The 2012 Obama campaign: Big Data Analysis and Gamification?

As the Republican presidential candidates continue to duke it out in contentious primary elections around the country, I’ve started to notice the increasingly public signs that the Obama campaign is gearing up for battle. Not surprisingly, I tend to focus on the Obama re-election team’s uses of digital technologies, where a number of shifts may result in important changes for both the voter-facing and internal components of the Obama For America’s (OFA) digital operations. I started writing this post with the intent of reviewing some of the recent news coverage of the campaign, but it turned into a bit more of a long-form reflection about the relationship between the campaign’s approach to digital tools might mean for democracy.

OFA 2.0: Bigger, Faster, & Stronger (Data)

A fair amount of media coverage has suggested that the major technology-driven innovations within OFA and the Democratic party this election cycle are likely to consist of refined collection and analysis collection of vast troves of voter data as opposed to highly visible social media tools (such as My.BarackObama.com) that made headlines in 2008.

As Daniel Kreiss & Phil Howard  elaborated a few years ago, database centralization and integration became core strategic initiatives for the Democratic National Committee after the 2000 election and the Obama campaign in 2008. These efforts have been expanded in big ways during the build-up to the current campaign cycle.

According to the bulk of the (often quite breathless) reporting on the semi-secretive activities of the 2012 Obama campaign, the biggest and newest initiatives represent novel applications of the big data repositories gathered by the campaign and its allies in previous years. These include the imaginatively named “Project Narwhal” aimed at correlating diverse dimensions of citizens’ behavior with their voting, donation, and volunteering records. There is also “Project Dreamcatcher,” an attempt to harness large-scale text analytics to facilitate micro-targeted voter outreach and engagement.

For a vivid example of what these projects mean (especially if you’re on any of the Obama campaign email lists), check out ProPublica’s recent coverage comparing the text of different versions of the same fundraising email distributed by the campaign two weeks ago (the narrative is here and the actual data and analysis are here).

(Side note: in general, Sasha Issenberg’s coverage of these and related aspects of the campaigns for Slate is great.)

What’s Next: “Gamified” and Quasi-open Campaign App Development

As the Republicans sort out who will face Obama in November, OFA will, of course, roll-out more social media content and tools. In this regard, last week’s release of heavily hyped “The Road We’ve Traveled” on YouTube was only the beginning of the campaign’s more public-facing phase.

The polished, professional video suggests that OFA will build on all of the social media presence and experience they built during and the last cycle as well as over the intervening years of Obama’s administration.

Less visible and less certain are whether any truly new social media tools or techniques will emerge from the campaign or its allies.  Here, there are two recent initiatives that I think we might be talking about more over the course of the next six months.

The first of these started late last year, when OFA experimented with a relatively unpublicized initiative called “G.O.P. Debate Watch.”  Aptly characterized by Jonathan Easley in The Hill as a “drinking game style fundraiser” the idea was that donors committed to give money for every time that a Republican candidate uttered particular, politicized keywords identified ahead of time (e.g. “Obamacare” or “Socialist”).

In its attempt to combine entertainment and a little bit of humor with small-scale fundraising, G.O.P. Debate Watch fits with a number of OFA’s other techniques aimed at using digital initiatives to lower the barriers to participation and engagement. At the same time, it incorporates much more explicit game-dynamics, setting it apart from earlier efforts and exemplifying the wider trend towards commercial gamification.

The second initiative, which only recently became public knowledge, has just begun with OFA opening a Technology Field Office in San Francisco last week.

The really unusual thing about the SF office is that it appears as though the campaign will use it primarily to try to organize and harness the efforts of volunteers who possess computer programming skills. This sort of coordinated, quasi-open tool-building effort is completely unprecedented, especially within OFA, which has historically pursued a secretive and closed model of innovation and internal technology development.

