Internet & Democracy Blog

As you’ve likely discovered from personal experience, timing is everything. And so the Internet & Democracy team is especially pleased to announce that just in time for this Sunday’s Russian presidential election, Karina Alexanyan, Vladimir Barash, Robert Faris, Urs Gasser, John Kelly, John Palfrey, Hal Roberts, and I are releasing a new paper that assesses the relationship between the Russian Internet and Russian political and social life: “Exploring Russian Cyberspace: Digitally-Mediated Collective Action and the Networked Public Sphere.” This work was made possible thanks to the generous support of the John D. & Catherine T. MacArthur Foundation.

In English and in Russian (thanks to the translation expertise of Gregory Asmolov), here is the full abstract for the paper:

**

This paper summarizes the major findings of a three-year research project to investigate the Internet’s impact on Russian politics, media and society. We employed multiple methods to study online activity: the mapping and study of the structure, communities and content of the blogosphere; an analogous mapping and study of Twitter; content analysis of different media sources using automated and human-based evaluation approaches; and a survey of bloggers; augmented by infra- structure mapping, interviews and background research. We find the emergence of a vibrant and diverse networked public sphere that constitutes an independent alternative to the more tightly controlled offline media and political space, as well as the growing use of digital platforms in social mobilization and civic action. Despite various indirect efforts to shape cyberspace into an environment that is friendlier towards the government, we find that the Russian Internet remains generally open and free, although the current degree of Internet freedom is in no way a prediction of the future of this contested space.

****

*Аннотация*

В данной статье представлены основные результаты трехлетнего проекта, целью которого было изучить влияние Интернета на российскую политику, средства массовой информации и общество. Для исследования общения и деятельности пользователей интернета мы использовали различные методы: отображение и исследование структуры, сообществ и содержания блогосферы и контента в Твиттере; опрос блогеров, контент-анализ различных средств массовой информации: как с помощью автоматизированных методов, так и с помощью экспертов.

Мы открыли существование живого и чрезвычайно разнообразного публичного пространства, которое представляет собой альтернативу более контролируемым официальным средствам массовой информации. Мы считаем возможным говорить об электронных платформах, на основе которых происходит социальная мобилизация гражданских действий. Несмотря на различные попытки превратить кибер-пространство в пространство, поддерживающее правительство, наше исследование показывает, что Российский Интернет все еще остается свободным и открытым. Тем не менее, несмотря на существующую свободу Рунета, очень сложно делать какие-то предсказания относительно его будущего.

**

Please note that we are working to provide a full translation in the future.
In the meantime, we welcome your comments at the Internet & Democracy Blog.

If we’ve whetted your appetite for more research on all things related to the role of the Internet in Russian society, we welcome you to take a fresh look at our October 2010 Russian blog paper, Public Discourse in the Russian Blogosphere: Mapping RuNet Politics and Mobilization.

Also, please keep an eye on our paper series page for future publications over the coming months, and check out the same site for a short description of each paper we’re planning to release.

Enjoy!

By Hal Roberts and Bruce Etling

Over the course of the weekend, a seemingly coordinated distributed denial of service (DDoS) attack flooded a number of the leading Russian independent media, election monitoring and blogging sites. Many users and content publishers, including the Global Voices RuNet project, have been reporting the attacks against sites including LiveJournal, Echo of Moscow, Novaya Gazeta, New Times, Bolshoi Gorod, Golos.org, ikso.org, ridus.ru, zaks.ru, and the online ‘map of violations’ created by the election monitoring group Golos (which has been the target since last week of a government campaign against ‘outside’ influence on the election (they are funded by US and European groups). LiveJournal, which is the biggest blog host in Russia and according to our research is the blogging platform where Russian political discourse is most prevalent, was also attacked. There are continued reports of LiveJournal’s inaccessibility inside Russia over the last couple days, and shorter term attacks on sites such as levada.ru, the Web site of the leading independent polling firm in Russia.

