Prediction markets are only effective if a lot of people participate. What are the best ways to encourage more traders and trading within internal company prediction markets?

The Google case on prediction markets argues that monetary gain is not very effective at getting people to participate in prediction markets.  I think money can be persuasive but the scale has to be sufficient.  Google's $10,000 pool was simply to low to motivate many to participate.  What did motivate the participants were the prizes that improved their reputation among their peers.  Simple t-shirts were more important to the winners of  the "10 most active trader" awards than the $1,000 prizes.  However if the awards were $50,000 or $100,000  or if there were the ability to earn money without winning an award, motivation to participate would certainly have been higher. 

In the context of your typical large company, setting up a prediction market with cash rewards sufficient to motivate is not likely to be an option.  For this reason, non-monetary motivation will be the most effective motivating strategy.  From Google's example, participants seemed to be motivated to participate because of the bragging rights they would earn by winning one of the prizes.  Googlers valued the awards and were proud to win the awards.  The challenge for anyone who sets up a prediction market will be creating a sense of value to any awards provided.  Participation in the market needs to be exciting and people need to be motivated to participate.  Companies can setup non-monetary awards like Google did with t-shirts and quarterly announcements of  winners.  Companies can also try to make participation a part of the corporate culture.  Management can help to instill the need to participate by participating themselves and encouraging their teams to participate.  Weekly emails can be sent out which summarize the ongoing results of various markets and the trending of those markets to help get people interested in the life of the markets. 

Companies should also not limit their marketplace to company relevant markets.  Google created markets for the world cup and NBA Finals.  Non-work related market opportunities should help keep employees interested in checking out the various markets that are available.

What are the similarities and differences between a community-driven product development process and a traditional product development process within a firm?

Traditional product development processes require a strategic plan that sets in motion research into ideas to satisfy the needs of the customers targeted by that strategic plan.  Once a product idea is determined, typically approvals of that product idea and the various features are acquired.  This includes the allocation of a budget to develop the product.  Typical traditional product development processes would then include prototype-evaluate-test-implement-test.  Finally the product is ready to market and sell...and the business must cross its fingers and hope for a winning product solution.  The process, however, can be lengthy and require many managerial signoffs before deployment of the product can take place.  The costs for traditional development are often substantial even though only a small number of individuals might contribute to the product's development.

Community based development often still includes prototype, evaluate and test phases but doesn't include most of the other aspects of traditional development.  Since the community provides their talents and effort, managerial approvals to transition from one stage of a project to another do not exist.  A significantly larger set of individuals will typically contribute to the development process of a community based project than in traditional product development.  If there is compensation for a community based project, this compensation is either small or limited to a few contributors(for instance the winner of a contest).  This limited compensation makes the cost of community based development often lower than traditional development.  Because of the sheer number of contributors, development time can often be shorter as well.  In addition the community will often provide extremely effective testing and market validation results.  If the product design and implementation makes it through the community development process, it by definition satisfies the needs of the community and likely satisfies the needs of the target consumers as well. 

LinkedIn recently hit 100M users. If you were in charge at LinkedIn, what would be your strategic goals for the next several years? How would you achieve them?

Clearly LinkedIn needs to prepare for the inevitable razing of the garden walls.   Since the time of this case, Facebook (among other sites) has opened up a login api that permits other sites to receive authentication credentials from Facebook.  Once authenticated, the sites also have access to some profile and network information.  Sites then list Facebook information on their own pages.  This helps provide even more access to Facebook and drives more traffic to Facebook. 

LinkedIn can follow a similar model.  They may want to restrict access to this API to only those organizations that pass a certain "Professional in Nature" test but the additional exposure will clearly drive more traffic to their site and increase the value of their service. 

Another area of seeming opportunity is on the Job Postings front.  LinkedIn seems poised to take control of the job postings/job search market.  Their user base already represents most of the companies in the US and I have heard that most, if not all, recruiters now use linkedIn heavily during recruiting efforts.  LinkedIn could take a larger chunk of this market without major efforts in terms of feature development.  A heavy marketing campaign may be able to win over a large portion of that market.

How do Wikipedia’s processes for creating and modifying articles ever lead to high-quality results? In other words, since anyone can easily edit Wikipedia, how is it that good (and usually accurate) content emerges?

I think the primary factor at play that helps produce generally solid results is the ratio of "good" contributors to "bad" contributors.  By "bad" I don't necessarily mean contributors with bad intentions.  Those folks certainly qualify as bad.  But I'm also referring to those contributors who provide erroneous or inaccurate information with no malice intended.  The bottom line is that the results of the contribution are of poor quality.  As long as there are significantly more contributors interested in accuracy and factual information than there are contributors that deteriorate the quality of entries, wikipedia will generally be an accurate encyclopedia.  Therefore wikipedia essentially depends on that ratio favoring the "good" contributors.  What that ratio actually needs to be to ensure quality entries, is a good question!