic pain sufferers, and we asked them to record
data around their pain symptoms. And then we built them a website
in which they can do a couple of things, they can first come and
see how their pain symptoms compared to everybody else in the data set. So you can kind of explore
your own experiences and how they compare
to everybody else. We also set up a system
where we mailed people through email weekly visualizations
of what the project was doing, so that we made sure
that they were — felt like they were part of
this project and not just kind of throwing their data
into a black hole where it might never come out again. But the big thing, the biggest thing
that I think we did in this project which I’m the most proud of
is that we produced a tool which allowed the contributors
to the project to pose their own hypotheses. So we call this a citizen
science project, and you’ve probably heard a lot
about citizen science projects. I think that most citizen
science projects fall into one or two categories, they’re
either citizen data collection or they’re citizen labor that they
couldn’t find a grad student to do. But not too many of them
are actual citizen science, like we’re not asking citizens
to actually be scientists, so what we’re doing here is we’re
asking them to do exactly that, to look at their data and make
a hypotheses, make a hypothesis. And so, what we end
up with is we end up with not only 10,000 people
participating in the study, we end up with thousands of
hypotheses, which tell us a, what people are thinking about
their symptoms and their conditions, but also, hey, this is
actually really good to look at those hypotheses. Maybe people are finding something
that we weren’t able to find. And this idea of like putting
people in the driver’s seat, giving them agency, is super,
super, super important to us and to the work that we do. What is it like to live in data? I believe that one of the primary
conditions of living in data is to be overwhelmed, to feel
like you’re underwater. We can’t really even look
at these data systems with our limited human
sensoriums because they’re so big and so complicated, right? And there’s this idea, I do
a lot of data visualization, and there’s this idea that
visualization is a remedy to this, that visualization can
help us with complexity. Well, I would argue that, actually, visualization doesn’t usually
help with what complexity. What it does is, it can get rid
of complexity, but getting rid of complexity is like
sweeping something very scary under a rug, right? We don’t actually want to get rid
of it, we want to be able to engage with it in ways that
help us understand it. And this is something we’ve
always had to do as humans, right, we’ve always had to
engage with complex things. And we’ve had to engage them in ways
that allow us to understand them. And sometimes complex things can
be the hardest because they might at the same time be right
and wrong or what they might at the same time be one
thing and the other. And so what have we developed
as humans over millions of years of evolution to do this? I believe that the mechanism
that we’ve developed is the arts. And we do a lot of projects that try to counter this very appealing
approach to do visualizations. So for example, this is
a map, yeah, it’s a map. This is all of the
hotels in the world. There’s 500,000 hotels
in this data set. And each one of them shows
as a little white dot. If I wanted to explain to
people what hotels look like from space, this is useful. If I want to explain to people
what hotels are actually like, this is not useful at all. And so we were asked by
the Vancouver Art Gallery to develop a piece exactly
about this, to try to talk about the diversity and
richness and complexity of the system of 500,000 hotels. And so instead of turning to visualization, we
turned to narrative. This is that an image
that Jack Kerouac drew to help understand the journey that his characters were
going to take on the road. And so we thought, oh,
this is interesting. What if we did this? What if we took hundreds of
characters from fictional novels and dropped them into
today’s world and told them that they would travel the
same route that they traveled in their books or in the novels,
and they had to stay at a hotel. So we dropped these people
into this imaginary world. It was projected in
real time continuously at the art gallery
for over three months. And during this three-month
period, these imaginary, these fictional characters, they lived within this
database of hotels. So here we’re going to see the
— I skipped the video here. These are the characters
from Lolita. They’re driving along the Eastern
Seaboard, and they’re about to stop at a hotel for the night. They don’t have a lot of money,
so they’re going to search for what’s available
in their budget. In this particular day
when the piece was running, and as you can see from some of
the returns, they don’t have a lot of luck in finding
a beautiful hotel. And so we showed the viewer in the
projections is kind of a combination of photo documentation
from the hotel itself but also from the users. We show them the names
of the hotels. We look up the price. And one of the interesting things
about this it’s a living system. So on one given night, they
might find these hotels. But the next one, there might be
a football game near that town, and there are no hotels available. So they have to drive to the next
town and find a place to stay there. And this idea of using
narrative to frame our way through a complex data set
is something that we’ve done over and over and over again. Because I think it does give us a
chance to investigate these things that are otherwise really
difficult to investigate. I believe that if aliens