DDoS and other sorts of cyber attacks on independent media have been common in recent years. One of the difficult things about understanding the cause and impact of DDoS attacks is that it is rarely clear who is behind the attacks. We have little or no evidence, for instance, that the Russian government is involved in these or other attacks. This is partly due to the nature of DDoS attacks, which often come from large collections of infected computers and so are very difficult to track back to the responsible actor. Governments have also avoided taking responsibility for these sorts of attacks, in constrast to the way that many government happily defend their filtering practices, perhaps because the attacks are often associated with the cyber-criminal gangs who build and run botnets.

What makes these attacks different is the number of sites attacked at the same time, and their close timing around the elections. We asked our friends at Arbor Networks, a leading provider of DDoS monitoring and protection services for Internet service providers and large content hosts, for any data they have on these attacks. Among other DDoS monitoring systems, Arbor has a large collection of taps installed in botnets, through which they are able to listen to the commands sent to the botnets. Jose Nazario reported back to us that starting on December 1 and continuing through the election on December 4, they saw commands come from just two botnet controllers to attacks the following list of sites, nearly all of which are independent media or election monitoring sites:

New Times (Oppositional news site The New Times)
Echo of Moscow (Leading Independent radio station Echo of Moscow)
Novaya Gazeta (Major oppositional newspaper Novaya Gazeta, often critical of the Kremlin)
Novaya St. Petersburg (St. Petersburg Novaya Gazeta site)
Kommersant (Major Russian news daily)
Public Post (online news site, had published stories about map of violations and Golos)
Slon (Online News site, partnered with Golos to publish ‘map of violations’ after Gazeta backed out)
Bolshoi Gorod (St. Petersburg news site)
Golos (Website of independent election monitor Golos)
Ikso (an outlier, the election commission of Sverdlovsk region)
Ridus (online news/citizen journalism site)
Zaks (a popular political website in St. Petersburg)
Pryaniki (a popular portal in Tula)
Map of Violations (Golos crowdsourced election violations map/site)
files.kartanarusheniy.ru (sub domain of ‘map of violations’ site)
LiveJournal (Major Russian blog platform)
Kotlin Forum (not accessible: Yandex search indicates a forum related to Kronshdat)
Kotlin (not accessible, Yandex search indicates news and info related to Kronshdat region)
GosZakupki (another apparent outlier in the group, a portal for Russian federal and local government tenders)
The Other Tver (oppositional Tver news and analysis site)
RosAgit (Web site connected to activist and blogger Alexey Navalny, which today is focused on promoting protests across Russia scheduled for December 10).

Botnets are often rented out for a variety of reasons, including spam, click fraud, and credit card theft, as well as DDoS attacks. It could be a coincidence that two botnet controllers were independently rented by a collection of actors to attack these sites during the election, but that coincidence seems highly unlikely. It is much more likely that some one or two actors was trying to take down a broad swatch of the Russian independent media landscape during the critical period of the election. We have see many, many attacks against individual media sources in the past in Russia, but we are not aware of any previous coordinated attacks against this number of sites at the same time.

The Arbor data, of course, says nothing about why these sites were attacked, but one argument put forward by editor-in-chief of Echo of Moscow Alexey Vendediktov (and many others), certainly seems plausible: “The attack on the website on election day is clearly an attempt to inhibit publication of information about violations.” Several, if not most, of these sites invited users to submit information on election violations, especially Golos, their violations map, Slon and Echo of Moscow. The timing of the attacks is also hard to see as coincidental, overlapping closely with the times that polls were open on Election Day. Most of the attacks also ended once the polls were closed. As is usual for these types of attacks, no one has claimed responsibility, even though they seem to clearly serve the interests of the government.

As the Berkman Center noted in its DDoS report last year, for media and NGOs that think they might be subject to a DDoS attack, putting data and information on major social media and Internet sites (like Twitter, Facebook, YouTube, Google, etc.) is a good back up plan, especially for smaller organizations with limited tech staff, since these major hosting sites are far more well prepared to defend against these types of attacks. For example, to our knowledge, the Google doc with over 5000 election violations created by Golos after its site was disabled, was never taken down. Alexei Sidorenko also has other details of how sites like Novaya Gazeta that were better prepared for the attack were able to help host Echo of Moscow blogs, which argues for these groups to support each other and host one another’s content, acting as a sort of ‘mutual aid society,’ which Jonathan Zittrain has written about. Also, we checked with one prominent Russian independent media site that we had worked with during the writing of the DDoS report about whether they had been attacked, and that site responded that they had used Twitter for all of their election coverage, specifically to avoid DDoS attacks. That site’s strategy was successful, as Twitter was either not attacked or withstood any attack during the election.