If the S.F. technology field office results in even one or two moderately successful projects – I imagine there will be a variety of mobile apps, games, and related tools that it will release between now and November – it may give rise to a wave of similar semi-open innovation efforts and facilitate an even closer set of connections between Silicon Valley firms and OFA.

Is This What Digital Democracy Looks Like?

I believe that the applications of commercial data-mining tools and gamification techniques to political campaigns have contradictory implications for democracy.

On the one hand, big data and social games represent the latest and greatest tools available for campaigns to use to try to engage citizens and get them actively involved in elections. Given the generally inattentive and fragmented state of the American electorate, part of me therefore believes that these efforts ultimately serve a valuable civic purpose and may, over the long haul, help to create a vital and digitally-enhanced civic sphere in this country.

At the same time, it is difficult to see how the OFA initiatives I have discussed here (and others occurring elsewhere across the U.S. political spectrum) advance equally important goals such as promoting cross-ideological dialogue, deliberative democracy, voter privacy, political accountability, or electoral transparency. (Along related lines, Dan Kreiss has blogged his thoughts about the 2012 Obama campaign and its embodiment of a certain vision of “the technological sublime.”)

All the database centralization, data mining, and gamified platforms for citizen engagement in the world will neither make a dysfunctional democratic government any more accountable to its citizens; erase broken aspects of the electoral system; nor generate a more deeply democratic and representative networked public sphere. Indeed, these techniques have generally been used to grow the bottom line of private companies with little or no concern for whether or not any broader public goods are created or distributed. Voters, pundits, President Obama, and the members of his campaign staff would all do well to keep that in mind no matter what happens this Fall.

Citizen journalism fail? I don’t think so

Much is being made of a false rumor about Steve Jobs’s health that started as a “user generated news item” on CNN’s iReport.com site.

Basically, someone using the handle “Johntw” claimed that Jobs had suffered a heart attack in an iReport posting around 9am on Friday morning. By 10am, when Apple rebuffed the claim, the Cupertino, CA firm’s stock had dropped almost 10% – or $9 Billion in shares.

Maybe if we just gave them all laptops... (by Olivander cc-by-nc-sa)

Predictably, the papers and news services reporting on the incident have rushed in to proclaim this merely another sad example of “the downside” (SF Chron) of “citizen journalism gone awry“(Bloomberg).

Even NYU Professor/Blogger/Media-studies-guru Jay Rosen chided CNN for their naivete, and is quoted by the SF Chronicle:

“I think if you are going to put closed and open systems under one brand, then you have to try to organize the open part so that it does not embarrass you…What the branded news companies cannot do is … create a low-cost open-gate ‘play area,’ where the citizens can do their thing and no one carefully watches over it, cultivates it, takes responsibility for building it or for steadily improving the quality of it.”

Rosen makes a great point – CNN should watch where it places its brand as it struggles to understand the shifting sands of the participatory web.

However, the fatal flaw in the rest of the coverage is that the authors and editors assume responsibility for the incident should be directed at CNN and it’s iReport platform. The sad truth of the matter is that none of the folks who repeated the rumor – and here I’m especially thinking of so-called industry experts such as Silicon Alley Insider’s Henry Blodget, who chose to broadcast the misinformation about Jobs on his influential blog – took the time to call a hospital or consider the very real possibility that information submitted to iReport just might be inaccurate.

Clearly, the fact that iReport uses the CNN brand likely facilitated the rumor’s credibility, but how does that make CNN responsible for the actions of an idiotic, gullible, or malicious user?

Blodget himself had the chutzpah to write, “‘citizen journalism’ apparently just failed its first significant test.”

To say that CNN or “Citizen Journalism” bears the blame for something as irresponsible as Blodget did is about as reasonable as saying Wikipedia is bad because college kids think everything on it is True.

It’s the users of iReport and Wikipedia who need re-educating, not the other way around.

Until information consumers – expert stock-pickers, journalists, and so-called everyday folks who get their news online – learn how to treat what they read with an appropriate level of critical skepticism, it doesn’t matter whether you have the editorial staff of the NY Times or an army of 10,000 monkeys writing the stories. There is always a risk that you will read something false.

