Inkwell: Authors and Artists

    
Inkwell welcomes Ethan Zuckerman, director of the Center for Civic
Media at MIT, and a principal research scientist at MIT's Media Lab. 
He's here to discuss his book, _Rewire: Digital Cosmopolitans in the
Age of Connection_, published this month by W.W. Norton. Publisher's
Weekly calls _Rewire_ "fascinating and powerful reflection on what it
means to be a citizen of the world in the Internet age."

With Rebecca MacKinnon, Ethan co-founded international blogging
community Global Voices. Global Voices showcases news and opinions from
citizen media in over 150 nations and thirty languages. Ethan's
research focuses on issues of internet freedom, civic engagement
through digital tools and international connections through media. He
blogs at http://ethanzuckerman.com/blog and lives in the Berkshire
Mountains of western Massachusetts.

If you're not a member of the WELL, and want to participate in the
conversation, just send your comment or question to inkwell@well.com.
  

    
Ethan, thanks so much for joining us! Before we get to the book, could
you expand on the evolution of your work from your days with Tripod,
to Geekcorps, Global Voices, and your current work with MIT's Media
Lab?
  

    
Hey Jon - 

Thanks for having me. I'm a long-time fan and follower of The WELL
from afar, and it's great to be here with you.

As you noted, I've been experimenting with internet technology since
the late 1980s. In the early 1990s, I was one of the early people
involved with Tripod.com, one of the web's first social media sites. We
let people host personal homepages and built tools to help them
connect with one another, in a way that would look familiar to people
who use LinkedIn or MySpace. 

For me, one of the most interesting aspects of Tripod was discovering
how many people from around the world used the service. In 1996, after
the US and the UK, we had the most users in Malaysia. After a bit of
research, I figured out that we were hosting many of the supporters of
Malaysian politician Anwar Ibrahim, who wasn't able to get much
coverage in Malaysian media so shared his message online instead. I
thought that was a pretty fascinating way to use digital media for
political speech, and started thinking about how the internet might
change political and financial systems in other parts of the world.

That led to my next two projects. Geekcorps was a volunteer corps that
sent tech people from North America and Western Europe to work with
tech startups in sub-Saharan Africa. That project got me fascinated
with how little much of the world knows about Africa, and I started
Global Voices (globalvoicesonline.org) with Rebecca MacKinnon, an
international aggregator of blogs and citizen media that gives you
perspectives on the world through the eyes, ears and voices of people
in other nations.

These days, I run the Center for Civic Media at MIT's Media Lab. My
students and I study how media and social change influence work
together, both by examining case studies (a recent one was a deep dive
into online and broadcast media around the Trayvon Martin shooting) and
by building new tools, including tools that help users monitor what
online and broadcast media sources cover and don't cover.
  

    
From your perspective, how has the Internet since it started getting
mainstream adoption in the 90s? 
  

    
Jon, the major change has to be ubiquity. In the early 1990s, there
was a sense that people online were "internet people", with a certain
amount of technical knowledge and interest in networks. As we moved
into the late 1990s, you couldn't make any assumptions about why anyone
was online - large percentages of the population in the US and Europe
were online. And then in the last decade, we've seen massive
populations in the rest of the world, including the developing world
come online.

In those early days, there was a sense that you might be able to
strike up a conversation with someone based around a common interest in
networking, online life, etc. As the internet has become more
ubiquitous, I think we're less likely to reach out to random
individuals, and more likely to (re)connect with people we already
know.

Think of it this way - if you're the first person in Tuvalu on
Facebook, all your friends are going to be non-Tuvalans. You'll be the
most cosmopolitan guy on Facebook, with all your contacts in other
countries. But as you introduce your friends and neighbors to Facebook,
suddenly you're mostly friends with your fellow Tuvalans. I think that
experience has happened to many of us who got online before the wave
of mainstream adoption.
  

