Ethan Zuckerman’s online home, since 2003
Currently Browsing: Geekery

Crisis Commons, and the challenges of distributed disaster response

Heather Blanchard, Noel Dickover and Andrew Turner from Crisis Commons visited the Berkman Center Tuesday to discuss the rapidly growing technology and crisis response space. Crisis Commons, Andrew tells us, came in part from the recognition that the volunteers who respond to crises aren’t necessarily amateurs. They include first responders, doctors, CEOs.. and lately, they include a lot of software developers.

Recent technology “camps” – Transparency Camp, Government 2.0 Camp – sparked discussion about whether there should be a crisis response camp. Crisis Camp was born in May, 2009 with a two-day event in Washington DC which brought together a variety of civic hackers who wanted to share knowledge around crisis technology and response. The World Bank took notice and ended up hosting the Ignite sessions associated with the camp, giving developers a chance to put ideas for crisis response in front of people who often end up providing funds to rebuild after crises.

The World Bank wasn’t the only large group interested in working with crisis hackers. Google, Yahoo! and Microsoft came together to found the Random Hacks of Kindness event, designed to let programmers “hack for humanity” in marathon sessions around the world.

While these events preceded the earthquake earlier this year in Haiti, that crisis was the seminal event in increasing interest in participating in technology for crisis relief efforts. A crisis camp to respond to the Haitian earthquake involved 400 participants in five cities and pioneered 13 projects. Over time, the crisis camp model spread to Argentina, Chile and New Zealand, with developers focused on building tools for use in Haiti, Chile and Pakistan. Blanchard explained that the events provided space for people who “didn’t want to contribute money – they wanted to do something.”

The camps had some tangible outcomes:
- I’m Okay, a simple application that allows people to easily tell friends and family that they’re okay, in an emergency situation, was developed at Random Hacks of Kindness
- Tradui, an English/Kreyol dictionary for the Android was developed during the Crisis camps
- Crisis camps also developed a better routing protocol to enable point to point wireless between camps in Haiti, writing new drivers in 48 hours that were optimized for the long ping times associated with using WiFi over multi-kilometer distances

Perhaps the most impressive collaboration to come from the Crisis Camps was work on OpenStreetMap for Port au Prince. Using satellite imagery released by the UN, a team created a highly detailed map, leveraging the work of non-programmers to trace roads on the satellite images and diasporans to identify and name landmarks and streets. As the map improved in quality, the volunteers were eventually able to offer routing information for relief trucks, based on road damage that was visible on the satellite imagery. A convoy would request a route for a 4-ton water truck, and volunteers would use their bird’s eye view of the situation – from half a continent away – to suggest the safest route. Ultimately, the government of Haiti requested access to the information, and Crisis Camps provided not only the data, but training in using it.

The conversation turned to the challenges Crisis Camps have faced in making their model work:
- About 1/3rd of the participants are programmers. The others range from the “internet savvy” to those with complementary skill.
- Problems and requirements are often poorly defined
- It’s challenging to match volunteers to projects
- There’s a shortage of sustainable project management and leadership
- Projects often suffer from undocumented requirements and code, few updates on project status.
- Little work focuses on usability, privacy and security.
- Code licensing often isn’t carefully considered, and issues can arise about reusability of code on a licensing basis.
- Projects can be disconnected from what’s needed on the ground
- Disconnection happens in part because relief organizations don’t know what they want and need and are too busy to work with an untested, unproven community
- Volunteer fatigue – the surge of interest after a disaster tends to dissipate within four weeks
- There’s a lack of metrics and performance standards to evaluate project success.

The goal is to move from a Bar Camp/Hackathon model to a model that’s able to build sustainable projects. This means bringing project management into the mix, and asking hard questions like, “Does this project have a customer? Is it filling a well-defined need?” It also means building trust with crisis response organizations and groups like the World Bank and FEMA, who can help bring volunteer technology groups and crisis response groups together.

Crisis Commons see themselves as mediating between three groups: crisis response organizations like the Red Cross; volunteer technology organizations like OpenStreetMap; and private sector companies willing to donate resources. Each group has a set of challenges they face in engaging with these sorts of projects.

Crisis response organizations have a difficult time incorporating informal, ad-hoc citizen organizations into their emergency response plans. There’s a notion in the crisis response space of “operating rogue” if you’re not formally affiliated with an established relief organization… which further marginalizes volunteer tech communities. Many CROs have little tech understanding, which means they aren’t able to make informed decisions about collaboration with technical volunteers. In a very real way, crises are economic opportunities for relief organizations – that reality doesn’t breed resource sharing, which in turn, gets in the way of sharing best practices and lessons learned.

