Crowdsourcing

Editor's Notes

• #5: (Linux) Threadless Chipolte Technically even American Idol.. Who selects the winner. Starbucks asks their employees for feedback. Implemented ideas are recognized and suggestor is given a T shirt (or something)
• #7: Actual Practice Started about 10 years ago.
• #8: Amazon Actually has users do a variety of low key low paying jobs. Pays change per transaction.
• #9: Describe screen some.
• #10: Need a coder to help you create the next 'killer app' or help you with a tough problem?  Just post your program or problem here and coders from around the world will email you bids on doing the work.     You can review each bidder's resume and reputation online, and when you're ready to make your decision, you can rent  your personal, expert, coder with just a few clicks! Its that simple!      Posting a bid request is free , and you are under no obligation to accept any bids sent to you unless you choose to.   Once you accept a bid, you will place your payment into escrow. The money is not released to the coder, till they complete the work according to your origina 1) Receive bid request from buyer (or find bid request online) 2) Make a bid 3) Do the work.File weekly status reports. Send end product to buyer. 4) Buyer approves work 5) Receive payment 6) Rate Buyer
• #13: People seeking supplemental income (moms, dads, college students. retirees
• #14: Money is a powerful motivator (more on that in a bit) Obama Campaign, the people running the campaign were surprised in how many people were asking how they could help and speak on behalf of the campaign.
• #15: Began in 2006 Ended in Summer 2009 10% improvement in their search results The Ensemble was a comgolmeration of several teams… hungry, Pa, New york, San Diego.. “ The crowd is indeed wiser than the individual .” But the real winner, already, is the burgeoning field of so-called prize economics — the idea that crowdsourced contests can produce better, cheaper results than a direct, concerted effort. Out of the contest’s chaotic early days emerged two highly-organized entities, each with collaboratively-produced algorithms significantly better than the best that Netflix’s engineers could come up with. Funny, how these things have a way of working themselves out. [i ] [i] http://www.wired.com/epicenter/2009/08/netflix-prize-continues/
• #16: Began in 2006 Ended in Summer 2009 10% improvement in their search results The Ensemble was a comgolmeration of several teams… hungry, Pa, New york, San Diego.. “ The crowd is indeed wiser than the individual .” But the real winner, already, is the burgeoning field of so-called prize economics — the idea that crowdsourced contests can produce better, cheaper results than a direct, concerted effort. Out of the contest’s chaotic early days emerged two highly-organized entities, each with collaboratively-produced algorithms significantly better than the best that Netflix’s engineers could come up with. Funny, how these things have a way of working themselves out. [i] [i] http://www.wired.com/epicenter/2009/08/netflix-prize-continues/
• #17: I – Prize form Cisco [i ] Cisco wanted an idea that would earn them $1billion dollars Prize for 1st place? - $250k and chance at a job. 1170 ideas submitted Winners were from Germany and Russia. (plan to optimize electricity via networking) Because of web 2.0 “you can get more ideas faster from more areas with far less work and resources” - Matthew Greeley CEO Brightidea (hosted the contest) [i] http://newsroom.cisco.com/dlls/2008/ts_071408.html
• #18: Search for Steve Fossett [i ] After adventurer was found missing Richard Branson Talked to Google for Google Maps Google Maps talked to their providers then Amazon. Amazon Mechinacl turk outsourced the dearch to thousands of internet users. Even though it didn’t work. The effort was groundbreaking. People from the other side of the world what about the future? A volunteer based collaborative Internet search could allow for more comprehensive search and rescue efforts when search budgets do not allow for highly advanced and expensive automated computer analysis of aerial photography. [i] http://www.npr.org/templates/story/story.php?storyId=14332329
• #19: White House solicited ideas from citizens at opengov.ideascale.com [ i] [ii ] Several Good Ideas however these were very “high” on the list. Remove Marijuana from Schedule I of the Controlled Substances Act  Verifying eligibility to be president of the USA  Open all classified UFO files. Lesson: Know your audience before asking them to do something. [i] http://opengov.ideascale.com/akira/ideafactory.do?id=4049&mode=top&discussionID=2294 [ii] http://www.nytimes.com/2009/06/23/technology/internet/23records.html?_r=1