If you read a user-contributed news item then make the decision to act on that information contained in that story, the institution of so-called citizen journalism is not at fault, you are.

Am I being overly-defensive here? Perhaps. Nevertheless, I’d like to think the point stands: when the corporate media blows a story they (rightfully) catch hell for it, but nobody suggests that the institution of professional journalism is to blame. Instead, we quibble about whether The New York Times should have fired Judith Miller sooner, or hired an ombudsman before the WMD mess and Colin Powell’s speech at the UN.

So-called citizen journalists and large-scale distributed news sites complement existing media institutions in numerous and valuable ways.

Indeed, the extent to which reputable news sources such as the NYT failed to perform the traditional role of the 4th Estate during the lead-up to the current War in Iraq is exactly what motivated some of the most important bloggers to jump into the fray.

Berkman@10: end of day one

I think I managed to make it out of the cocktail hour at the end of the day without any trouble. Now I get to relax and prepare for tomorrow’s schedule.

Today included a number of great talks. I also enjoyed my conversations with the other attendees.

This is the first conference I’ve been at with such intensive self-coverage: between the IRC backchannel, the twitter feed, the del.icio.us tags, and the live-question tool hosted by the amazing Berkman geek-cave crew (yeah Mike and Sebastian!) the information overload is pretty intense. Sorta like drinking from the firehose…

In case you’d like to follow the self-coverage, you can subscribe to the full feed of conference material (as tagged and submitted by the attendees)

What cumulative advantage means for debates about the networked public sphere

Everybody and their cousin’s got a link to Duncan Watt’s recent NYTimes Magazine piece on cumulative advantage. It’s a nice bit of public sociology and an interesting application of experimental methods to understand how social networks function.

The argument also has implications for the ongoing debates about the nature of the networked public sphere. Watt’s results suggest that there may be some merit to the position of folks like U Chicago’s Cass Sunstein, who claims that big media has historically promoted civic virtues by exposing us to ideas we wouldn’t encounter by googling alone. If, as Watts says, people’s interests are over-determined by knowledge of what is popular, then search algorithms predicated on popularity (like “PageRank“) could produce a feedback mechanism that stifles diversity in public debate. To adopt Matthew Hindman’s phrase, we’d be left with “googlearchy.”

Yochai Benkler has disagreed with both of these views for a while [full disclosure: I currently work as a research assistant for Benkler], in part because they both idealize the state of public discourse prior to the creation of the Internet. Watt’s data could just as easily be turned around to make the claim that traditional print media and television – much like the recording industry – were giving us a false impression of popularity (or importance) that only reflected the prejudices of a handful of editors and industry executives. By disseminating these perspectives widely, the big media therefore imposed an elitist politics and outlook on the public as a whole.

I’ll have to do some more thinking and reading to figure out where I fall on this issue – but for the moment, studies like Watts’ shed important light on the complex nature of social networks and reputation on the role of information in society.

Re-establishing community norms: Kos responds to Clinton-supporter “strike”

Yesterday, Kos responded to the pro-Clinton strikers. His post argues the following:

1. The site has never claimed to include everyone on the left or all Democrats.
2. The site has embraced a particular vision of political mobilization and the transformation of the Democratic Party (50 state strategy; grassroots oriented; change the DLC; reject consultants; no PAC money, etc.). Hillary Clinton actively opposes that vision.
3. Clinton is not winning the primaries and cannot win without dividing the party and staging a coup against the popular vote via superdelegates.
4. Clinton’s response has been to foment “civil war” within the party and for that she deserves whatever the blogosphere, Keith Olbermann, and others can throw at her.

The long quote below sums it all up nicely. The bold text was in the original:

To reiterate, [Clinton] cannot win without overturning the will of the national Democratic electorate and fomenting civil war, and she doesn’t care.