Volunteer tech communities frequently don’t understand the processes used by CROs, and frequently fail to understand that there’s often a good reason for those processes. While VTCs provide tremendous surge capacity that could help CROs, if there’s no good way for CROs to use this surge capacity, it’s a waste of effort on all sides. At the same time, tech communities inevitably suffer from the “CNN effect” – when crises are out of sight, they’re out of mind, and participation slumps. This is particularly challenging for managing long-term projects… and tech communities have massive project management and resource needs. Finally, successful VTCs can find themselves in a situation where they have a conflict of interest – they’re seeking paid work from relief organizations and may choose to cooperate only with those who can support them in the long term.

Private sector partners are usually participating in these projects led by their business development or corporate social responsibility divisions… while cooperation with the other entities often requires technical staff. Response organizations are often the clients of private sector players – the Red Cross is a major customer for information systems – which can create financial conflicts of interest. And working with large technology companies often raises intellectual property challenges, especially around joint development of software.

Meeting with a subset of crisis response organizations, Crisis Commons understands that there’s a need for long term relationships between tech volunteers and relief organizations, tapping the innovation power of these charitably minded geeks. But this requires relief organizations to know what solutions are already out there and what are reasonable requests to make of volunteers. And volunteer organizations need to understand the processes CROs have and how to work within them.

The hope for Crisis Commons is to become an “independent, nonpartisan honest broker” that can “bridge the ecosystem and matrix the resources.” This means “translating requirements of the CRO to the crisis crowd, helping the public understand CRO requirements,” and the reasons behind them. This could lead towards being able to set up a service like “Crisis Turk”, which could allow internet savvy non-programmers to engage in data entry tasks during a crisis.

In the long term, Crisis Commons might emerge as an international forum for standards development and data sharing around crises. Building capacity that could be active between crises, not just during them, they could direct research projects on lessons learned from prior disaster relief, could build a data library and begin preparing operations centers and emergency response teams for future crises. Some scenarios could involve managing physical spaces to encourage cooperation within and between volunteer tech teams and providing support for future innovation through a technology incubation program.


Starting from the shared premise the Crisis Commons founders presented us with – “Anyone can help in a crisis” – the discussion at Berkman focused on the structure Crisis Commons might take. The goal behind a “commons” structure is to be able to be an independent and trusted actor in the long term, to be able to be objective source of tech requirements, and to be able to bring non-market solutions to the table. But the founders realize that this is an inherently competitive space, and that volunteer organizations might find themselves in conflict with professional software developers in providing support to relief organizations, or with relief organizations if volunteer organizations began providing direct support.

It’s also possible that another player in the space could compete with Crisis Commons in this matchmaking role. Red Cross could develop an in-house technology team focused on collaborating with technology volunteers. Google could use the power of their tech resources to provide services directly to relief organizations. A partnership like Random Hacks of Kindness could emerge as the powerful leader in the space. Other volunteer technology organizations – Crisis Mappers, Strong Angel – might see themselves providing this bridging function. FEMA could start a private-public partnership under the NET Guard program. What’s the sweet spot for Crisis Commons?

One of our participants suggested that Crisis Commons could be valuable as a developer of standards, working to train the broader community about the importance of standards, and on the challenge of defining problems where solutions would benefit a broad community.

Another participant, who’d been involved with several Crisis Camp events worried that “the apps, while neat, never really made it into the field,” suggesting that the problems raised are real, not theoretical. It’s genuinely very difficult for tech volunteers to know what problems to work on… and hard for relief organizations under tremendous pressure to learn how to use these new tools.

This, I pointed out, is the problem that could prove most challenging for Crisis Commons in the long term. When crises arise, people want to help… but it’s critical that their help actually be… helpful. Clay Shirky told the story of his student, Jorge Just, who’s worked closely with UNICEF to develop RapidFTR, a family tracking and reunification tool. It’s been a long, engaged process with enormous amounts of time needed for the parties to understand each other’s needs and working methods… and it’s easy to understand why it might be difficult to convince volunteers to participate to this depth in a project.

I offered an observation from my time working on Geekcorps – I meet a lot of geeks who are convinced that the tech they’re most interested in – XML microformats, mesh wireless, cryptographic voting protocols – are precisely what the world needs to solve some pressing crisis. Occasionally, they’re right. Often, they’re more attached to their tech of choice than to addressing the crisis in question.

As such, the toughest job is defining problems and matching geeks to problems. At Geekcorps, it often took six months to design a volunteer assignment, and a talented tech person needed to meet several times with a tech firm to understand needs, brainstorm projects and create a scope of work, so we could recruit the right volunteer. While that model was expensive – and ultimately, made Geekcorps unsustainable – I think aspects of it could help Crisis Commons find a place in the world.