• #21: 1. BE FOCUSED Vaguely defined problems get vague answers. Current TV is explicit about the goal of its viewer-created ad messages (v-cam's): to develop new advertising to run online and on the cable network. And it allows companies such as L'Oreal or Sony (SNE ) to mine the v-cams for fresh advertising ideas. The more infrastructure you build into the creative process, the more success you will have. Current TV has clear rails to support social networkers. It provides a specific focus such as the one for Sony Ericsson's Walkman Phone: "With a Sony Ericsson Walkman Phone you're always connected to your music. The question is: where will it take you?" It also provides a Sony Ericsson logo and a smattering of graphics, as well as instructions that include a time limit -- up to three minutes -- and a deadline -- September 8. Current TV also spells out directly in its rules that the network retains the power to reject anything that paints the brand in a negative light. 2. GET YOUR FILTERS RIGHT Crowdsourcing often produces a wealth of ideas, and companies need effective filters to pick the gems. Consider IBM's (IBM ) innovation jam, a two-part brainstorming session launched in July designed to tap the collective minds of employees, family members, and customers to target potential areas for innovation. CEO Sam Palmisano will put $100 million into promising ideas. IBM identified four large themes, providing interactive background information on each one, employing moderators to keep conversations focused, and setting a 72-hour time limit for the first session. By the end of it, IBM had collected 37,000 ideas. IBM will use its own crowd to filter the ideas. The company has made transcripts available to the 140,000 people who logged in to the first session and teams will review the posts. In early September, the company will host a second session, where everyone will again log on to the jam session to vote on the ideas with the most potential. Then senior executives will sift through this short list to make recommendations about which should be funded. Palmisano will have a hand in making the final choices. 3. TAP THE RIGHT CROWDS At YouTube and probably within any new social network, only 1% of the users are active content creators. Another 10% interact with the content and change it. The remaining 89% passively observe. Smart companies want to assemble the crowds with the most sophisticated knowledge about their business problems to maximize the impact of the small percentage of idea generators within them. Consider InnoCentive, a social network created by Eli Lilly (LLY ) where companies like Procter & Gamble (PG ) and Boeing (BA ) can pay a steep fee to post the knotty problems they can't solve internally -- like a process for the extraction of trace metal impurities, for example. The idea is that individual problem solvers -- retired scientists, obsessive hobbyists, university students -- might be able to lend a hand. If they solve the problem, they receive a hefty cash reward. This network is dependent on a crowd of extremely talented scientists with highly specific skills. To attract them, InnoCentive recruits at universities, where young, smart minds have not yet entered the workforce. To date, the network has signed agreements with 25 Chinese universities, including the prestigious Chinese Academy of Sciences and the National Natural Science Foundation of China. The network also promotes itself at industry events and advertises in trade publications. Also key, Innocentive's rewards for solutions are sizable. Crowdsourcing is not cheap. Just as in corporate America, top talent is expensive, and companies will have the most luck when they are willing to pay up for the inventions that lead to potential innovations. 4. BUILD COMMUNITY INTO SOCIAL NETWORKS Cash is key to getting people to participate, but successful crowdsourcing taps into a well of passion about a product that stretches beyond monetary incentives. Cambrian House, a software company founded by Michael Sikorsky, relies entirely on crowdsourcing. Everyone who contributes an idea receives royalty points, which function as equity in the final product and can eventually be cashed in for stock. But Sikorsky has also created glory points, which reward members who collaborate. He says one key element in keeping people engaged long-term is for them to build friendships with other members. It's not so different from the eBay (EBAY ) model, where buyers and sellers rate each other and offer commentary. These interactions foster trust and keep people active in the community.