That’s why she has earned my enmity and that of so many others. That’s why she is bleeding super delegates. That’s why she’s even bleeding her own caucus delegates (remember, she lost a delegate in Iowa on Saturday). That’s why Keith Olbermann finally broke his neutrality. That’s why Nancy Pelosi essentially cast her lot with Obama. That’s why Democrats outside of the Beltway are hoping for the unifying Obama at the top of the ticket, and not a Clinton so divisive, she is actually working to split her own party.

Meanwhile, Clinton and her shrinking band of paranoid holdouts wail and scream about all those evil people who have “turned” on Clinton and are no longer “honest power brokers” or “respectable voices” or whatnot, wearing blinders to reality, talking about silly little “strikes” when in reality, Clinton is planning a far more drastic, destructive and dehabilitating civil war.

People like me have two choices — look the other way while Clinton attempts to ignite her civil war, or fight back now, before we cross that dangerous line. Honestly, it wasn’t a difficult choice. And it’s clear, looking at where the super delegates, most bloggers, and people like Olbermann are lining up, that the mainstream of the progressive movement is making the same choice.

And the more super delegates see what is happening, and what Clinton has in store, the more imperative it is that they line up behind Obama and put an end to it before it’s too late.

I agree with Kos’ assessment of the primary situation and the problems with the Clinton campaign’s reprehensible actions. I also agree that the pro-Clinton “strike” on the site is a violation of the norms established many years ago. This was clear from Allegre’s diary entry announcing the strike, in which s/he argues for a strange vision of Democratic unity in which party members don’t criticize each other (in a really bizarre twist, Allegre then mis-attributes that idea of unity to Barack Obama…this is polemical bunk). The Daily Kos leadership and community have never embraced that kind of vision. From a strategic perspective, I agree that they never should.

Kos’ post interests me for other reasons then. In it, he re-iterates the norms governing the community through a reference to the founding ideals of the site and an extension of those ideals to the current primary election situation. The preservation of the site’s original ideals depends on such occasional interventions from the community leader. In turn, the ideals and norms maintain the basis for large-scale collaboration and conversation.

But if that’s the case, does it negate what I wrote earlier about the significance of defection from large-scale collaborative communities? I don’t think so. Highly symbolic defections like this one still matter even if they are not grounded in an accurate interpretation of community norms. This skirmish, no matter how mundane or over-blown it has been, is part of the ongoing process of managing discursive production on the site.

Writers’ Strike at dailyKos?

If, like me, you spend too much time lurking around in the blogosphere you already have a sense that the dailyKos and MyDD communities have squared off behind the Democratic candidates. This was hardly surprising and certainly didn’t seem out of place in the flame-friendly environment of web-based conversations.

But this might be something new: yesterday evening, one of the many prolific community members on the Kos site, Alegre, announced a writer’s strike of Hillary supporters in her diary entry. Responses – both positive and negative – piled up quickly, including this one apparently from WGA member Jeff Lieber.

I am not really interested in discussing the merits of the arguments in the strike debate. Rather, I want to draw attention to the questions a strike raises about the nature of governance in Web 2.0 communities.

I blogged about this a few weeks back in one of my early posts. Building effective institutions of governance in the context of collaborative peer-production presents a number of deep problems. On Kos, conversation management is probably the most obvious of these. To understand how conversations are managed on Kos, though, you have to understand a little bit about the technical platform of the site.

Currently, Kos uses Scoop, a conversation management tool originally developed by Rusty Foster for the Kuro5hin community. The features of Scoop have helped Kos scale from the one-man show it once was back in 2002. Scoop includes a built-in, decentralized system for generating and measuring reputation among site users. Basically, every registered user can rate the other registered users’ comments. In addition, by commenting on other users’ diaries, community members can bring that diary increased visibility. The result is a massive conversation that pretty much runs on autopilot.