I ended up suggesting that Crisis Commons act as:
- a consultant to relief organizations, helping them define their technical needs, understand what was already available commercially and non-commercially and to frame needs to volunteer communities who could assist them
- a matchmaking service that connected volunteer orgs to short term and long term tech needs, preferably ones that had been clearly defined through a collaborative process
- a repository for best practices, collective knowledge about what works in this collaboration.

Unclear that this is the right solution for Crisis Commons or the road they’ll follow, but I came away with a strong sense that they are wrestling with the right questions in figuring out how to be most effective in this space. Very much looking forward to discovering what they come up with.


Counting International Connections on Facebook

My friend Onnik Krikorian has become a Facebook evangelist. Onnik, a Brit of Armenian descent, living in Armenia, is the Global Voices editor for the Caucuses, which means he’s responsible for rounding up blogs from Armenia, Georgia, Azerbaijan as well as parts of Turkey and Russia. This task is seriously complicated by the long-term tensions in the region. Armenia and Azerbaijan are partisans in a “frozen” conflict – the Nagorno-Karabakh war, which lasted from 1988 – 1994, and remains largely unresolved.

It’s taken Onnik years to build up relationships with bloggers in Azerbaijan, relationships he needs to accurately cover the region. Azeri bloggers are often suspicious of his motives for connecting and wonder whether he’ll cover their thinking and writing fairly. But Onnik tells me that Facebook has emerged as a key space where Azeri and Armenians can interact. “There are no neutral spaces in the real world where we can get to know each other. Facebook provides that space online, and it’s allowing friendships to form that probably couldn’t happen in the physical world.” (Onnik documents some of the conversations taking place between Azeri and Armenian bloggers in a recent post on Global Voices.)

Picture 1
Graph from the front page of peace.facebook.com

Onnik was talking about his love of Facebook at an event hosted by the US Institute for Peace, where I and colleagues at George Washington University and Columbia were presenting research we’d carried out on the use of social media in conflict situations. Onnik’s hopes for Facebook as a platform for peace were echoed by Adam Conner of Facebook, who showed the company’s new site, Peace on Facebook. The site documents friendships formed between people usually separated by geography, religion or politics. Some of the statistics seem clearly like good news – 29,651 friendships between Indians and Pakistanis per day. Others are rather dispiriting – 974 Muslim/Jewish connections in the past 24 hours.

I’m a data junkie, and there’s little more frustrating to me than an incomplete data set. Basically, by showing us a very small portion of the nation to nation social graph, Facebook is hinting that the whole graph is available: not just how many friendships Indian Facebook users form with Pakistani users, but how many they form with Americans, Canadians, Chinese, other Indians, etc. Obviously, this is info I’m interested in – I’ve been building a critique that argues that usage of social networking tools to build connections between people in the same country vastly outpaces use of these tools to cross national, cultural and religious borders.

Without the whole data set, it’s hard to know whether these numbers are encouraging or not. Are 29,651 Indian/Pakistani connections a lot? Or very few, in proportion to how many connections Indians and Pakistanis make on Facebook in total? In other words, we’ve got the numerator, but not the denominator – if we had a picture of how many connections Indians and Pakistanis make per day, we might have a better sense for whether this is an encouraging or discouraging number.

I made a first pass at this question this morning, using data I was able to obtain online. Facebook tells us that the average user has 130 friends – a number that might be out of date, as the same statistics page lists “over 400 million users”, not the half billion currently being celebrated in the media. (Ideally, we’d like to know how many new friends are added per day so we can compare apples to apples, but you got to war with the data you have…)

We also need a sense for how many Facebook users there are per country. Here, we turn to Nick Burcher who publishes tables of Facebook users per country on a regular basis. Nick tells readers that the data is from Facebook, and the Guardian appears to trust his accounts enough to feature those stats on their technology blog. They are, alas, incomplete – Burcher published stats for the 30 countries with the largest number of Facebook users, and revealed a few more countries in the comments thread on the post.

Because we don’t have data for Pakistan, we can’t answer the India/Pakistan question. But we can offer some analysis for Israel/Palestine and Greece/Turkey.

Facebook for Peace tells us that there are 15,747 connections between Israelis and Palestinians for the past 24 hours. The term “connection” is not clearly defined on the site – it’s not clear whether a reciprocated friendship is 1 connection or 2 – because I’m going to count the number of Israeli friends and Palestinian friends, it makes sense to count a reciprocal friendship as two connections. (If Facebook is counting differently than I am, my numbers are going to be half what they should be.)

3,006,460 Israelis are Facebook users… a pretty remarkable number, as it represents 39.92% of the total population of the nation and roughly 57% of the country’s 5.3 million internet users. There are very few Palestinian internet users – 84,240, or 2.24% of the population… This mostly reflects how few Palestinians are online, as Facebook is used by 21% of Palestine’s 400,000 internet users.

