Skip to content

Cloud computing, cloud commuting and risk management

14-May-09

I’m a big fan of Zipcar for many reasons, among which the least-discussed is that it lets me never worry about car maintenance. I’m one of those auto n00bs that mechanics love to see come through the door: ignorant, anxious, and trusting. So owning my own car is an ongoing maintenance liability: every “check engine” light is yet another opportunity to blow a few hundred dollars on repairs of dubious value.

Zipcar lifts the burden of car ownership and gives me what I want: a convenient way to get to the outlet mall and back. I don’t need to take adult ed classes on auto maintenance nor turn car ownership into a hobby.

Zipcar does for cars what The Cloud does for computing. It divvies up labor and lets specialists deal with issues with far more expertise and much better economies of scale than distributed ownership. We don’t need to know how to change the air filters or set up MySQL to drive to the beach or post a photo album.

On top of efficient division of labor, cloud computing/commuting also distributes risk appropriately. This means that the inevitable lemon car or DOA hard drive is handled as part of a larger batch rather than dumped, hot-potato-like, on individual hapless victims. This also means that consumers, in aggregate, make better choices. When presented with two hard drives — one of which is $10 more than the other, but also 5% more likely to fail — an individual is likely to go for the cheaper option and roll the dice. The cloud, on the other hand, is more likely to make rational cost-benefit analyses. An sysadmin who buys 100 hard drives knows that 5% failure rate means 5 dead drives, not a random gamble.

This kind of logic extends to all sorts of capital goods, including housing. Putting so much capital into a single investment strikes me as somewhat feudal in an era when capitalism argues for diversification and specialization (that is: buy REITs and outsource your real estate management).

What I like best about the cloud approach is that it’s eminently capitalist while capturing the flavor of socialism. We pool our resources, but we pay for what we get within a robust marketplace. (Zipcar will have really succeeded when they face a viable nationwide competitor).

Now I don’t believe that we should completely alienate our cars/condos/computers to some vendor and end up at its mercy. Even as I keep more and more of my stuff on Google and other clouds, I also want the option of backing it up on my own personal hard drives. And yes, some people take deep pleasure in ownership, tinkering with the car or repainting the shed. (I myself just built a new computer this week). But for those of us who aren’t expert mechanics, programmers, or construction contractors (nor friends with one), trustworthy cloud services can help mitigate the risks associated with ownership while tapping into expertise not otherwise accessible.

Video game interfaces for real-life war

05-May-09

XBox sniper controls?

XBox sniper controls?

As war becomes increasingly virtual, will it also become increasingly inhuman and thus inhumane? PW Singer lays out issues related to this question at TED, posted recently, in which he specifically cites Grand Theft Auto as evidence that “we do things in video games we wouldn’t do face-to-face.” He quotes one soldier who specifically says, “It’s like a video game.” Yet Singer also acknowledges that Predator Drone pilots apparently suffer higher rates of PTSD than their on-the-ground counterparts.

Will video game interfaces make what Singer terms “cubicle warriors” cold-blooded killers? Right now these remote-controlled robots largely borrow hardware interfaces from video games — see the image linked from this FOX News story or check out minute 10:30 in Singer’s talk. But what happens if and when they begin borrowing software interfaces from games as well? (The remote-control systems do already feature crosshair targets — but video games had first taken that from real guns.) Is an Ender’s Game
scenario — when the soldier doesn’t even realize he is fighting a real battle — possible?

Interface design isn’t quite the same as “codelaw” — that is, embodying laws in code — but in some ways it’s even more powerful, and therefore more potentially insidious. Many of the examples of choice-shaping that Thaler and Sunstein cite in Nudge are, in fact, interface innovations. But if interfaces can dehumanize, can they also re-humanize? Video games are not known for their emotional range, but I agree with those who believe that’s a matter of historical accident, not destiny. If video games can evoke authentic emotion, can we infuse it into our military software interfaces? The fact that Predator drone pilots suffer PTSD suggests that a digital screen need not cripple our humanity.

Cross post from Valuable Games.