    
Mentioning "the most cosmopolitan guy" brings us to your subtitle,
"Digital Cosmopolitans in the Age of Connection." Who, and where, are
the digital cosmopolitans? 
  

    
Short external link to this conversation is
http://bit.ly/ethanz-rewire

If you're not a member of the WELL, but you'd like to participate,
send comments or questions to inkwell@well.com and we'll post them
here.
  

    
The digital cosmopolitans are the people who use the internet to make
meaningful and helpful connections to people in other nations,
languages and cultures. The core argument of the book is that it's
helpful to build these connections, because it makes us more likely to
see problems and opportunities that arise from global connection, and
is likely to make us more creative and effective problemsolvers,
leveraging cognitive diversity. But it's hard, because there's a
tendency to use the internet to connect with the people and topics we
already know about.

Two groups of digital cosmopolitans are particularly interesting.
Bridge figures connect two or more disparate communities. They are
people who know two cultures well and can act as a broker between
cultures - think of a Nigerian student in the US explaining his home
country, or a foreign correspondent who's spent a long time in China
and can report on China to a US audience. Xenophiles are not bridges,
but people who want to cross those bridges - people who systematically
seek out encounters across cultures.
  

    
You and others (like Eli Pariser, who writes about a "filter bubble")
express a concern about an echo chamber effect with the Internet, as
you suggest in your first paragraph above. How is that effect baked
into the structure of the Internet as it's evolved? What are some of
its effects? 
  

    
Eli worries about filter bubbles, where algorithms make it difficult
for you to access conflicting opinions. I worry more about echo
chambers, where we surround ourselves with familiar opinions through a
combination of choice and habit and hear few perspectives that
contradict what we already believe.

The internet makes it much easier to find what we want. For the most
part, this is a good thing - we can explore topics in depth that we
never would have bothered to pursue in a library, and we routinely get
lost in discovery. But the ability to choose news and opinion content
can have a powerfully isolating effect. The rise of search was a first
step towards this sort of isolation, making it possible to find the
content we wanted without much wandering. Personalization was a second
step - once we told Google News what we wanted to know about and what
we wanted to ignore, we removed ourselves from a good bit of random
encounter. The personalization and recommendation algorithms Eli
worries about are a third factor, and the rise of social media as a
tool for recommendations is a fourth factor, as it is likely that our
friends have a good bit in common with us, and are likely to recommend
similar sources.

The main fear of echo chambers, articulated by Cass Sunstein, is that
they tend to polarize us. When liberals deliberate with other liberals,
Sunstein finds, they become more extreme in their views - same for
conservatives. He suggests that we need to exit echo chambers to
maintain a functioning democracy.

But I'm worried that there are other effects as well. It's worth
remembering that echo chambers are three dimensional - we tend to get
isolated not just in terms of left and right but in terms of us and
them. We have a strong tendency to see the world from our national and
local point of view, and here, the net effect may not be polarization
so much as blind spots, an inability to see what's important and
unfolding in other parts of the world because we're overfocused on
what's important locally.
  

    
Before the Internet appeared and evolved, when we had a very limited
number of channels for news and information about the world, we
depended on professional curators/editors to tell us what information
was important, and to deliver just that information. Information
delivery has been disintermediated, and at the same time the
information channel has changed from a narrow dripping faucet to a
firehose blast. Attention is limited and we may be focusing more of it
on less authoritative and reliable sources, and also on the very
process of information-gathering. We also increasingly build our
world-view from randomized references in Twitter feeds and Facebook
posts, to name a couple of popular sources. This means that our
worldview is more of a patchwork, often built from less authoritative
sources. 

Quite a mess, wondering how we sort it out? Meanwhile it's tough for
news organizations to fund real journalism, in-depth reporting.

You talk about ways to connect, build bridges, cultivate the emergence
of digital cosmopolitans. How can we make them smart about the world
they're in? How can we preserve the profession of journalism in this
digital era?
  

