Crowdsourcing

Open Government Panel at 3 PM Today

I'm generally not one for pimping panels (okay, okay, unless I'm on them), but I really do want to get the word out about this one. The Center for Strategic & International Studies is hosting a panel on "open innovation in government" at 3 PM today. I know—nothing like a handful of government bureaucrats quoting from white papers to liven up an otherwise drab, spring day. But if you're open gov geek (and isn't every red-blooded American?) this is a must-see. It features interesting thinkers ("open government champions," according to the press release) from an array of federal agencies, including USAID, NASA, and the State Department.

But most importantly, Beth Noveck, the New York Law School prof and longtime open government theorist, will be moderating. Noveck broke open gov ground with her Peer-to-Patent program (featured in—shameless plug alert—my crowdsourcing book), and is currently serving as the President's director of open government initiatives. She has consistently proven to have one of the most sophisticated visions of how government could use technology to collaborate with citizens, and her presence in the executive branch speaks well of the White House's genuine belief in technology's democratizing potential.

Appropriately, the panel will be streamed live, and Noveck will be taking questions from the peanut gallery (which is to say, us). Hint: Don't ask about marijuana legalization.

Cross-posted from the Epicenter Blog.

Crowdsourcing and the President: A Failed Marriage?

The blowback from President Obama's interactive town hall has been intense and widespread. In dismissing a legitimate policy issue the President seems to have shown an uncharacteristic degree of political tone deafness. There are many excellent reasons to rethink the War on Drugs—that most ill-fated of American conflagrations, and mostly bad ones for staying the course. Many in Obama's base felt betrayed by the brush off. And they weren't the only ones. A former police chief and mainstream newspaper columnists also cried foul. Donations to NORML spiked last week.

It's all terribly interesting, though not for any of the reasons people think. The incident signifies the end of one, increasingly troubled stage in the courtship between the President and social media, and — we can only hope — the beginning of another, more realistic and mature stage. At this critical juncture I'd like to offer some relationship counseling.

It's perceived by many that the forces of drug reform "hijacked" the White House’s Open for Questions platform. Indeed, decriminalization is nowhere to be found in any list of what Americans think are the most important issues facing the country. But this conclusion assumes the technology used by the White House is capable of creating a representative sampling of popular opinion. The tech doesn't do that, and we shouldn't expect it to. We possess other, highly effective tools for that job — they're called polls.

Open for Questions fits squarely within a genre of crowdsourcing I call "idea jams." These are often called suggestion boxes on steroids, or some such silly thing. But in reality they constitute their own evolutionary branch of brainstorming. Users don’t just submit ideas, but also vote and (usually) comment on them as well.

Idea jams are a big hit with the private sector. Companies like Starbucks, Dell, IBM and even General Mills have all adopted them, for the excellent reason that they’re a cost-effective method for product innovation, and inspire good will with your customers to boot. The best-publicized incarnation involves Dell's "IdeaStorm," which the computer maker used to tap its most loyal (or at any rate, most vocal) customers. They've now integrated some 280 suggestions into their product line. Tellingly, Dell used the same Salesforce.com platform that the Obama transition team used to produce the quickly — and justly — discarded Citizens' Briefing Book.

So if the idea jam format works for companies, why isn't it working for our President? A few reasons:

First, the White House isn't matching the right tool to the right job. "The whole point of [such exercises] is not to find the question that the whole group wants to ask and that is predictable – but to enable cognitive outliers to ask the unpredictable question — to promote ways of thinking about problems (and solutions) that are uncommon," writes Kim Patrick Kobza, CEO of Neighborhood America, which develops social software for business and government.

In other words, idea jams are built to allow people to discover the fringe question (or idea, or solution), then tweak it, discuss it and bring the community's attention to it. When Dell launched Idea Storm, it was "hijacked" by Linux die-hards which suggested (nay, insisted) that Dell release a Linux computer. These folks were "trolls" to the same extent the drug legalization lobby swamping White House servers are, and Dell struggled with how to deal with them.