were to touch down today, and try to understand our culture, they would look towards these
gigantic data sets to try to understand what was going on. So how would they do that? Well, maybe they would do
it using a similar tactic to what we’ve done here. This is a data set which was
mentioned in the introduction. This is MoMA’s permanent
collection, 127,000 objects. And we were approached by MoMA to
do an artist residency and to try to understand how we could make
an artwork about this data set that would help people
understand it. And that would help people
understand what is the character of this data set and what
are some of the possibilities that are entrenched within it. So what did we do? well, what do you do with a CSV? We load it into a database
and we start querying it, we start asking it questions. And so some of the first
things we did was to try to look for some pattern. And so I wrote a little this
is called a recursive query, which takes a title, and then it
looks for a title which is similar to it and a little bit longer. And then we do it again
and again and again. And we get these kind
of constructions, these poetic constructions,
of the titles of artworks. And I like these, first of
all, because they’re evocative. They’re fun to read. But second because they’re this
kind of strange curatorial. We’re mashing things together. I think of it as a kind of cocktail
party, where these artworks get to meet for the first time. So this one artwork is like,
oh, you’re a trapeze girl, I’m another girl on another planet. And across the wall they’re
like, I’m [inaudible]. Like they have never
been able to meet. But through some commonalities
in the way that these titles are
structured, they’re able to kind of meet and talk to each other. I meant to give this warning
at the beginning of the talk, and that is that there is some
obscene language in this talk, and it’s coming up, not in
the slide but the next one. And so we were very
interested in these structures, particularly the ones around
girl, because there was this kind of entrenched record of kind
of the sexualization of art and this gendering of art
inside of these titles. Here’s every obscene
title in MoMA’s database. And you know, these are also
historically interesting. They came in a period where artists
for a long time had been trying to like challenge expectations
of galleries. They’re like, here’s a painting,
it’s all just black, right. And originally curators
were like whoa, and now they were —
they accepted that. And then so the next step was like,
here’s a painting, it’s called — I’m not going to read it, but
it’s called something like that. And then Mike Hansen who is one
of my collaborators, was sitting and doing some database queries,
and he asked a very simple question. What is the most common
name, first name, of an artist in the MoMA’s database? The most common first name is John. There are 200 artists named
John who have a piece or more than one piece in the
MoMA collection. These are the top four. Here are the top 40. I’m giving you a second. And so it was this piece that really
set the tone for what we were going to do because these are names. And what is the natural
human thing to do with names? That natural human thing to
do with names is to read them. And so we’d worked in the past with this experimental theater group
called the elevator repair service. And we decided that we
would perform the database. And so for 40-minute
performances, we did eight of them. These actors would essentially
read the database in the gallery, but they would do it in a way
that was quite performative. So I’m going to play
you a section of this. Again, there’s some
obscene language. And it’s actually the
performing of this data set that I’m just showing you right now. [ Video ] So we show the names of every artist
in the database during this piece. [ Video ] And part of this was a response to
try to understand how can we act against people’s expectations for what they’re going
to receive from data. But really it was an
opportunity to present the data in what I thought was one of
the most engaging possible ways, to have it read and
to have it performed. And again, these mechanisms,
these things like performance, they’re built for these
types of things, to be able to present complex
issues to people in ways that don’t give them a prescribed
answer, but instead give them a set of questions that they
might leave with. This is Manchester, where we
did that Cloudy with a Chance of Pain project that
I mentioned before. And I want to show you one
other project which talks also about unusual ways to show data. We took some of the data from that
study, actually all the people from Manchester, and we constructed
a 60-foot long essentially bar graph, a histogram, that people
could walk under and see the data and help to sort of guess
whether it’s correlated to certain weather conditions. And what I loved about this
is that, how often do you get to experience data as a
collective, like as a family, or as a class, or as a group? I think that our methods
for examining data typically through a computer or maybe through a phone are really
meant for one person. Whereas the problems
that we’re dealing with that data is his
encoded in or data is involved in really do require us to deal
with those things as a group. And this also gave me a new
metric for a successful project, which is that any successful project
we do involving data has to be able to fit an entire marching
band underneath it. You know, we use the word “public”
a lot when we talk about data. But I would argue that
most of our engagements with data are inherently not public. I said this before,
but they’re public in the way the White House