    
This is both the best of times and the worst of times for journalists.
It's the worst of times because the revenue model for journalism in
the US is broken. We had a very good run with ad supported journalism,
and I think we are now reaching a point where relying on ad revenues to
fund a critical public service is a bad idea. We need to take a close
look at what other democracies do to support public media, and we need
to consider models that don't just serve a well-educated elite that's
willing to subscribe to high quality media, including charitable and
public service models.

But it's also the best of times, in that it's easier than it's ever
been to feature the voices of real people in your stories. I think of a
friend like Andy Carvin who's found new ways to cover events like the
Arab Spring by listening to and amplifying voices through social media.
Whether or not reporters can afford to travel to cover stories, there
are almost always people on the ground reporting on what they see and
experience and they provide richness and detail to stories that was
absent in previous years.

You're right to identify the challenge in curation and verification.
When we can hear from dozens or hundreds of different sources, the
challenge is deciding who we pay attention to and who we ignore. The
heart of my argument in "Rewire" is that we need to think through those
decisions carefully. Right now, we tend to pay disproportionate
attention to wealthy nations (the long-standing bias of US media), to
our friends and family (the bias of social media) and to topics we
already think are important (the bias of search), and not enough
attention to much of the world, to people culturally or politically
distant from us, and to topics that may be emerging in importance.

Not only do we need to find a way to continue paying for investigative
reporting, we also need to find a way to pay curators for the work
they do helping drive us towards unexpected but important stories. Some
of these curators will be professional journalists, while others will
be bridge figures who take on the job of curating what's important in
their local conversations and sharing it with global audiences.
  

    
Can you give some examples of "bridge figures" and how they work?
  

    
As a consequence of the social media-catalyzed movement we call "Arab
Spring," Mubarak of Egypt was out and the Islamist Morsi was elected.
His administration bombed, and military's removed him. The Muslim
Brotherhood appears to have lost all influence there.

A question I and others had about the so-called Arab Spring was about
its longer term effects. You can use social media to catalyze a
revolution, but with revolutions the question is always "what next?"
This reminds me of Steven Johnson's points about clustering vs coping
in the Howard Dean campaign: once you've built a following
(clustering), you have to manage and lead (coping).

How might digital cosmopolitans address this problem in creating more
participatory governance in the Middle East and elsewhere?
  

    
On the topic of bridge figures:

In the book, I talk about my friend Erik Hersman as a notable bridge
figure. Erik is American, but born and raised in East Africa. He lives
in Nairobi and has co-founded a number of key Kenyan tech institutions,
including Ushahidi, iHub, the BRCK and Afrigadget, a fantastic blog
that features hacking and technological innovation in Africa. He's
deeply embedded in the global tech scene, so he understands how to talk
about African innovation in a global context, but he also has a deep
understanding of local culture in Kenya, which gives him access to
sources and ideas that I don't have as an outsider. 

In general, bridge figures have roots in at least two worlds. They're
often people who've in two cultures for a significant piece of their
lives and who understand both cultures deeply, which allows them to
broker ideas across cultural boundaries.

On Egypt:

There's still scholarly debate about the role of social media in
catalyzing protests during the Arab Spring. I am comfortable pointing
to the role of social media in the Tunisian Spring, where activists
documented protests using social media, which when amplified by Al
Jazeera, helped spread participation in demonstrations. (See
http://dspace.mit.edu/openaccess-disseminate/1721.1/78899) In Egypt,
social media likely had a mobilizing effect, but massive offline
mobilization probably played a greater role.

One way or another, we've seen online activism assert more influence
in opposition to ideas than in proposing constructive alternatives.
Think of the protests last year against SOPA/PIPA - activists were
effective in making their outrage at proposed legislation known and
felt, but attempts to create alternative legislation through online
channels have been anemic and unsuccessful thus far.

The Egyptian Spring ousted a government that had lost all public
confidence. In the vacuum that resulted, the Muslim Brotherhood was the
best organized party and was able to claim power. They, too, lost
legitimacy and were ousted. The question now is whether anyone can
claim and hold power or whether Egypt's new tradition of revolution is
so strong that it will challenge anyone's hold. 