I’m just waking up to discover that, coinciding with today’s Russian Duma elections, there has been a series of major DDOS attacks that have at times brought down a number of leading independent media outlets, the LiveJournal blogging platform, and the online ‘map of [election] violations’ by election watchdog group Golos. Key independent mass media sites include the very influential Echo of Moscow radio and newspapers Kommersant, Novaya Gazeta (which is often critical of the Kremlin, has been the victim of DDOS attacks previously and has also had a handful of its journalists killed over the last few years), Bolshoi Gorod, slon.ru, and the more oppositional New Times. The election watch dog group Golos has been the target since last week of a government campaign against ‘outside’ influence in the election (they are funded by US and European groups). They were the subject of a thirty minute NTV special last week after warnings about outside interference in the election from Putin. The primary Russian blogging platform LiveJournal, which hosts the majority of blogs focused on politics and public affairs has also been attacked. The number of sites attacked at once seems unprecedented, and taking place during the Duma election cannot be considered a coincidence. As usual with DDOS attacks, it will likely be difficult if not impossible to determine who is behind the attacks.

Gregory Asmolov at Global Voices has the most exhaustive list of sites attacked that I’ve seen:

thenewtimes.ru, echo.msk.ru, novayagazeta.ru, kommersant.ru, publicpost.ru, slon.ru, Bolshoy Gorod (bg.ru), golos.org, ikso.org, ridus.ru, zaks.ru (Saint Petersburg), pryaniki.org (Tula), crowdsourcing platform “Karta Narusheniy” and Livejournal platform.

Alexey Sidorenko has a couple updates on his Twitter feed @sidorenko_intl

And the Moscow times election live blog also has some details and updates.

U.S.-Japan relations have never been stronger, but change during the month of August is perhaps the greatest since 1955.  On August 20 the U.S. Ambassador John Roos assumed duty of the Embassy in Tokyo, bringing with him from his perch as CEO of a top Silicon Valley law firm, a wealth of Internet, high technology, and legal expertise:

“Throughout his tenure, John helped lead the firm during the various waves of innovation in Silicon Valley, from the growth of software and communications to the Internet Age, the emergence of biotechnology to the present focus on clean technology and renewable energy.”

Concurrently Japan, the world’s second largest economy, is to host its national elections. A Parliamentary Democracy, Japan has both an Upper House of Councillors and a Lower House of Representatives. On August 30, elections in the latter will determine the distribution of power across the 480 seats, and it’s interesting because the Liberal Democratic Party (LDP) has been in power (except for 10 months, 20 days) since 1955. Largely seen as the guardian of the post-Cold War U.S.-Japan relationship, the LDP has been relatively unopposed in ruling seats in the Japanese Diet.  However, polls indicate that the Democratic Party of Japan (DPJ) could alter Japanese history.

Currently the DPJ holds 112 seats in the Lower House.  They would need 220 to have a “relative majority,” 241 for a “majority,” and 269 for an “absolutely stable majority,” according to Takako Hikotani of Japan’s National Defense Academy, formerly at Harvard. A potential LDP loss, and DPJ win would unsettle an institutional path dependence, and shake up ties in a country built of relationships. Such a DPJ win could have profound impact on US-Japan political, security, and technology relations.  With issues such as the Support of Forces Agreement, a potential US-Japan Free Trade Agreement (FTA), Okinawa U.S. troop deployment, and Japan’s perennial “Article 9” debate on the purview of its military, a DPJ win next Sunday could create substantive bilateral and regional change.

On security issues the U.S. collaborates with Japan on regional air defense, missile defense, and maritime security, but it’s the economics that are perhaps even more central to the bilateral US-Japan amity. The US and Japan cooperate on multiple fronts, not least of which is technology development. Recent Japanese GDP data indicated 3.7 percent growth, but confirmation that the country is still reliant on export demand in the West.  “Decoupling” remains but a sound-bite on CNBC. Domestic groups such as the Japanese Business Federation state that a DPJ win could mean labor reform, new targets on unemployment, and potential for immigration policy alteration.