The company's ultimate reaction is instructive. First, they merged all the Linux comments into one thread, giving much-needed daylight to other ideas. Next, they saw the value in what the Linux folk were saying. The loud and clear demand for an open source OS had revealed that there was a "constituency" large enough to justify enacting this particular "policy." Put another way, there was adequate demand to support a new product line. Three months after launch, Dell released three computers pre-installed with Ubuntu.

In this sense, last week's virtual town hall performed a valuable function. It highlighted an important, if non-urgent issue and stimulated an ultimately useful public dialogue. The problem was that the President's "Director of Participation" wasn't part of that conversation. Which brings me to my second point: Participation goes both ways.

"Idea management is really a three-part process," says Bob Pearson, who as Dell's former chief of communities and conversation rode heard on IdeaStorm. "The first is listening. That's obvious." The second part, Pearson says, was integration, "actually disseminating the best ideas throughout our organization. We had engineers studying IdeaStorm posts and debating how they could be implemented."

The last part is the trickiest and most important: "It involves not just enacting the ideas, but going back into your community and telling them what you've done." Starbucks, which maintains its own version of IdeaStorm, employs 48 full-time moderators whose only job is to engage the online community. In other words, Starbucks is investing the vast share of its resources in the second and third parts of the idea management cycle.

By contrast, the White House essentially used its platform as a listening device, and failed to participate in the ensuing conversation.

The White House faces technological and legal hurdles that Dell and Starbucks don't have to worry about, to say nothing of the political considerations of seriously entertaining a policy of decriminalization at the very moment when the White House most needs GOP votes.

If the goal is to allow citizens to express themselves, mission accomplished. But if President Obama truly wants to engage his constituents in a national conversation, to involve them in the hurly-burly of law-making, he'll need to evince a much better understanding of how the knowledge, opinions and, yes, wisdom, of a large populace can best be harnessed. For one, he could push Google Moderator to allow users to comment on each other's ideas. Disabling this otherwise standard feature neuters the Idea Jam process from the outset.

In its current iteration, Open for Questions isn't really enabling democracy, unless if by democracy we mean the "never-ending, small-bore struggle for advantage among constantly shifting coalitions of interest groups," a conception of politics articulated by the early 20th Century political theorist Arthur Fisher Bentley. This isn’t quite as uplifting a vision as the one we were treated to during Barack Obama’s campaign, but it may—in the end—be a more realistic one.

Cross Posted from the Epicenter Blog.

Links of the Week

My crowdsourcing radar used to consist of Google Alerts and a RSS feed from Technorati. Then I installed TweetDeck on my laptop. I'm not sure if it was the best or the worst thing I've ever done. I had thought of Twitter as a broadcast tool, but it's become far more valuable to me as a listening device. I used to say that keeping track of crowdsourcing's growth was a full-time job (the punchline being that I already have a a few full-time jobs).  Now it would take an entire newsroom (okay, a small one) to cover the diverse, imaginative—and occasionally wrong-headed—ways crowdsourcing is manifesting in our culture. Anyway, here were a few of the more significant developments from the week:

Open for Questions, or Open for Vote Rigging?
President Obama held the first crowdsourced press conference. The administration used Google Moderator to collect questions from citizens, and vote for those already posed. Results were decidedly mixed. The marijuana lobby turned out in force, effectively stuffing the ballot box for decriminalization topics. My colleage Nick Thompson has done my work for me here, here and here. The long and short of it is that one interest group gamed the system to vote their concerns to the top. As scores of other people have pointed out, decriminalization is a legitimate topic for a press conference, but hardly represents a pressing issue. My read: Open for Questions highlights that the ideastorm model of crowdsourcing is still very much in a beta phase, useful for some applications and counter-productive for others.

An Index of Crowdsourcing
Boy has this been a long time coming. Anjali Ramachandran, a strategist at London-based digital agency Made by Many, posted a wiki with 135 companies currently engaging in some form of crowdsourcing. It's a great start, and Anjali is asking us all to help expand it. Such efforts are crucial to the maturation and understanding of crowdsourcing. The phenomenon has grown so rapidly (and so haphazardly) that it's exceeded any single person's capacity to track it. I know there are a lot of crowdsourcing junkies reading this post. I'd encourage you all to go contribute.