At 3,090,700 Palestinian and Israeli Facebook users, we should see almost 402 million friendships involving an Israeli or a Palestinian. If we extrapolate from 15,747 friendships a day to 5.7 million a year, we’re looking at Israeli/Palestinian friendships representing 1.43% of friendships in the Israeli/Palestinian space… with all sorts of caveats. (The biggest is that the use of a year-long interval to calculate total friendships is totally arbitrary and probably not supportable. If you’ve got better data or a suggestion for a better estimation method, please don’t hesitate to speak up.)

We get very different results from looking at Greece and Turkey. 2,838,700 Greeks are Facebook members (25.11% of the national population), while 22,552,540 Turks (31.08% of the population) are. That’s roughly 3.3 billion friendships projected, and our year-long approximation finds us just over 4 million Greek/Turkish connections. That suggests that only 0.12% of friendships in the pool are Turkish/Greek friendships.

What explains the disparity between these numbers? While there’s certainly a long history of tension between Greece and Turkey, the last major military confrontation between the nations ended in 1922. Israel and Palestine, on the other hand, are involved with an active conflict and Israel’s recent incursion into Gaza ended a few months ago. What gives?

It’s possible that the numerous efforts designed to build friendship between Israeli and Palestinian youth are having an impact, much as Onnik’s work in Armenia and Azerbaijan is showing positive results. But there’s another possibility – 20% of the Israeli population are Arab citizen of Israel, and the majority of this set is of Palestinian origin. It’s certainly possible that the high percentage of Israeli/Palestinian friendship includes a large set of friendships between people of Palestinian origin in Israel and Palestinians… indeed, given the difficulty for both populations in meeting in physical space, we’d expect to see increased use of the internet as a meeting space to compensate for the difficulties of meeting in the physical world. This could be a factor in explaining India/Pakistan friendships as well, as well as Albanian/Serbian friendships, as the emergence of new nations through partition and conflict left groups united by cultures, separated by borders.

My goal in this post isn’t to belittle the power of Facebook for providing a border-transcending space where friendships can be built – Onnik’s story makes it clear that Facebook is a real and powerful tool for good, at least in the Armenian/Azeri space. But I continue to think that we overestimate how many of our online contacts cross borders and underestimate how often these tools are used to reinforce local friendships. I’d invite friends at Facebook to correct my numbers or my math… and mention that we could do a much better job of answering these questions if Facebook would release a data set that shows us all the cross-national connections made on the service.

—-

Ross Perez has created some great interactive maps that visualize the adoption of Facebook around the world, using Burcher’s data – worth your time.


Democrats, Republicans and Appropriators

I had the good fortune to catch a small part of a conference at Harvard yesterday on text analysis. Good fortune, because I was there long enough to hear Justin Grimmer‘s talk on his dissertation, Measuring Reputation Outside Congress. Grimmer is interested in an important – and tough to answer – question: how responsive are the people we elect to their constituents?

We could look for ways to answer this question by studying the voting record of legislators (qualitatively or quantitatively), examining their work in Washington (through Congressional literature) or through examining their communications with constituents at home. This latter set of questions is referred to as the “Home Style” of a politician, following the work of Richard Fenno (1978).

Home style tells us something about politicians that their voting record often doesn’t, Grimmer tells us. He invites us to compare Senators Jeff Sessions (R-Alabama) and Richard Shelby (also R-Alabama). If we consider them simply in terms of their voting behavior, they look nearly identical – they vote together the vast majority of the time and both can be described, in voting terms, as conservative republicans.

But anyone who knows Alabama politics will tell you that Sessions and Shelby are vastly diferent guys. Grimmer characterizes Session as “an intense policy guy” who will bore you to tears with incredibly long, thorough explanations of issues when all you wanted was a photo with him. Shelby, on the other hand, is all about bringing home the bacon… and there are Shelby Halls at two Alabama universities to prove it.

Evidence suggests that representational style – policy versus pork, heavy versus light communicators – cuts across party lines. And it’s likely that politicians have diverse, stable, nonpartisan home styles. If we can find ways to characterize these differences – Grimmer proposes studying the difference in communications with constituents that claim credit and those that discuss policy – we have the opportunity to compare across senators, and connect these differences to what senators do within the institutions of power.

When Fenno studied the “home style” of politicians in 1978, he engaged in “soaking and poking” – intense participant observation, which involved following 18 members of Congress over 8 years. This method, Grimmer observes, is expensive, underrepresentative (and really hard to replicate as a graduate student.) Instead, we might study texts produced by senators. One candidate is newspaper articles… but editorial bias makes it hard to use editorials as representative of senatorial communications. We might use the constituent newsletters produced by Senate offices… but they’re sent using the constitutional Franking privleges and are very hard to get hold of.