But at the end of the day, a DPJ win will not likely alter Japanese green technology development as a priority, and U.S.-Japan technology transfer as a silicon cornerstone of pan-Pacific partnership. While the acronym associated with Lower House Diet seats may change, while business-to-government relationships may need to refresh, and while this is no doubt historic, bilateral technology cooperation will remain a core issue. President Obama’s appointment of John Roos, a lawyer skilled in Internet venture capital, heralds a cooperative continuation. Perhaps it’s fitting that Ambassador Roos’ transition from Silicon Valley to Akasaka can be monitored across Japan on Palo Alto-based Facebook, and described on Deputy Chief of Mission James Zumwalt’s blog, “Z Notes.”

The Silicon Valley has yet to create true forecasting technology, but certain online tools are providing voyeurs with the ability to interpret political events in terms of observable changes over time. In some cases, comparing relative change over time with against an expected baseline of activity can indicate predictive deviations. Explosive growth in use of the term “SBY” across Internet platforms corroborated what polling said off-line: Incumbent President Yudhoyono was re-election bound. And Google Trends data stood in contrast to polling expectations that Jusuf Kalla would lead Megawati in second place.  Google, and not polling data, corroborated actual electoral ordering.

Last Wednesday, on July 8th, Indonesia swiftly completed the second direct democratic presidential election in the country’s history. According to a national polling group, Lembaga Survei Indonesia (LSI), incumbent Susilo Bambang Yudhoyono (popularly referred to as “SBY”) won 60.82 percent of the vote, with the opposing Democratic Party of Struggle (PDI-P) candidate Megawati winning 26.57 percent, and the Golkar party candidate Jusuf Kalla taking 12.61 percent. Despite tepid claims that 5.9 million fictitious names had been included among the eligible voters (made by Megawati’s billionaire financial contributor, Hashim Djojohadikusumo), the election took place without incident.

LSI Regional Presidential Election Data.

Observation of relative trends over time has been used in many contexts. For example, Raymond Fisman, co-author of Economic Gangsters, observed corruption by monitoring stock prices and news. Under the Indonesian Suharto regime, insider-information as to Suharto’s health moved the then-Jakarta Stock Exchange before news became public. In this case, insider information, driven by concealed concern over political change, facilitated opportunistic buying and selling of stock that, in moving the market price, helped reveal corrupt practices.

Today, Internet users reveal themselves publicly in a variety of ways that collaboratively paint a picture of preferences and concerns that, if not generally applicable to the populous, are immediately reflective of the online demographic in the region. In Internet ecosystems such as the Netherlands, where 90 percent of the population is online, or in the United States, where 72 percent of Americans have access, Internet trends can be more widely extrapolated to indicate the public ethos. In Indonesia, despite its low Internet penetration of roughly 5 percent, the Internet is still a useful tool to observe opinion on important issues as well as regional strongholds of support.

Over the course of 90 days leading up to the election, top Google queries across Indonesia almost exclusively included references to popular networking sites such as Facebook and Friendster. As such, among connected Indonesians, use of social networking platforms is important. “Facebook Lexicon,” a tool that allows one to observe trends of terms or topics used in “wall posts” between friends, becomes relevant in indicating shifting ethos. Within the Indonesian Facebook demographic — admittedly a small and likely young group– Facebook Lexicon reveals that over the last year there has been significant change in topics of on-site political discussion.

While Facebook is a networking and discussion platform and its use typically involves conversation, active use of Google search engine indicates explicit interest. And over the same period, data from Google Insights for Search queries confirms a swelling Internet interest in incumbent candidate SBY. Since January 1, 2009, Google queries across Indonesia on “SBY” grew by 625 percent, compared with 40 percent on “Mega.” While relative search on “Jusuf Kalla” increased by 1100 percent, his absolute search volume was roughly 90 percent lower than SBY.

Telephone polling without regard for demographics.