Um, Do You Know How We Can Get Out of This Mess?
Ireland's economy has been especially hard hit by the Great Recession. Enter "The Ideas Campaign," or "the People's Campaign for Economic Growth," an open innovation project operated and paid for by Irish Web consulting firm, AMAS. Here's Springwise: "Launched just a week ago, the Ideas Campaign is asking the citizens of Ireland to propose innovative ideas to boost economic activity in the country across 19 key areas including manufacturing, technology, construction, retail and education." Great idea. Horrible execution. The ideas are collected, vetted and posted back to the site without any opportunity for the community to vote, comment or otherwise interact with them. A perfect use for aforementioned Google Moderator. Would the Dope Lobby—or more to the point, some special interest—have gamed this site like they did President Obama's? Possibly, but that's an argument for enhanced moderation, not locking out the crowd's input on, well, the crowd's input. If it works for Ireland, maybe Iceland will give crowdsourcing a spin.

Smartsourcing vs. Crowdsourcing
Pete Peterson has a very thoughtful essay at techPresident this week in which he examined the Obama administration's mixed results from such crowdsourcing experiments as the Citizen's Briefing Book (another ideajam that was stampeded by the drug lobby). He advocates bringing together "select group of citizens."

The government can serve a vital role as convener – bringing together smart people from a variety of viewpoints to collaborate and debate online over particular policy initiatives. These policy discussions – whether in wiki or blog format – can be kept transparent and open to the public’s view, but as we have seen during the Transition, these must be controlled in such a way as to prevent hijacking by small, organized groups."

I couldn't agree more. I would just like to point out that smartsourcing (great term!) is crowdsourcing. In my original article on crowdsourcing as well as in my talks, I've always said, "First—pick the right crowd." This has theoretical underpinnings. Scott E. Page, who's probably done more work than anyone in collective intelligence, calls this a "crowd of models." Diversity will trump ability, he notes, but only if a certain level of talent and ability are mixed in with that diversity. At any rate, Peterson's post is the read of the week, in my view.

Links from the Week

I'm just rolling in from an open jaw that took me from Austin to Boston and back. (I went to SXSW to moderate a panel and read from my book, then to the Berkman Center for a talk. More on these later.) What did I miss? To judge by my tweetdeck alerts, quite a lot. Here are some highlights, in no particular order:

Crowdsourcing Schadenfreude
A few days after sitting on my SXSW panel the Forrester analyst Jeremiah Owyang published an unfortunate post about Mzinga, one of the companies he follows. Within he reported that he'd been hearing rumors of financial difficulties at the company, and advised his clients to "stall any additional movement till they brief me next Monday" (italics his). Without recounting the resulting shit storm (Which you can read about here and here), I would like to pass censure on the (purported) use of crowdsourcing to determine whether Forrester should give Owyang the ax. Basically, a social media consultant asked his readers to vote on Owyang's fate. A) This is only crowdsourcing if Forrester were to base a decision on it (which they surely won't); B) it gives form to the sort of anonymous trollery that all is social media's least attractive feature.

Crowdsourcing Congress
I've been ignoring the rise of crowdsourcing in governance for the excellent reason that I'd have to drop my day job in order to due the subject justice. But that doesn't mean I've been lurking. Guest blogging at O'Reilly, Rep. Mike Honda, D-San Jose brought Gov 2.0 one step closer to fruition by asking the crowd how Congressional Websites might best take advantage of, well, crowdsourcing.

Crowdsourcing Recycling
Brilliant, common-sense use of community to disseminate recession-era home ec.

Army of ... Many?
Tantalizing, right? Tune in later when I actually have time to complete a friggin' blog post.

A Belated Congratulations: ProPublica Gets on the Bus

Last week a friend and erstwhile colleague, Amanda Michel hit not one but two home runs. Michel was the driving (and as our mutual friend David Cohn notes, unheralded) force behind the Huffington Post's citizen journalism project, Off the Bus. OTB was arguably the first truly successful attempt at citizen journalism, and the Columbia Journalism of Review wisely tapped her to write an in-depth feature on her experience running it, called "Get Off The Bus - the future of Pro-Am Journalism."