Instead, Grimmer has been studying the press releases that senate offices produce – over 64,000 in all. The average senator issues 212 press releases per year, and while the quantity produced has a wide range (some produce only a few dozen, while Hillary Clinton’s senate office produced over a thousand a year), there’s no strong correlation between political party and usage of the tool.

After collecting the releases, Grimmer used machine learning techniques to separate transcripts of floor statements (which are usually released as press releases) from pure press releases, which let him study how a senator chooses to speak to her constituents. Once that sorting has taken place, the task is pretty simple – determine the topic of a press release. This is simplified by the fact that congressional aides try hard to ensure that press releases are on a single topic.

Grimmer’s work clusters senators by the topics discussed in their press releases. His research reveals four basic clusters:

- Senate Statespersons. These folks speak like they’re running for president… and they may well be. Their releases discuss the Iraq war, intelligence issues, international relations and budget issues. John McCain’s office communicates this way.

- Domestic policy. These senators are also policy wonks, but their focus is domestic – the environment, gas prices, DHS, and consumer safety.

- Pork and policy – Communication from these senators includes discussions of water rights grants, but also has serious discussion of health and education policy. Sometimes this is because the office simply issues lots and lots of releases – (former) Senator Clinton’s office fits in this camp.

- Appropriators – These guys communicate about the grants they’ve won – fire grants, airport grants, money for universities, and for police departments.

As well as clustering press releases based on topic, Grimmer’s work considers another metric – how often a press release claims credit for an appropriation. There turns out to be a vast spectrum, ranging from John McCain, who basically only issues statements about policy, and a guy like Mike DeWine, an Ohio Republican, where virtually every press release claims credit for an appropriation. There’s a very strong correlation between the topic clusters in releases and the percentage of releases claiming credit. (That’s at least in part because claiming credit is one of the topic clusters – you’re correlating between, in part, the same factor. Interesting nevertheless.)

What’s most interesting is that this classification – either by type of politician or by their place on the credit spectrum – is tightly correlated to their voting behavior on a particular issue: votes on appropriations rules, or as Grimmer puts it, “How do legislators self-regulate the porkbarrel”. These votes aren’t partisan – the late Ted Kennedy voted with Richard Shelby on these sorts of votes, which suggests truth to the truism that there are three parties in Congress: Democrats, Republicans and Appropriators. In other words, the way a Senator communicates with constituents is strongly predictive of their legislative behavior, specifically on how they allocate funds.

I thought this was excellent stuff – I hadn’t seen someone take a large database of political communications and subject it to automated analysis, and I thought the demonstration of this “third party” was particularly compelling.


ROFLCon: From Weird to Wide

An audio version of danah and my keynote is now available for download online. I recommend a background of lolcats – preferably multilingual ones – as you listen.


I gave a dozen public talks last month, and it’s possible that ROFLCon was the most intimidating of the bunch. I was asked by Tim Hwang, internet researcher (and Berkman Center affiliate) co-founder of The Awesome Foundation and of ROFLCon, to kick off the event by co-keynoting with (dear friend) danah boyd. danah actually works in the deep swamps of contemporary internet culture, so ROFLCon – a conference that takes both a loving and scholarly look at the phenomenon of internet memes – is close to home turf for her. I, on the other hand, tend to study things like the impact of cellphones in political organizing in the developing world, and wondered if there was any possible way to connect the sort of issues I work on with a conference that featured Mahir Cagri (of I Kiss You fame), the owner and videographer of Keyboard Cat and the author of Garfield Minus Garfield.

Turns out I was underestimating ROFLCon. Yes, there were panels where the main question seemed to be, “What’s it like to be a microcelebrity”… which may have included the panel danah and I moderated. And yes, there’s nothing to make you feel old and decrepit like walking into a panel where you don’t know a single one of the internet memes being celebrated. (No, I’d never heard of cornify. No, my world has not been substantially broadened by listening to their founder, wearing a unicorn mask, discuss vampires.) On the other hand, the panel on race – I can haz dream? – was one of the best conference panels I’ve ever attended. (If any network execs are reading this blog, let me just point out that a late night show based around Baratunde Thurston and Christian Lander would kill.) And many of the people at the conference seemed to be deeply engaged in the sorts of issues danah and I were talking about – Who creates internet culture? Whose voices are amplified and whose aren’t? What happens when marginal, weird cultures become mainstream?

Alex Leavitt did an excellent job of liveblogging our talks. I thought I’d post my notes and some of my slides as well – the full slide deck is online, though isn’t real useful without accompanying notes, which follow below.