Engineering a better virtual town hall

26-Mar-09

President Obama and his new media team are rightfully receiving kudos for their inaugural online town hall. Roundup at Personal Democracy Forum. But as a pilot, there’s room to improve, as the first commenter on the linked PDF post points out. Moving forward, the new media team should focus on re-tuning the technology to hit the core values and purposes of town halls and citizen participation:

1. Patch vulnerabilities. Whether or not you believe legalizing marijuana is a top-echelon issue facing the country, most of the top-rated MJ questions had little or passing relevance to the categories they dominated. The last category of question listed, “Budget,” became a honeypot for swarms of legalization advocates (the first seven of the top ten questions were on that topic), with only the addition of the word “tax” differentiating it from similar questions voted up in the “health care” and “green jobs” categories. I’m inclined to believe this was an authentic grassroots movement, but it could just as easily be engineered as a bot or mechanical turk astroturf campaign. What’s particularly pernicious about using crowd-sourced moderation is that the campaign wins either way: at a minimum, millions of Americans will be forced to read their submissions, even if only to vote them down.

2. Nuance the moderation: I voted on some 40+ questions and quickly began to realize that a straight up/down/abuse vote wasn’t capturing my opinion. For one thing, it became clear that if I wanted my interests to rise, I should vote against everything else (much like the way voters game multi-choice elections with bullet voting). It’s important for the system designers to realize that they are developing a game — a set of rules that determines winners and losers. For another, I found I had more specific things to say about each one: that a question was off-topic, or didn’t really ask a question, or was too generic, etc. In fact, I guess what I really wanted was:

3. Allow interaction: If the White House wants real civic engagement, it shouldn’t be built as spokes on a single hub (citizen -> President). The beauty of the Internet, like democracy, is that it’s many-to-many. I recognize that allowing citizens to talk to each other opens huge and difficult problems that make the deluge of posts demanding to see the President’s birth certificate seem trivial by comparison. Perhaps it’s up to civil society to pick up where Open for Questions leaves off — given enough lead time, citizen associations can build their own events off the town hall to host more robust discussions that can’t happen in the Presidential site. Still, this experiment is one of the closest things to a true public commons on the Web we’ve seen so far, and it’d be a shame if the only way to run it were a state monopoly that shunts citizen discussion off to private spaces.

4. More personality: One of the strengths of the town hall format is connecting abstract public policy to the lives of real, visible people. The format of Open for Questions (very limited space, no nuanced voting), however, privileged generic questions that went straight to the point and didn’t give a strong sense of who the person is and what their circumstances are. I felt a very strong difference in affect between Obama’s interaction with online questions (which was practically a press conference) to the video and especially live, in-person questions (which felt much warmer and more personal).

5. Or focus on the Internet’s strengths. Scratch that last suggestion. Maybe nothing will ever beat the face-to-face conversation for warmth and authenticity. Why not focus the online town hall on the very kinds of questions that town halls are terrible at: those best answered nonverbally (whether numbers, illustrations, or charts) or which require the President to draw on his advisors and not just the talking points he’s memorized. (We want the President to manage a team, not to be a one-man savant, after all). Stretch the new media team’s capabilities and see if they can create interactive charts, videos, or even games to frame or illustrate the President and his advisor’s responses.

Finally, let us acknowledge what has just happened: President Obama and his team have engaged over 93,000 people in an online town hall conversation. I hope this is just the first step towards a more robust system of citizen engagement.

The environmental mortgage crisis

25-Mar-09

David Owen’s recent New Yorker comment, Economy vs. Environment, draws an apt analogy between the mortgage and ecological crises we face. Countless living beings over millennia deposited their saved energy as carbon so that we might burn the resulting coal and oil with less heed than a homeowner on her third subprime equity loan. When you consider how quickly we’ve sent these millions of years up in smoke, our energy profligacy outweighs our recent real estate binge by astronomical orders of magnitude. Yet the vice is the same: squandering hard-earned assets with little thought of caring for tomorrow.

And yet after several years of virtuously foregoing the clothing dryer, I’ve concluded that spending my hours hanging up wet laundry is a backwards step in evolution; we cannot build a modern economy on stone age technology. Clean, sustainable power to fuel a healthy an growing economy — one that lets us all enjoy the amenities of technology, including the dryer — is the only realistic way forward. If we are to redeem our prodigal dissipation of the riches that our ancestors bequeathed us in the form of fossil fuels, it is to make that downpayment on this future way of life.

Obama’s Presidential Library should be virtual

25-Feb-09

The Boston Globe’s Mark Feeney asks, “Where would an Obama Library make most sense: Hawaii? Kansas?”