Amanda, David and I worked together on Jay Rosen's Assignment Zero. When it ended I wrote that it had been a "beautiful failure," which was to say that while we hadn't achieved our goals, we'd learned a lot about how the crowd might be tapped to produce quality, investigative journalism. I've been inexpressibly pleased to watch Amanda make lemonade out of those lemons with Off the Bus.

As it happens, she now has the opportunity to further develop the distributed reporting model. The same day the CJR piece came out the non-profit, investigative journalism outfit, ProPublica, announced it had hired her to head up its own citizen journalism effort. This is a big deal in our small world of pro-am journalism. Michel is possessed with a big brain, a big heart (an essential attribute when managing dozens-nay-hundreds of unpaid contributors) and a tireless work ethic.

Here's Knight Foundation's Michelle McLellan on what it all means for the journalism community at large. 

And if Crowdsourcing is Beautiful, This is its Soundtrack ...

Less profound than the Koblin/Massey joint, but much more listenable. Go here, then here to figure out wtf you just heard. Whatever we want to call this, I love it.

If Crowdsourcing is Evil, This is its Soundtrack

Aaron Koblin and Daniel Massey, crowdsourcers extraordinaire, strike again:

Is Crowdsourcing Evil and Other Moot Questions ...

Crowdsourcing, by its very name, encourages a comparison to outsourcing. But when Wired first published the article that entered the term into the popular lexicon, it was far from clear whether the phenomenon would realize its disruptive potential. Three years later it seems increasingly obvious that it will. Aided by a new generation of sophisticated start-ups, ever cheaper creative tools, and — most of all — a recession that is forcing cost-saving measures on businesses, crowdsourcing is rapidly migrating from the fringe to the mainstream.

Witness the upheaval currently afflicting the design industry, sparked by the rise of so-called "spec design" sites like crowdSPRING and 99designs. On these sites customers post creative briefs directly to the community, which then competes to create a design that best fits the clients' needs. A typical "assignment" will draw dozens of submissions. The winner receives a nominal fee (as little as $200), and the client receives a logo or Website design at a fraction of what a professional agency might charge. The losers get zip, which goes a long way to explaining why working on spec (aka 'on speculation," or without guarantee of payment) has always been considered the work of last resort for writers, designers and other creative professionals.

So one might expect crowdSPRING and 99designs to wither away like so many other seemingly ill-conceived Web 2.0 start ups. Instead, they seem to be flourishing. 99designs says it has paid out over $4 million to its community of 30,000 artists, and crowdSPRING expects to be profitable by next year. The success of crowdsourced design has sparked a vibrant, highly emotional debate within the design industry. (The brouhaha will go live at SXSW next week, when I moderate a panel on the subject of spec work in design.)*

Alarmed by the popularity of the spec model, a group of designers formed a protest group called NO!SPEC to persuade their colleagues (and prospective clients) to just say no to design contests. Their effort has not been in vain. The trade group AIGA — composed of some 22,000 designers — has gone so far as to stake out an official position on spec-work: "AIGA strongly discourages the practice of requesting that design work be produced and submitted on a speculative basis in order to be considered for acceptance on a project."

The controversy shifted into high gear last month after Forbes published an airy, one-sided look at crowdSPRING. In over 100 comments to the article one could read persuasive, articulate variations on a single theme: "Fuck you and crowdSPRING too." Several prominent design blogs posted their own jeremiads lambasting crowdSPRING. "Spec work has become a major force in devaluing the perception of graphic design in the business world," writes eyeCinq. "The folks that run these outfits have managed to figure out a way to get thousands of people — some skilled enough to earn a decent living — to work for them gratis. It’s an amazing sleight-of-hand," writes the logo factor.

It would seem that the squabble has united the design community against the barbarians at their gate. And that would seem to bode ill for the future health of the spec sites, right?