With the exception of Kos, ct the tech wizard, the contributing editors, and the handful of community members with front-page privileges, this reputational system structures the conversation among the site’s users. The problem is that the system isn’t fool-proof – or flame-proof. Angry blocs of polarized users can “approve” negative comments, thereby generating a feedback loop of discontent. Their opponents then retaliate in kind and create their own “peer-reviewed” circuit of annoyance. As the negative momentum builds, the argument attracts more attention through the site’s reputation systems and results in further polarization and negativity. The threat of a widespread disengagement becomes credible as more and more people find reasons to defect from the cooperative knowledge production model that makes the site so interesting in the first place.

My analysis may or may not be sound here, but the most important question – as Lenin always understood – remains: What is to be done?

Telling everyone to “just be nice” doesn’t work. The actions of the flamers already run counter to the behavioral norms set out in the FAQ on dKosopedia. Similarly, the strike seems unlikely to achieve the stated aims of those who have threatened to leave the site. Rather than bring about a return to civililty, the departure of Clinton supporters for MyDD and other sites will only reproduce the fragmentation of the political blogosphere within the progressive community. It is too early to know what this would mean for the netroots’ project of building a new democratic politics. But does it really have to come to that?

Instead, why not design a new technical platform to manage the conversation? What would such a platform look like? It would need to incorporate many of the existing features that have made Scoop so powerful and effective. However, it would also require some mechanism for recognizing negative/hateful feedback loops and enabling the community to address them in a more productive way. Political dialogue – whether between members of opposing parties or merely supporters of two Democratic candidates – will continue to be divisive. The trick is to leverage that divisiveness (is that a word?) so that it can produce a self-sustaining community that continues to generate strategic political advantages.

in other news…focus on the blogosphere and social networks

Forever behind on the feeds…

HuffPo fronts Pachacutec’s thoughtful analysis (originally at Firedoglake) on the somewhat unsubstantiated views of the democratic candidates that circulate in the netroots community. Pachacutec weighs the evidence and concludes that the possibility that either Obama or Hillary might “pull a Lieberman” is actually quite high and the blogosphere needs to recognize that it will likely have to hold either candidates feet to the fire if they ever make it into the oval office.

Charlene Li reads the social networking tea-leaves over at Groundswell and argues that the current social software ecosystem will look primitive in 5 or 10 years. Memorable quote:  “I believe that in the future, social networks will be like air.”

The Guardian goes out on a limb with a list of “the World’s 50 Most Powerful Blogs.” The list is lovely and certainly indicative of something, but without more details on their methodology for making the rankings, I’m inclined to question its validity.

I can’t wait to watch/listen to the debate on bloggingheads between Cass Sunstein and Henry Farrell over the roll of the Internet and the Blogosphere in American politics.

Lastly, and on a (somewhat) unrelated note: I’ve been writing about WIPO like it’s my job for the past few days, but that doesn’t mean the rest of the internet governance debate has stopped. Ian Lamont of The Industry Standard mulls ICANN’s apparent desire not to renew the US Commerce Department’s oversight of its activities (link is to the Slashdot version of the story). If this goes through it will be an important step towards ending the current US private sector hegemony over the management of the net.

studying web-based collaborative communities

I’m working on some preliminary research for a study of DailyKos and some of the other political blogs that continue to define the networked public sphere.

In the process, I had to start thinking more seriously about what it means to do an ethnography of this kind of community.  I’ve done some work on this before, so I had a few ideas, but the challenges and scope of Kos are a bit daunting.

As a result, I’m focusing on breaking down an initial assessment of the site into several categories (see below). I then use these to structure my observations and to help define problems that I’ll have to solve later (possibly with more than just a tab-happy browser window, the Internet Archive, and two tired eyeballs).

The main categories are:

• History and Evolution of the site – including community structure, software, interface, layout, etc.
• Organizational/Institutional Structure (behind the scenes stuff like money, hosting, contractors, etc.)
• The Current Community
  • social network topography (i.e. is it just another bow-tie?)
  • practices, norms, governance, etc.
  • signs of life off-line?
• Technical Platform & Software
• Content (production & consumption)
• Networks and public-sphere functions (linking, SEO, connections to the media, political parties, etc.)