The answer, obviously, is cyberspace. As our first Web-savvy President, Barack Obama should put his Presidential Library online. If his Transparency and Open Government Initiative succeeds, most of the Library will already be built by the end of his term. Then it’s a matter of working with his brilliant Web team to design, curate, and future-proof the space.

Then instead of raising money for one library, put the funds into the public library system nationwide, so all 50 states benefit. That would be a legacy all Americans can be proud of.

WBUR supports Fair Trade

11-Feb-09

This year’s WBUR annual Valentine’s Day fundraiser features Fair Trade flowers from Winston Flowers! This means that both workers and the planet are treated right.

Find out more about the Flower Label and Transfair Fair Trade Certified flowers.

Winston Flowers

Donate now to WBUR and support local news.

Obama’s non-reductive rhetoric

21-Jan-09

Whatever the accolades for the speech that Obama delivered at his inauguration, it seems it won’t generate a singular sound bite as in JFK’s “Ask not…” or FDR’s “Fear itself” (Many of the major papers picked themes, rather than pluck quotes, although a few took to “hope over fear“). Pundits have hailed Obama as a gifted orator and skilled speechwriter, but generally overlook one aspect of his speaking that distinguishes it from his peers’: its complex structure resists distillation down to a single quotable phrase.

Non-quotability is often fatal to those who survive on media exposure, and in the early days of 2007 it seemed Obama’s campaign would drown in his words. Yet thanks to a deep and wide funding base, he survived long enough to turn that liability into a core asset. The Obama campaign is credited with doing an end-run around the media, going straight to the people through email and Youtube, but the candidate’s rhetoric aided that strategy. Summarizing his speeches is like paraphrasing a poem, which drives the public to seek out the original — the full text of Obama’s inaugural speech currently sits in the NY Times’ top 10 most emailed, and of course his so-called “race speech” famously convinced millions of Americans to sit down for a 40+ minute talk about one of America’s most difficult issues. By using complex constructions that resist distillation, Obama minimizes out-of-context critics, although he cannot mute them (witness the “bitter” comment).

Obama reached the pinnacle of oratory in his New Hampshire concession, which turned narrow defeat into triumphant victory. But brilliant as it was, the speech would have languished in the circles of hard-core Obama supporters were it not for Will.i.am and Jessie Dylan’s recognition that its core, can-do optimism needed a fuller articulation than the mainstream media could provide. So they set the speech to song, and suddenly many millions more were willing to stretch their attention from a 10-second soundbite to a 4:30 journey.

It was a stroke of brilliance for Will.i.am, and maybe of luck for Obama. Never since the rise of mass media has a campaign succeeded on assuming not only the basic intelligence of voters, but also their willingness to hear out a complex argument. The technology to bypass top-down media is one cornerstone of Obama’s success as a communicator. His nonreductive rhetoric is another. And if he continues to convince Americans to dig deeper into complex issues and not settle for the pat answer, we are already on our way to the change we need to take back our country.

Liveblogging the Internet & Politics conference 1

10-Dec-08

Building Collective Capacity : New Forms of Political Organizing

I’m here at the Internet & Politics conference at Harvard University, co-hosted by the
Berkman Center for Internet & Society and the Institute of Politics. The purpose of
this event is to gather leading practitioners and scholars to reflect on lessons
learned from the recent Presidential election and preliminary thoughts on moving
forward from here.

The majority of the conference will be held by Chatham House rules — no attribution.
But the keynotes are open, and here’s the first one, featuring Prof. Marshall Ganz
(Harvard Kennedy School) and Jeremy Bird (Obama for America).

Marshall is giving a backgrounder on organizing as a general matter. He has significant
resources available on this topic elsewhere, but here is the quick summary:

What’s needed for purposeful collective action?

1. Leadership: Achieving shared purpose in the face of uncertainty.

2. Community: A collective entity capable of exercising agency.

3. Power: A community able to use its resources to achieve its purposes.

Enablers:

1. Shared values (broader than interests — they are the sources of motivation)

2. Peer commitments

3. Shared structure

4. Shared strategy

5. Shared action

6. Action that is clear, specific, intentional, and can be learned from

To what extent can new technologies support these activities? (Or detract from?)

The Obama campaign emphasized carpenters, not tools.

Now for Jeremy’s response:

The “Jazz” and “Classical” metaphor from 2004 describes the connection well. Start with
the startegy and look at technology as a resource. Four stories that illustrate the
interdependence between technology and strategy.