Don’t count on it. A similar debate was taking place in the stock photography world when we published “The Rise of Crowdsourcing” back in that bygone era of June 2006. The fact this debate has been largely settled — in favor of the barbarians — speaks volumes about where graphic design, and, for better or worse, most other creative fields, are heading.

I made this explicit comparison on my blog last August. The demand for low-end design has ballooned in recent years alongside the profusion of start-ups and small businesses. Conveniently enough, so has the supply of what we might call "low-end designers" (amateurs, recent grads and the like). According to Forbes there are 80,000 freelance designers in the US alone. Most of these are, proverbially speaking, waiting tables. When someone matches demand and supply, well that's kismet!

iStockphoto and other so-called "microstock" agencies capitalized on a similar disparity. The result was the total disruption of the $2 billion stock photo industry. iStock is now the third-largest purveyor of stock images, and some 96 percent of its "workforce" is comprised of amateurs. In my crowdsourcing book I posed the question of whether stock photography was an isolated case, or just the canary in the coal mine. It was an open question as of April 2008 when I submitted the final changes to my galleys. Now it ain't. The canary is prone, lying motionless on a bed of its own droppings. It looks like it's time to find another mine.

* Two of my co-panelists have written their own distinctive takes on the debate. Please check out Threadless.com's Jeffrey Kalmikoff on the spec debate, as well as Forrester analyst Jeremiah Owyang's advice to designers.

Herdict—Using the Crowd to Fight Censorship

Last week Harvard's Berkman Center for Internet & Society launched Herdict, which encourages Internet users around the world to report blocked, or otherwise inaccessible Websites. The name is a portmanteau of "herd" and "verdict," and true to its name, Herdict allows people to track blackouts in order to determine whether the problem is innocuous and temporary, or the result of government censorship. Dig the trailer (Does anyone launch anything without a trailer anymore?)

The project is the brainchild of the author, law professor, and all around brillionaire Jonathan Zittrain, who not only co-founded Berkman ten years ago, but also conducted pioneering research on Internet filtering earlier in this decade. He eventually helped create the OpenNet Initiative (ONI), which tracks online censorship in countries like China, Iran and Uzbekistan. Herdict is an outgrowth of ONI's work. But whereas ONI gathers anecdotal evidence and technical analysis for academic study, Herdict hopes to aggregate massive amounts of user data to create real time reports. It is essentially, if I understand the technology correctly, a public version of the testing software that ONI operatives (my word choice) used to test sites used to monitor government filtering from within the borders of repressive states. Herdict, then, is part of a larger movement to enlist crowds in an effort to create transparency in government. While it can't stop censorship, it can cast a light on it.

The site's designers clearly took to heart one of the most elemental (and most ignored) principles of crowdsourcing: modularity. Translation: Break a task down into snack-sized bites so it can be performed by what one might call the "coffee break labor force." You can test a few dozen sites in less than five minutes. Modularity is what made Google Image Labeler and the NASA Clickworker such a success. It's fun, and it takes something like .003 seconds to contribute. (For you crowdsourcing geeks out there: Berkman fellow Yochai Benkler has written about modularity and how the relationship between the granularity of tasks and a project's ability to elicit participation. See pages 100-103 in The Wealth of Networks.)

To succeed Herdict will depend on large-scale participation—and not just by the folks in the English-speaking blogosphere. To this end Herdict has put out the call to help translate the site into other languages (using the excellent community translation site, dotSUB). If you're bilingual and happen to believe in a free Internet (and really, who doesn't?), pitch in to help put the herd into Herdict.

8020 est Morte! Vive le 8020!

The funeral dirge for newspapers continued as the 150-year-old Rocky Mountain News published its final edition last Friday. The casualty list is sure to grow. The Washington Post's Howard Kurtz summed up the dire signals in his State of the News story today: "Newspapers are killing sections and closing bureaus, particularly in Washington. The Detroit News and the Detroit Free Press have cut back home delivery to three days a week. The Washington Times has dropped its Saturday print edition. The Christian Science Monitor is switching to Web-only publication in April. Gannett Co., publisher of USA Today, is forcing staffers to take a week-long furlough. Hearst plans to close the Seattle Post-Intelligencer unless it gets a buyer."