I don't have much confidence that the way forward for Egypt comes from
online political deliberation. I suspect it comes from a much longer
process in which people who've been active in ousting governments
become the people involved and engaged in building governments.
  

    
Responding specifically to your question about digital cosmopolitanism
and participatory governance:

I think the tensions in Egypt reflect just how difficult it is to get
full representation in a diverse society. Many of the people who
overthrew Mubarak weren't particularly excited about political Islam,
or the Muslim Brotherhood. A more inclusive political structure would
have given those secular Egyptians more of a voice post-Mubarak, and
might have avoided the protests that ouster Morsi.

One challenge for digital cosmopolitans is to avoid the dualities that
characterize conventional politics. Egypt went from being
pro/con-Mubarak to pro/con-Morsi. In truth, it needs a politics that
recognizes tensions between those who support political islam, those
who want a secular state, those who want strong protections for
minority rights, those who see large or small roles for the military.
It's a multipolar system, not a binary one.  If digital cosmopolitanism
can help us see multiple points of view on key global issues, then
it's a helpful concept for situations like the Egyptian one.

As far as what it advises for people building participatory
governance, we need to start with the assumption that we have a
politics that moves beyond dualism and supports multiple points of
view. Then we need systems that help us understand and value those
multiple perspectives.
  

    
"Then we need systems that help us understand and value those
multiple perspectives."

Ethan, what things are unique to digital platforms that might help in
understanding and valuing multiple perspectives? Is there still more
work to be done on that front, and is anyone doing it?
  

    
That's a great question, Ted.

Just want to add this thought: the problem of needing to be inclusive
and respect minority views is crucial - and why I suggest that we avoid
the term democracy, which is too broadly interpreted as "majority
rule" - he who has the most votes, wins. What we hope for is a
participatory consensus, which is what "democracy" is really about =
everyone who's eligible participates equally. (That question of who's
eligible is important, too.) In that sort of structure, compromise is
inherent.
  

    
Ted, I think there's tons to be done on that point.

There's a strong tendency, at least in US media, to assume two sides
to any story. In part, that's a convenient convention - in the days
when the "fairness doctrine" ruled American airwaves, a station could
be in compliance by making sure a second opinion followed a strongly
expressed opinion. We've lost the requirement but kept the formula - we
see a lot of terrible television journalism that shows two sides
disagreeing on an issue and goes no further.

One potential the internet gives us is a much larger "shelf space" -
we're freed of some of the time constraints that govern broadcast media
and can explore many more perspectives than a simple duality. We might
start by looking into an issue and casting as wide a net as possible,
then curating clusters of responses or proposals. 

Many of the systems I'm considering try to avoid the notion of "sides
of an issue" altogether - they don't try to identify people on the left
and the right and try to feed each the other side's links. Instead,
they identify people as the product of the pages they've read and
recommended and then try to move them towards highly-ranked links
outside their community of practice. We don't need to label one pile
"conservatives" or "liberals" or even "Americans" and "Nigerians" - we
can simply say "Here's content that other people, who've got some
things in common with you, are finding interesting and which you're
very unlikely to find on your own."

I think it's also not purely a technical issue. I suspect that valuing
perspectives has to do with respect for people articulating those
perspectives. One of the reasons I find Twitter such a useful source of
links is that I'm able to follow people like Howard French, a
journalist I respect deeply on issues of China and Africa. When Howard
expresses an opinion or offers a reading suggestion, I take it very
seriously. I'd like to find a way to identify those bridges and
curators who could introduce people to topics they'd like to be more
knowledgeable about.
  

    
Ethan, in your next to last paragraph above, you seem to describe an
existing app - does that app exists, or is your description
aspirational?

Also re identifying bridges and curators: if you had an algorithm that
could identify them, what would the next step be?
  