It’s not easy being an academic at a conference like ROFLCon. The stars are the folks who’ve done something wonderful, weird, unforgetable, or so wonderfully weird it’s unforgetable. Those of us who are trying to make observations about the field feel a little like musicologists studying Bach – we can study his compositions exhaustively, but we’re acutely aware that we’re not going to write a mighty fugue. No matter how much I might study internet memes, I know I’m never going to accomplish something as majestic as keyboard cat… and I have to live with that truth every day of my life.

Unlike danah who can actually tell you something about internet culture, I study information in the developing world. Basically, I’m interested in the question of whether the internet, mobile phones and community radio can make people healthier, wealthier and more free.

slide4.004

If you work in this field for very long, you’ll end up realizing that the basic question behind development economics is “Why are some people rich and other people poor?” There are better and worse answers to these questions. Some of the smartest answers focus on which parts of the world had animals and plants that were easily domesticated and which had endemic diseases. Other smart answers look at the ways in which colonialism held back development or look at the problems of bad governance and persistent conflict. Bad answers to the questions focus on the idea that some people are inherently, biologically smarter than others. This idea – “scientific racism” – surfaces throughout history, as the basis for eugenics and more recently in psuedo-scientific analyses of IQ scores.

If you’d like to understand just what a stinking heap of bullshit scientific racism theories are, I recommend spending some time in very poor nations. You’ll discover that many of the people you meet display extraordinary creativity as they navigate the challenges of everday survival. And you’ll start learning about people like William Kamkwamba, whose near death from famine in Malawi didn’t prevent him from building a fiendishly ingenious power-generating windmill from an old bicycle and some recycled PVC pipe.

My time in the developing world suggests to me that intelligence, creativity and humor are evenly distributed throughout the world. People’s ability to express their intelligence, creativity and humor – and our ability to encounter said traits – are heavily geographically constrained, but the basic distribution is near constant.

slide7.007

All of which leads us to the question at hand today: Daddy, where do memes come from? I suspect Drew will be asking me this question any day now, due to Rachel and my egregious tendency to misuse Cafe Press and the fact that we gave him the middle name “Wynn” in part so we could title his blog “For the Wynn“. In answering these questions, I find that I’m usually referring to Randall Munroe’s brilliant
Online Communities map, and to the fertile equatorial regions that extend from the Gulf of YouTube through the Ocean of Subculture. Within this region, there are areas whose soils – turned black with the charring of endless flamewars – are especially fertile for the cultivation of new memes. (sup, /b/?)

slide10.010

I’m interested in mapping memes in a different way. Here’s a quick and dirty map of internet memes extracted from Know Your Meme. Yes, the US and Japan dominate global memetics (or, at least, they do based on the site, which has its own – recognized, now being addressed – cultural biases). But there’s a huge number of memes coming from almost all corners of the globe.

In development economics, we pay special attention to the so-called BRIC countries – Brazil, Russia, India, China – who we expect to become increasingly important over the next few decades due to their large populations, natural resources and rates of economic growth. And so we shouldn’t be surprised to find distinctly regional memes emerging from each of these countries – I offer as a gallery of superheroes Brother Sharp from China, Golimar from India, Glazastik from Russia and the legion that is Tenso from Brazil. You may not know who these viral wonders are, but the people who live in these rapidly developing nations do.

Assume I’m right and that creativity has a near-constant distribution. Assume also that access to the internet continues its explosive spread. The inescapable conclusion is that the next wave of internet memes is going to come from the developing world.

It’s already happening – I just watched the first major Kenyan internet meme come to life. The Nairobi-based band called “Just a Band” released a video for a song called “Ha-He” off their new album. The video’s absurdly good – it’s shot by the guys in the band, and it introduces a new superhero: Makmende.

Actually, “Makmende Amerudi” means “Makmende has returned”… “Makmende” was what you called a kid in the neighborhood in Kenyan in the 1990s who wanted to be Bruce Lee. I heard it and assumed that it was a sheng word – “sheng” is the blend of Swahili and English that’s Kenya’s unofficial national language – turns out that “Makmende” is what happens when Kenyans say “Go ahead, make my day”.

So Makmende kicks the ass of all comers in this video, gets the girl… who he promptly ignores, and spouts some incomprehensible but pithy aphorisms. This video went crazy in the Kenyan blogosphere – which is an extremely creative space – and we started seeing Makmende magazine covers, a 10,000 shilling note and lots of video remixes.

Above, we see a local television reporter come to a rapid and bad end when he has the misfortune of finding Makmende’s house… in sort of a Nairobi version of the Blair Witch project. And yes, Hitler’s upset about Makmende as well… But the best stuff actually has pretty low production values – it’s the website aggregating the sort of Makmende one-liners that shot across Twitter for a week or so after the video became popular. Sure, lots of the content here could have appeared on Chuck Norris Facts, but much of what’s there is indigenous to Kenya, and may not make sense if you’re not Kenyan.