So it seemed somewhat incongruous to learn that 8020 Publishing—purveyor of the lush, user-generated photo magazine, JPG Magazine—had found an investor that would allow it to keep trucking. Former CEO Mitchell Fox, wrote in an email that a joint venture between camera retailer Adorama and private investors had formed "for the purpose of executing on the unique vision that led to the creation of JPG Magazine, jpgmag.com and everywheremag.com." Was it a fire sale? Possibly. Fox has said 8020 had lots of suitors, but they evidently weren't serious enough to keep 8020 from shuttering earlier this year.

But who cares? The fact remains that JPG—the magazine and the vibrant community around it—provides a robust model for how communities and publishers can collaborate to create compelling content. In these most grim of times, newspapers could do worse than to follow its lead.

The Return of Crowdsourced Border Patrol

One of the earliest (and creepiest) uses of crowdsourcing appeared in Texas where the governor hooked surveillance cameras up to the Internet, allowing anyone with an Internet connection the ability to play virtual border guard. The program failed to attract enough users though, and the governor shuttered the "Virtual BorderWatch"—warning: Texas trademarked the name!—a few months later.

Now it's reappeared, and is getting much the the same wide-eyed press attention lavished on it the first time around. I'll decline to enter the fray over whether this constitutes the triumph of the panopticon. That ship sailed long ago. We have met Big Brother and it is us. But I am interested in whether the Virtual Border Watch can attract enough users to justify the $2 million the state is spending to maintain the site. If it suceeds, this would seem to imply that social production-slash-crowdsourcing-slash-distributed-labor has penetrated the mainstream enough to support even the most skull crushingly boring of its uses. Since the program re-launched in November, 43,000 people have logged in, resulting in the retrieval of 1,500 pounds of marijuana. That seems like a drop in the bucket to me, but then drug war economics are always an exercise in absurdity, so who knows how this is playing in Austin.

Whither Journalism? A Reading List

Journalists have always been a little too ready to write the story behind the story, which is to say, we devote a disproportionate number of column inches to covering our own industry. Stories devoted to the rise and fall of the short-lived Talk magazine alone could fill a small town library. This all seemed like so much self-obsessed froth until last fall, when the news of our own demise seemed, suddenly, not so premature. As ad revenues plunged and venerable papers began declaring bankruptcy or shuttering their print operations, the navel gazing took on considerable gravitas: How do we save ourselves?

Over the last several weeks a number of possible answers to that question have been put forth. I moderated a roundtable in Boulder that will touch on the future of media, among other questions, and I thought it might be useful to try to organize some of the most prominent voices in this conversation into some sort of order. It's worth noting that the mere fact this debate is taking place is a positive sign. I spent much of my time at Wired chronicling the decline of the recording industry, which seemed intent on marching silently into its own dark, digital night. So we in journalism at least have our very public self-regard going for us. Below pls find an incomplete reading list on the future of journalism:

Start with "End Times," Michael Hirschorn's piece from the January issue of The Atlantic Monthly. He set off a firestorm by pondering—persuasively—whether The New York Times could go out of business as early as this May, and what form a world without the Grey Lady might take. "Ultimately the death of The New York Times—or at least its print edition—would be a sentimental moment, and a severe blow to American journalism. But a disaster? In the long run, maybe not."

With this dystopian vision in your head, you'll want some reassurance. Pick up historian Jill Lepore's New Yorker essay, "Back Issues: The Day the Newspaper Died," to be reminded, as Twain said, that while history doesn't repeat itself, it does rhyme. "The last time the American newspaper business got this gothic was 1765, just after the first gothic novel."

The "Micropayments" Argument:

On January 12 the New York Times media columnist David Carr proposed a novel solution to the newspapers' dilemma: Create an iTunes for news content. In summing up the music label's complaint about Apple, Carr writes: "Those of us in the newspaper business could not be blamed for hoping that someone like [Jobs] comes along and ruins our business as well by pulling the same trick: Convincing the millions of interested readers who get their news every day free on newspapers sites that it's time to pay up."