is public, and not public in the way the library’s public. They’re public in that
we’re told it’s for us, but we can’t really go in. We don’t have the tools
or the access to do so. And so how do we make public
data more like libraries? And I would argue that it is about understanding this
collective experience. Something very interesting
has happened to data over the last few years, which is that our public understanding
of it has changed. And when I give a data talk,
sometimes people come up to me and they say, did you know that
data is collective and is plural and that you should be
saying datum in the singular. And I’m like, I did
know that actually, but the average person
doesn’t know that. We use it in a different
way in society these days. We use it as a mass noun. These are some other mass nouns. So the way that the
average person uses data, as something that is indivisible, which I think is very,
very, very interesting. I had the great opportunity to
work with the poet and photographer and author Teju Cole, and we
built this project called Time of the Game Together. And what Teju did is he
asked his Twitter followers when they were watching the
World Cup to take a photograph of their television and to post it
with the hashtag time of the game. And we pulled thousands and
thousands of these images and we ran them through an algorithm which automatically
centered the television so they were exactly
on top of one another. To try to give this idea that there
was this collective experience happening at the same
time around the World Cup. And Teju, who is 900 times
more eloquent than I am, had these really nice things
to say about the concept. He said that the time of
the game becomes almost like a pilgrimage time,
like a public time which is the chronological
equivalent of public space. And I love this idea of public time. I love this idea of these
mechanisms that can allow people to come together, not necessarily
in the same physical space, but at the same time,
to be able to engage with what was essentially
a very large data set. And the other thing that I love
from this quote is this idea of data having the ability
to allow ourselves to testify to each other’s existence. At my studio, we’ve been
working on one last project — one very new project, and I’m
not going to show you much of it because there’s not
much of it to exist. But what we’re doing in St. Louis,
is we’re building a community space in which people can come and create
very large scale maps together about their lived experience
in St. Louis. And then to use those maps as
instruments to explore civic data. So you may draw a map
of your neighborhood, and then we can project on top
of it demographic information, traffic information, access to
healthcare, so on and so on. And the idea here is like,
how do we balance maps like these ones, you know. This is a typical map
room, a kind of war room. With these ones, these
hand-drawn very beautiful maps. This is a map of somebody’s paper
route when they were a child. How do we balance maps like
this, this is the census dot map of St. Louis which shows
the incredible racial divide that exists within St. Louis. How do we balance maps like this
one, a historic map of blighted, so-called blighted districts in St. Louis during the 1800s,
with maps like this one? This is a hand-drawn map by an early
resident of St. Louis, Gene Taylor, who drew what the city
looked like in 1877. And here I think lies
the central conflict of, how do we deal with data. How do we present the objective
in line with the subjective? How do we present the kind of
clinical, cold clinical magnitude of data, alongside with human story? And I don’t think I
have an answer for you. But I was doing some
research around this project and I stumbled upon this
phrase called geosophy, which is the opposite, maybe not the
opposite, but a pair to geography. Whereas geography is about what
the world “actually looks like,” Geosophy is the study of world as people conceive
of it and imagine it. And I love this idea that we
might move away from using data as this purely objective
thing and instead as a data to understand knowledge from
any or all points of view. JK Wright who came up
with this idea writes about how geosophy is really
this idea of geography but geography for everyone. And in tandem for this, I want us
to consider, what does data look like if data were for everyone? What is it like to live in data? It’s to be used. It’s to be without agency. It’s to be overwhelmed
by complexity. What if we were to turn that around and leave this so-called big data
behind and instead enter an era of data humanism, where we
might first design data systems for the well-being of the
people, from whom it was taken. We might wherever possible
provide mechanisms for feedback. We might honor the complexity of
individual and community realities. We might create real and
functioning data public. You know, everybody in this room in case I think is a data
professional in some weird way. And I think it’s up to
us to build these worlds. You know, I’ve been
thinking about this mostly because I have a 13-month-old son
who has to live in this new world of data that we are all
collectively creating. And I hope that if we consider these
things, and we primarily do this, we put another layer of wisdom
at the bottom before we go about our endeavor
of collecting data, to make sure that these structures
that we build our livable. And if we do that, maybe
the answers are different. What is it like to live in data? To be empowered? To be engaged? To be equal? To be given new ways of seeing? To be given new ways of being? And maybe most importantly,
to be at home. Thank you. [ Applause ]>>This has been a presentation
of the Library of Congress. Visit us at LOC.com.