Makmende’s so badass that he raises two philosophical questions for me. The first is, “Who gets to decide what’s a meme?”

slide21.021

Brilliant and funny lexicographer Erin McKean tells us that new worlds enter the language because people love them enough to use them. Lexicographers aren’t the bouncers at the language club; they’re anthropologists, discovering and documenting how language gets used. This is clearly how memes work as well – if people adopt it, love it and transform, it’s a meme… and what anyone else says doesn’t matter.

But it sure as hell helps if it ends up in Wikipedia. Getting Makmende into Wikipedia was one of the first things Kenyans tried to do… and getting things into Wikipedia is a lot harder than it used to be. The article was deleted a couple of times before the authors realized that they needed to make the case that Makmende was Kenya’s first major internet meme, which made it notable. It hasn’t made it into Know Your Meme yet – it was summarily deadpooled when last submitted.

My hope is that all of us who are interested in internet culture can be anthropologists, not bouncers. Yes, not everything that gets posted online is worthy of our study and amplification… but it’s worth keeping in mind that we sometimes don’t understand the unfamiliar at first and would find it intensely cool if we took a bit more time to try and understand it.

My second question is: “Who gets to play along with an internet meme?” On the one hand, there’s not much preventing you from adding some Makmende facts to the mix. On the other hand, a lot of the funny stuff already posted doesn’t make much sense unless you know the language and the culture. “Makmende hangs his clothes on a Safaricom line” only is funny if you know that Safaricom is Kenya’s largest mobile phone company and doesn’t have any traditional phone lines.

My sense is that most memes don’t cross between cultures because we don’t understand the language, don’t understand the references or weren’t paying attention to that corner of the internet to start with. Those that do tend to be funny in a way that’s independent of language. The Back Dorm Boys are pretty funny, and it’s not hard to figure out how to join in the fun.

This question parallels one that internet scholars are spending a lot of time on: Do we have one internet or many? When a country like China heavily censors their internet and encourages the growth of a parallel internet, do we hit a point where it just doesn’t make sense to talk about “the internet” anymore? Perhaps we’ve got to talk about internets, and how they interconnect. And if 340 million Chinese internet users look mostly at Chinese sites, laugh at Chinese memes, maybe it makes sense that the Chinese internet will eventually run on its own protocols, which might make it easier to censor or control. Go far enough down this road and you can imagine diverging internets, each trying to best meet the needs of their users, and no longer having a world where we readily peer into each other’s internets.

slide 26.026

If we care about a single, united internet, it is imperative that we develop, discover and disseminate internet memes that we can laugh at together. When governments censor political sites on the internet, they alienate the small portion of their populations who already identify as politically dissident – and they can make the case that they’re protecting their citizens from terrorism or incitement to violence or pornography. But when they block our access to videos of cats flushing toilets, we see them for the heavy-handed bullies that they are. The cute cats serve as cover traffic for more serious political speech – so long as chinese users want to laugh at our cat videos, we’re encouraging people to circumvent censorship and potentially encounter all sorts of stuff on YouTube.

The Chinese have developed cute cat technology. Even a cursory glance at Youku shows that the once apparently insurmountable cat gap has been thoroughly bridged. And not just simple cute cats – Youku features cats flushing toilets! And not just western style toilets – squat toilets as well! If we accept my assertion that it’s politically critical for us to LOL together, we need not just to be studying Chinese net memes – we need to develop memes we can LOL at across cultures.

When we cross cultural borders in internet memespace, we’re usually laughing at someone else. Engrish, funny though it is, is basically the act of laughing at someone for failing to speak your (absurdly complex and irregular) mother tongue. I’m deeply impressed with people like Mahir a?r? who managed to turn the experience of being laughed at by the entire internet into laughing along with the joke. It takes an unusual personality to pull this off – I’m not sure that laughing at and inviting folks to laugh along is always the best way to go.

I’d rather take the example of Matt Harding, the video game developer who spent years travelling the world, dancing badly. After the success of his first video, Matt discovered that the piece of music he’d used – “Sweet Lullaby” by Deep Forest – had a problematic history. The very short version – the French musicians behind Deep Forest used a lullaby from the Solomon Islands to record their hit song, without seeking permission from the woman who sang the song and over the explicit objections of the musicologist who recorded it. Worse, they presented it in such a way that most listeners thought it came from central Africa, not from the south Pacific.

Matt could have dismissed this story as an ugly footnote to his adventures with internet fame. To his great credit, he didn’t. Instead, he went to Auki, a small town in the Solomon Islands, to interview a nephew of Afunakwa, the woman who’d recorded the original song. It was his way of apologizing for the complex past of the song, and his way of using the weirdness of internet fame to make his world – and all those of us who’ve watched the video – a little wider.