And then a few weeks later no less a luminary than Walter Isaacson endorsed the broad strokes of such an effort in a Time Magazine cover article. "Under a micropayment system, a newspaper might decide to charge a nickel for an article or a dime for that day's full edition or $2 for a month's worth of Web access. Some surfers would balk, but I suspect most would merrily click through if it were cheap and easy enough."

Sounds reasonable, right? Well, no, reasons Clay Shirky, it doesn't. First, he points out that the term micropyaments doesn't even apply, as it refers to payments measured in cents or even fractions thereof. More importantly, it just wouldn't work: "Such systems solve no problem the user has, and offer no service we want. As a result, conversations about small payments take place entirely among content providers, never involving us, the people who will ostensibly be funding these transactions."

And then there was a simultaneous debate occuring over another proposal altogether: Having newspapers mimic non-profit institutions like universities.

The Endowment Argument:

Yale Chief Investment Officer David Swensen and financial analyst Michael Schmidt got the ball rolling on January 28th in a New York Times op-ed entitled, "News You Can Endow." The authors argued that "by endowing our most valued sources of news we would free them from the strictures of an obsolete business model and offer them a permanent place in society." It lent the imprimatur of two respected financial minds to a topic that's been discussed in newsrooms for several years.

Later that day New Yorker writer and former Washington Post managing editor Steve Coll weighed in with some compelling math: "If the Washington Post had a $2 billion endoment, it would be able to fund a very healthy newsroom. And this is before revenue from continuing operations—advertising, circulation, etc. ..."

But this prospect comes with its own considerable problems. Coll returned to the subject a few days later to address some of the concerns brought up in the wake of his original post. For one, there's a limit to the amount of philanthropy dollars available to news organizations—and some of that is already going to support innovative newsrooms like the nonprofit Voice of San Diego. Slate's respected media critic Jack Shaefer poured some cold water on the endowment idea on February 3, pondering why, if these news-gathering institutions are so vital to 'our democratic constitutional system' (Coll) ... not enough paying customers can be found to support them. "There's something disconcerting about wanting to divorce the newspaper from market pressures." This argument received a lot of traction in the blogosphere, whose members would like to see newspapers innovate their way out of their struggles. Easier said then done.

As for me, I have my own, somewhat half-baked idea: Collective licensing. These are generally mentioned in the context of the music industry, the general outlines of the proposal being that ISPs would collect some small monthly sum from its subscribers that would then be doled out to the labels based on complicated formulas relating to usage. The smart money is on collective licensing as the only route to sustaining a label in a world that shows no decline in illegal file-sharing. Which isn't to say it's gained much traction—outside the Isle of Man anyway. I wonder if such a program might be the way to support Internet content broadly.

The Crowdsourcing Book: A Correction

It's been brought to my attention that I made an unfortunate error in the book. In the final chapter I briefly examine a Canadian crowdsourcing firm called Cambrian House, which had intended to use online communities to: A) Generate ideas for new products; B) Determine which of those ideas had merit; and C) Build said products and bring them to market. The crowd turned out to excel at parts A and B, but the company had a difficult time persuading their members to invest the considerable sweat equity involved in launching a new business. And therein lay a lesson about crowdsourcing.

However, I summed up their experience by pondering whether "the failure of Cambrian House signifies the failure of crowdsourcing." (My unequivocal answer, of course, is no it doesn't.) The fact is, Cambrian House didn't fail, even if their original business model did, which is what I meant to imply. The company has shifted to focusing on licensing its software as a sort of crowdsourcing operating system. This shift was underway when I was reporting on the company, but I left that nuance on the cutting room floor. The fact is, I was and remain a big fan of the people behind the company, which embodied some of the best aspects of crowdsourcing, from community respect to total transparency.

I also noted that the company had "sold its assets to the VC-firm Spencer Trask." At the time of writing, such a deal was under discussion. In the end, Spencer Trask only licensed CH's software. Again, my apologies to the good folks at Cambrian House for the error.