1. April 11, 2007 — Florence, South Carolina. Not necessarily the most tech-savvy
state. In putting together tickets, were planning to capture emails, but then decided
to also capture cell phones. In December with Oprah, asked 30,000 to text the campaign
and also capture their numbers. Texting underappreciated — were able to text just the
team leaders. Or have volunteers send back pictures to keep other teams motivated. South Carolina house meeting program. Sam Graham-Feldson came to shoot video.
Despite the written program, no one knew what they were doing. What the video did was
tell the story: both to the rest of the campaign and to the community. (at 7:31, all the volunteers knew we’d won via the text message program).

3. Maryland. Teams who organized themselves using the MyBO tools. With two weeks left when Jeremy arrived to GOTV. This was a very different environment with much tighter connectedness. Through the ‘Net, bring together the volunteers into trainings, sufficient to hit every voter 3 times before the primary.

4. Pennsylvania. 8 weeks to go while TX and OH is going on. Took the online tool, PATeams tool, that allowed volunteers to log in and target neighbors. It was the “classical” and the “jazz” coming together. It enabled the volunteers to set and hit goals without setting up an office, to connect folks together and not just “go online and make calls” — they felt they were part of a community. Eventually led to the neighbor-to-neighbor tool.

5. Ohio (general election). We started to shoot all sorts of video. It was one of the most important things we did, because it told the story of what we were doing. Nationally, we set up VoteForChange.com that allowed people to download and turn in voter registration forms. As every individual downloaded the form, it gave organizers information about voters — but it turned out it was the most rich source of volunteers. These were young people who sought this out themselves.

These are still designed with field and new people sitting together. In 2008 we’re still figuring out if new media is a separate thing. We’re trying to figure out how to make organizing and online organizing work together.

In Maryland, a statewide group of 150 were already meeting every Saturday, all volunteers leading their own teams created through MyBO. Is it possible to use this technology for smaller campaigns? How to do it without the 2,500 paid campaign organizers that the Obama campaign had? The person who raises their hands first to be the leader may not be the best leader. One of the key questions is how to build leaders – how to define, how to select, are there tests? The hard part is that many of these are interpersonal skills; it’s not like learning geography. Marshall is trying to develop a distance course, but people will enroll as teams, not individuals. “Self-organization” is a chimera, a wish. It takes skill and practice. Buffy, in CA, was able to produce more calls per organizer than most other states using the technology to leverage. This was not the traditional leadership structure: we launched interdependent teams with shared norms, which diverged from the usual top-down individuals who burn out or have other issues. Coaching plays a critical role here. (Just because it’s face-to-face doesn’t mean it’s traditional).

Videos to ask people to sign up were very effective – saw great numbers. A good video, connected to a real narrative, and a way to capture people who say they want to do something connected to that. One of the real challenges is communication of emotion, affect, via the Internet. It’s easy to express emotion but harder to experience it, lacking the empathetic component. Video enables empathetic communication.

Counterfactually, what if Cesar Chavez had different tools – what difference do the tools vs. the carpenter make? If the Farmworkers collapsed because of a lack of accountability, then this risk is heightened without empathetic interaction. There was a time when the Farmworkers tried to market rather than organize the boycott – disastrous – perhaps the Internet would make this worse.

What to do when the wrong person becomes a team leader? Fire them. In PA, with only 8 weeks, we messed up. We spent too much time trying to figure out how to support bad leaders. Is there was more transparency because of the Internet tools – more data to measure outcomes.

  Politics | Comments (0) »

My.BarackObama.com — 2008 Game of the Year

16-Nov-08

It featured minimal graphics, no sound effects, and deeply flawed gameplay. Yet one of the most important game titles of 2008 was played by thousands and helped change the face of American politics. I’m writing about My.BarackObama.com.

Game designer and scholar Ian Bogost considered it a washout election cycle for political games. McCain had his “Pork Invaders” arcade gimmick, and Obama bought ads in Xbox Live (largely an indulgence). But I would argue that 2008 represents a watershed moment for video games, a moment when the medium showed that it can, indeed, change the world. My.BarackObama.com (“MyBO”) didn’t just communicate ideas. It encouraged people to go and do something.