Retrospective analysis is always problematic. One must be wary of ex-ante conditions observed ex-post, and the ease of false attribution. And as was seen in the telephone polls of American voters prior to the 1948 presidential election, the misinterpretation of niche trends for alterations in public opinion can yield headlines such as “DEWEY DEFEATS TRUMAN,” perhaps the greatest Chicago Daily Tribune gaffe. But as Google.org and the CDC have shown, aggregated search engine query data can, by observing online health-seeking activity, “accurately estimate the current level of weekly influenza activity in each region of the United States.” In politics, if the frequency of Internet user queries about candidates correlates with the percentage of votes cast, then perhaps this information could also be applied to pre-election statistics, which could augment offline polling data on connected demographics.

Pre-election polling by LSI and other Indonesian polling groups showed SBY leading followed by Jusuf Kalla.  Putative belief was that Megawati was entirely out of the running. Pre-election Google search query data pointed to an online reality that Megawati was still garnering much attention.  In fact, nationally, the results for actual votes cast more closely followed the numbers of Internet search query data –not domestic polling numbers. Perhaps such observation is mere coincidence. Perhaps Internet Search Analytics is an increasingly important data point to cohere with off-line demographic polling.  Today the online information-seeking behavior of a geographically diverse sample of connected Indonesians is perhaps illustrative of broader pre-electoral interests. It’s not perfect, and its scope is limited to Internet penetration, but I’m betting that tomorrow’s crystal ball could be made of silicon.

Photo: Daily Dish

The LA Times and other media outlets are reporting that a resilient group of Iranians has defied the government by publicly protesting the nearly month old election results. The opposition chose today to commemorate the 10th anniversary of student-led protests against the closing of reformists newspapers by conservatives during the Khatami presidency. Andrew Sullivan and others are back to relying on Twitter for updates. According to the LA Times, over a thousand protesters turned out in Tehran, and they have been met with violence by the Basij and regular security forces. In an attempt to thwart protest organizers, the government has cut off cell phone access for the last few days, although they apparently released 2,000 who had been arrested after earlier protests. On Tuesday, Ahmadinejad called the election, the “freest ever,” which alone was probably enough to push the opposition back into the streets.

Yesterday, on Sunday July 5, Mexico hosted its mid-term elections, bringing change to its lower house of Congress, six governor and hundreds of mayoral offices. The reform agenda of President Felipe Calderon will depend on his ability to secure a majority in the 500-member Lower House, or Chamber of Deputies, or “Deputatos.” Calderon’s party, the National Action Party (PAN) competes with the Institutional Revolutionary Party (PRI), the old garde in Mexican politics until Vicente Fox upset the trend nine years ago. Though Calderon’s PAN hold more Senate and Deputy seats, they lack a majority, as the Democratic Revolution Party (PRD) occupies highly coveted seats.

Sunday’s elections are pivotal in the ability to provide majority. While no Senate seats are up for grabs, should PRI or PAN secure sufficient seats in the lower Chamber of Deputies –350 of whom are elected, 150 appointed– the President’s ability to push reformist policies will alter. With the onset of Swine Flu, the disengagement of tourists, rising unemployment, and destabilizing macroeconomic climate, the mid-term election offers potential for facilitated policy change, not to mention a tacit referendum on Calderon half way through his six-year term.

Today, Mexico has roughly 27M people, or 25 percent of its population, online. Although looking at Internet data in  low-connectivity nations can be problematic –as many online in Mexico are “Panista,” meaning they support the Calderon’s PAN party– such data can still be illustrative if used to describe proper demographics.  Before Sunday’s elections, online trends point to pockets of support across demographics and regions. It can point to issues of regional importance. Relevant to Mexico’s youth demographic, Facebook’s Lexicon displays prevalence of terms on Facebook Wall posts. When comparing “PRI” with “PAN,” the margin of difference for Calderon’s PAN has increased since March 2009. In fact, Facebook Wall reference volume on PAN is triple PRI, and peaking around June 15. As yet, there is still no means of parsing Facebook Lexicon data by geography.

Facebook Lexicon PRI vs PAN Wall Posting Data

Regarding use of Google, 30-day moving averages of relative search data indicated that Calderon’s PAN was leading in regional online interest, with strongholds in states of Sonora, Jalisco, Distrito Federal (Mexico City), Nueva León, and Baja California.