    
The description is aspirational, Jon - we're in the early stages of
working on a product called "The Weekly Different". The idea behind
Weekly Different is that it will monitor your Twitter behavior,
probably by following the links you post, then match you to people who
have similar, or dissimilar behaviors. We should be able to calculate
degrees of similarity and we'd want to test and see whether people
prefer getting news from people who are significantly dissimilar. Our
hypothesis is that some level of dissimilarity is helpful for discovery
and that the level of discomfort/risk needs to be counterbalanced with
those benefits.

Another possible input into Weekly Different is information on where
you are, and where your followers are, using geoinformation from
Twitter. We might look for someone who's recommending very similar
stories, but comes from a very different place, perhaps helping you
find someone with common interests and a different context.

We might be able to use that information to identify people likely to
be bridge ties in networks - people who have lots of ties to people in
two largely disconnected communities. There's a difference between
bridge ties and bridge figures - bridge figures make a decision to
broker information between cultures. But bridge ties are a great place
to look for bridge figures.

If we had a good algorithm for identifying bridge figures, we could
start building indexes and directories of them. I can imagine a very
different sort of discovery engine that tries to connect you to someone
who's knowledgeable and passionate about a community and likely to be
able to connect with you on the topic.
  

    
Out of curiosity -- do you think it's likely that is something the NSA
can do -- or could learn to do -- with phone and email interaction
records? 
Would that provide insight into people who might leak/inform/spy?
  

    
Ethan, anything you wish you had added to your book, or new
reflections having written it?
  

    
Interesting data point:
http://www.thenetworkthinkers.com/2012/10/2012-political-book-network.html

Valdis Krebs has been analyzing book purchases using data from Amazon,
algorithmically determining political leanings of purchasers and
determining which books were bought on the left vs the right, also
looking at the overlap, if any. The resulting clusters suggest how
polarized we are (or, at least, how polarized Amazon's customers are).

In 2012 he saw a greater number of books being read by both clusters,
thought the polarization is still evident. Could those authors be seen
is bridge figures?
  

    
Gail, I think the NSA likely has enough data that they could do a very
good job of identifying bridge figures in networks - if the level of
government surveillance we're experiencing didn't deeply anger me, I'd
consider asking them if we could license the data. :-)

I don't think they're going to have a ton of luck using this data to
identify spies. I've done some work with very large data sets. It's
hard to build an algorithm that detects something that's very rare -
you're basically trying to train an algorithm based on very few data
points. I expect that whistleblowers and leakers are so rare that it's
difficult to detect characteristics that they have in common and few
others have, though I could certainly be wrong. 
  

    
Ted, I'm about to write something for my blog responding to a
thoughtful, critical review of my book in Book Forum by Astra Taylor.
(http://www.bookforum.com/inprint/020_02/11685) She rightly points out
that I'm far too easy on internet companies and probably place too much
trust in them to take on the problems I'm identifying. Ultimately, I
end up arguing that we need risk to overcome homophily. She points out
that internet companies, which depend on advertising, don't want risk.
Neither does Netflix, which wants to make you happy by giving you
movies it safely predicts you'll like (perhaps at the expense of movies
you'll love.) I wrote the book in the hopes of persuading people in
the tech community to join me in caring about these issues, and I think
I probably wasn't critical enough of the corrosive role of advertising
in making the internet safe and relatively unexciting.

I tried not to pick a fight with Evgeny Morozov and others who are
arguing that people turn too often to technology to solve social
problems - I tried to make the case that technology was often
problematic and doesn't automatically address issues of diversity, but
argue that we need to see technology as political and strive to make it
embody the politics we aspire to. I suspect there's no ducking that
fight, and I probably should have made a stronger case for intervening
using technology. It's not that information technology is the best
solution to all problems - it's that our software is so protean that we
can experiment quickly and at low cost to try to find directions worth
exploring that it seems foolish to dismiss the idea that we might
accomplish change through code.