MyBO awarded Obama supporters with points for taking real-world actions that would likely help the candidate win the primaries and the general election: making phone calls to voters, hosting gatherings, and donating money. MyBO wasn’t the first website to use game mechanics to stimulate real-world action. In 2004, ILoveBees sent thousands of players on a worldwide treasure hunt to promote the traditional console game Halo 2. In 2007, World Without Oil had participants imagine a world where oil prices become astronomical, then adjust their lifestyles in response. Over 18,000 people joined in, recording changes large and small that prefigured what people really did do in the actual oil shock of 2008. These Augmented (or Alternative) Reality Games all found ways to blend the virtual and real.

MyBO was the first serious ARG deployed by a political campaign. Sure, I’m stretching the term “augmented” a bit (unless you’re one of those who believed that all Obamabots lived in an alternate reality). And aren’t fundraising thermometers also a reality-based game where putting in $50 makes the mercury rise? I suppose – but what made MyBO revolutionary, and what puts it in the same category as World Without Oil, is that it also asked participants to engage in non-digital, non-virtual activity. You can donate money without leaving your bed or interacting with another human being. But calling voters requires an authentic human touch, even if the medium is digital (as it was for a colleague who Skyped voters on November 3 from Cairo, where she was at a conference).

Gameplay on MyBO was far from perfect. Part of the problem is that the boundary between digital and real remains only semi-permeable. For example, in January, my partner and I drove down to South Carolina and spent a week in the trenches, eventually helping to run a bellwether staging location. For this – and for our subsequent work in MA, VT, and PA, we scored a big fat zero, because there was no way to let MyBO know what were doing. Meanwhile, others were apparently gaming the system by hosting bogus events or flipping through phone numbers without actually calling anyone, perhaps hoping to win various awards. (The site did limit the number of numbers it would give you within a specific period of time to limit this kind of abuse – or, I suppose, wholesale data-mining).

A typical quest (note the in-game manual):
MyBO -- call quest

MyBO pointsPerhaps the biggest problem of MyBO as a game was its failure to scale. It was disheartening to log in and see that you were in 266,442nd place. True, the points and ranking were meaningless (except for the ten lucky phonebankers who got to meet Sen. Obama), as they are in any game, and I suppose you could argue that the fact that there were 266,441 other people doing more work than you also said something important about the campaign. But the system would have been far more motivating if your cohort group was more local: all Obama supporters in your state, city, or your MyBO groups. After all, the strength of the grassroots resides in its person-to-person connections.

MyBO - Activity TrackerThe scoring system never did go local, but in early August 2008 the developers swapped out points in exchange for an Activity Tracker. Instead of winning absolute points, supporters “leveled up” the ranks from 1 to 10 (10 being highest). Groups as well as individuals also scored points, which helped people find others who were actually doing real work. Previously, it was hard to get a sense of how you compared to other volunteers: 266,442 sounds pretty low on the totem pole, but not if there are over a million registered users!

Some were upset by the change, which demonstrated that the points really did motivate some. Wrote one of the top 500: “GIVE ME MY POINTS BACK!!!! THEY DO NOT BELONG TO YOU!!!!!” – words not unlike an MMO player whose epic weapon has been nerfed. But for those lower on the scale – which would include all n00bs, the lifeblood of any campaign or MMO – the switch removed the sense of futility that pervaded the game before. (Points also decayed over time, which also gave n00bs a fighting chance. Consider it an estate tax for scores).

For most supporters, the points likely functioned as a curiosity. Still, the point system helped signal what kinds of activities really mattered, and it probably had something to do with the over 200,000 events hosted and 27,000 groups created on MyBO – an impressive number even after you discount some set of bogus ones put on to game the system. And then there’s two other scores, 203 and 8,481,030, the margin of victory for Obama in the electoral college and the popular vote, respectively.

A resounding victory for President-Elect Obama. And, I suspect, for the future of reality games in political and civic campaigns. (Full disclosure: including one I’m now working on a civic engagement game for Fair Trade).

From campaigning to governance 1: civic engagement

11-Nov-08

“Yes we can,” as an election slogan, implies a relatively simple mission: get more people to cast a ballot for your candidate than for the other one. But as Barack Obama’s creed pivots from a battle cry to a governing philosophy, what, exactly, “we can” becomes a much larger and more complex matter. So, too, is the potential role technology can play in an Obama administration.

In this series of essays I’ll look at how Obama’s new CTO might transform American democracy in three areas: civic engagement, administrative transparency, and legislative advocacy.

(Cross-posted at techPresident)
More…