Geographic Distibution of Relative Google Search on "PAN"

The 30-day domestic Mexican Internet search volume puts PAN at roughly 40 percent greater volume than PRI, however recent spikes in traffic and initial post-election results are putting PRI roughly ten points up on PAN. Internationally, the importance of the mid-term election as partial referendum is indicated by high volume on President Calderon. Outside Mexico, greatest interest in Calderon –as a proportion of domestic search– comes from Puerto Rico and Costa Rica. Outside of Mexico, the issue remains important across much of Latin America, in Peru, Colombia, Venezuela, and even Spain.

Global Geographic Overview of Google Search on "Calderon"

Initial post-election results show Calderon’s PAN trailing the PRI in new Lower House seats won. Diminished PAN influence in the Chamber of Deputies will hamper Calderon’s reform movement, and strike an initial blow to the Mexican economy, with lack of Congressional majority undercutting likelihood of decisive leadership. As of 9:40AM, stocks and the Mexican Peso had fallen by half of a percent against the Dollar. Initial results indicated that in Mexico online interest in party and candidate terms by connected voters did not translate into a reflection of ballots cast. But the Internet demonstrated that such electoral events matter not only in Mexico, but also indicated a leading interest in Mexican politics across Latin America and across much of the globe.

The summer months of 2009 have already played host to game-changing elections in the world’s largest Hindu and Shiite Muslim nations, India and Iran respectively. On July 8, Indonesia – the world’s fourth-largest by-population nation, the world’s largest Muslim country as well as largest Muslim democracy– will hold its presidential elections.

On July 8, Demokrat party incumbent Susilo Bambang Yudhoyono will face off against the incumbent Vice President Jusuf Kalla, now the Golkar party presidential nominee, and against 2001-2004 Indonesian President Megawati Sukarnoputri, also daughter of Indonesia’s first President Sukarno. Megawati is the leader of the opposition party known as Partai Demokrasi Indonesia Perjuangan, or PDI-P. Her controversial career soldier running mate, Prabowo Subianto, is the son-in-law of Suharto and the well-heeled founder and former Presidential nominee of the Gerinda party.

30-day domestic relative data on most popular candidate terms

While the perennial elite continues to vie for Indonesia’s top office, political engagement is moving from the streets to the information superhighway. Despite religious differences, the most salient non-domestic interest in the Iranian elections came from Jakarta, where –according to Google Insights for Search– Indonesian (Bahasa) trailed only Persian as the language of choice for entering Google search queries on Iranian presidential candidates. Outside of Iran and its diaspora, Indonesian interest in Iranian politics underscores religious trans-national solidarity, and an increasingly politically active youth demographic.
Read the rest of this entry »

By John Kelly and Bruce Etling

While Twitter is getting a lot of attention in the current Iranian crisis, it’s good to know that the robust Iranian blogosphere also remains active in the face of the government’s interference with the Internet. The figure below shows new blog posts on blogfa.com, the dominant Iranian blogging platform, over the past three weeks. While some Blogfa users are outside Iran, the vast majority are inside. We can see significant, through sporadic, disruption of Iranian blogging for a period of about two and a half days beginning a day after the disputed election. After that, posting returns to roughly pre-election levels.

What are bloggers talking about? A scan of text reveals high levels of discussion about politics. Many bloggers continue to link to websites supporting Mousavi (such as mirhussein.com), whereas linking to the main site supporting Ahmadinejad  emtedadmehr.com) has nearly stopped, including among conservative political bloggers.

Read the rest of this entry »

The Berkman-affiliated Web Ecology project, lead by the Internet & Democracy’s own Tim Hwang, has done some amazing and very timely research on Twitter in Iran. This adds some more quantitative data to our Op-Ed in the Post last week. The key findings:

* From 7 June 2009 until the time of publication (26 June 2009), we have recorded 2,024,166 tweets about the election in Iran.
* Approximately 480,000 users have contributed to this conversation alone.
* 59.3% of users tweet just once, and these users contribute 14.1% of the total number.
* The top 10% of users in our study account for 65.5% of total tweets.
* 1 in 4 tweets about Iran is a retweet of another user’s content.

You can download the full PDF report here.