Obviously, these overlap a lot and the list can get much more detailed (indeed it does in my notebook). The important stuff – at least the stuff that a lot of smart Internet research has identified as important – seems to be accounted for…

researching the political blogosphere

A series of questions I’m working on at the moment and some of the resources I’ve found:

• How do ideas move through the political blogosphere?
• What role does the political blogosphere play in “agenda setting” within the mainstream media (msm), political party elites, and networks of expertise (think tanks, consultants, etc.)?
• Do these roles vary for blogs on the right versus blogs on the left
• Also, how does blog governance operate across the political spectrum? Is there a predominant model of community organization that has emerged? Are there patterns that correspond to whether the blogs come from right or left?
• How do large, collaborative blogs produce stable community and governance structures? To what degree are they self-organizing and to what degree do they rely on various “levers” to reproduce stable patterns of collaboration and sufficiently low rates of defection?

Some interesting tools that should help me approach these problems include the following:

• Chapter 7 of Yochai Benkler’s The Wealth of Networks
• Existing tools that measure blog influence such as Blog Pulse from Nielsen and an interesting looking French project called blogopole
• Several papers and reports such as: “The Emergence of the Progressive Blogosphere” (which incidentally, seems to answer a few of those questions about the differences between right and left – at least, as of 2006); the work of Lara Adamic (I think she designed some of the blog tools like Blog Pulse); Matthew Hindman’s research on stuff like “googlearchy”; and whatever it is Ed Chi and the folks working on “Augmented Social Cognition” are up to at PARC in Palo Alto..

There are others (and I’ll try to keep adding them as I dig them up), but this is a good start. The big questions that I can try to answer here really have to do with the way this architecture (in the sense that a community design is often unplanned) relates to the “culture” of the political blogosphere. How does citizenship – or something like it – emerge in the blogosphere and other social spaces of the collaborative web? Why are the power-sellers uniting and what are they going to do?

the institutional side of Web 2.0

A recent piece by Chris Wilson on Slate.com as well as some conversations with Yochai Benkler at the Berkman Center have gotten me thinking more seriously about the institutional side of web 2.0 and the organizational structure of social production.

It seems like there is a growing realization that the revolution in networked production has involved more than friendly collaboration among like minded amateurs. Anyone who frequents web 2.0 sites already knows this – wikipedia has its chaperones, amazon has its super-reviewers, digg has its preferred posters, etc. Furthermore, anyone with a background in organizational theory would expect this. There’s no reason to believe that pure peer-production could scale without costs or without the creation of disciplinary institutions of some kind. Nevertheless, Wilson’s piece reads like an expose – similar to this earlier Slate piece by Garth Risk Hallberg on Amazon Book Reviews. The rhetoric of openness and collaboration may yet come back to haunt these projects.

From a more analytical perspective, though, these pieces raise some really interesting questions. For example: how do the different kinds of regulatory systems work within each of these sites/communities? What systems scale most effectively? What mechanisms determine the emergence or success of one web 2.0 organizational structure over another?

I’m going to be doing some posting about these questions in relation to the mother-of-all-web 2.0-sites, Daily Kos in the near future (part of my work with Benkler). However, in the meantime, I wanted to take advantage of these articles to start thinking about how to design researchable questions about to these issues.

You might start with several different kinds of questions:

(a) What characterizes the institutional design of the community? What are the levers of control, manipulation, power, and consensus?

(b) How did the institutional design evolve over time?

(c) What problems does the institutional design solve? In particular, how does it elicit the collaboration of participants?

(d) Where are the contradictions and tensions most likely to emerge given the structure of the field elicited by the institutional design and the forms of power involved? Where does the institutional design break down?

All of these lend themselves to very different kinds of research strategies.