My conclusions?
- We can go from weird to wide, as Matt did, using the strange and quirky corners of the internet to prod us into curiosity
- It’s worth asking ourselves if we’re laughing at, or laughing with. And if we don’t like the answer, perhaps we need to change our behavior.
- Anthropologists are cooler than bouncers.
- If we don’t laugh at Chinese internet memes – the first step towards getting Chinese users to laugh at global memes – the censors win.
- “Erinaceous” is a totally awesome word.


Highlights of presenting the talk included:

- Co-presenting with danah, which encouraged significantly sillier behavior than I generally engage in when on stage. I’d like to believe that I would always be willing to crouch behind a podium wearing a fluffy red hat before delivering a keynote… but it’s just not true. Add danah to the mix and it suddenly is.

- Matt Harding jumping up when his name was mentioned and dancing in the audience. I’m thankful that he came on stage after the talk to introduce himself and apologize if I freaked him out by spontaneously hugging him. I just think he’s wicked cool and deserves recognition for using the internet to show us (one facet of) how wide and wonderful the world can be.

- Meeting Mahir, who turns out to be utterly lovely in person. Yes, he immediately started filming our meeting via flip video and digital camera, and yes, he did invite me, my wife and infant son to visit him in Izmir… but I got the sense that it wasn’t in any way an act, just his particular version of friendliness. It felt more wonderful than weird.

- Talking with the guys from Know Your Meme, who are working really hard to ensure that their site is global and inclusive, and who are trying to take some pages from the Global Voices playbook, recruiting local editors who understand memes in their corners of the world. I’ve got high hopes of a Makmende article in development soon, and hope perhaps for a GV/KYM alliance where we source and research global memes.

In other words, I had a blast. Thanks to everyone involved and hope you had as much fun as I did.


Eric Rescorla – How paranoid should we be?

Eric Rescorla is a man who speaks frankly about internet security. You might have guessed that from the name of his consultancy – RTFM – an acronym politely translated as “read the friendly manual”… except that it’s usually not politely translated. He’s the co-designer of secure HTTP, and now works on security issues with Skype.

In a conversation about cyberwarfare at Princeton, his assertion – “The internet is still too secure” – raises a few eyebrows. But his key message – “things could be a lot worse” – is a useful counterpoint to the rhetoric of cyberwarfare that Dr. Lin explores, and helps show the tensions between the computer security community and the defense community (not to mention with the human rights community.)

Rescorla asserts the following:

We have nearly unbreakable crypto primitives.
AES resists all practical attacks, RSA isn’t quite as strong as we’d like – but there is good new stuff in the pipeline, and there’s been significant progress made on addressing collisions in SHA-1. “There is no serious concern we’re going to run out of crypto any time soon.” The problem instead is getting the new stuff into use – we’re not using the strong stuff that we already have.

We think we know how to build secure protocols.
There’s been no basic change in security design since 2000. We’ve got good protocols for data object security (S/MIME, PGP), for stream security (SSL/TLS, SSH), for packet security (IPsec, STLS) and for authentication (Kerberos). Our work isn’t on deploying new protocols – it’s on addressing flaws in existing implementations – updating to newer hash functions after MD5/SHA-1 attacks, for instance. And it’s on gluing together existing protocols, none of which have really changed since 2004.

We can’t build systems that are reliably secure.
Good implementations are hard. We recently saw an attack on OpenSSL where a single “record of death” can crash the whole system. Debian managed to break their psuedorandom number generator, which meant that Debian keys were highly predictable and crackable with a table of only 32,000 keys. This probably affected 1% of internet connected machines.

This is just the security critical code. As Steven Bellovin at Columbia has pointed out, “Software has bugs. Security software has security relevant bugs.” And we’re bad at finding these bugs – audits are time consuming and there’s no evidence that they significantly affect the discovery rate for bugs. And affected implementations are slow to go away.

User practice is appaling.

Users are careless. They install random software from untrusted sources. They ignore our carefully constructed messages designed to prevent man in the middle attacks. And crypto doesn’t do much for you if your system’s been compromised.

Things could be a lot worse.

Despite all this, things aren’t so bad. Why is computer security so poor despite twenty years of work on the topic? We’ve been working on personal security for 10,000 years – why aren’t human beings invulnerable? Security people always think about the worst case scenario, but the actual attacks we experience are fairly primitive.

The Debian PRNG bug is about as bad as it gets, but there was no evidence of practical attacks in the field. You can mount a DOS on the whole internet pretty easily – publish bogus BGP routes, as Pakistani ISPs did while trying to block YouTube. As a practical matter, we can almost always get to YouTube. The internet shouldn’t work, but it does. Perhaps the question we need to be asking ourselves is “What’s a rational level of paranoia?”


« Previous Entries Next Entries »

Powered by WordPress | Designed by Elegant Themes