Rick Wash

By Rick Wash and Jeff MacKie-Mason

Social computing systems collect, aggregate, and share user-contributed content, and therefore depend on contributions from users to function properly. However, humans are intelligent beings and cannot be programmed to behave; system designers must provide incentives to encourage users to contribute. We explore the behavioral consequences of one simple incentive mechanism: require users to contribute a minimum amount of information before they are granted access to the system. Users with a high marginal cost of contribution will stop using the system, but users with a moderate marginal cost will increase their contribution, frequently leading to greater benefits for everyone still using the system. Additionally, if contributions are collaborative and build upon each other, then existing contributors are likely to slightly decrease their contributions, leading to a more ’equal’ distribution of contributions. We show that this mechanism often leads to increased contributions, and provide concrete design advice for using this mechanism in social computing systems.

Rick Wash and Jeff MacKie-Mason. “Using a Minimum Threshold to Motivate Contributions to Social Computing,” Working Paper, June 2009.

Download: PDF

By Rick Wash

Piracy of digital goods has become a large problem on the Internet due to its ability to distribute large quantities of content to large quantities of people at very low cost. As a result, many digital content producers have turned to technology to insure their revenue stream. These technologies, collectively known as Digital Rights Management (DRM) technologies, use encryption and other security technologies to artificiallyrestrict the uses of a digital good.

I develop a idea that was first observed by Acquisti (2004). In that paper, Acquisti develops a model of platform adoption in the presence of network effects. The DRM technology makes transferring between two networks (open and DRM networks) possible, but one direction is costless (open -> DRM) and the other costly (DRM -> open). In addition to other results, Acquisti derives a result that popular content is more likely than niche content to be transferred to the open platform if it is initially only available on the DRM platform. He therefore concludes that the success of a DRM platform depends on `user-generated’ content (as opposed to `widely-popular vendor-generated’ content.

I use this idea as a basis for studying incentives for innovation in a content industry that uses DRM technologies. I study the situation where there is a negative network effect in the costs of `breaking’ the DRM, or transferring the content from the DRM platform to the open platform. As content becomes more popular, the average cost of breaking decreases since the content will more likely fall into the hands of people capable of breaking it, hackers will be more interested in breaking it because it can be shared farther, or there exist returns to scale to breaking DRM technologies. In this situation, I hypothesize that content producers will prefer high-value `niche’ content over low-value `mass-market’ content when using DRM. Here I studythe implications of such a preference.

Rick Wash. “Content Provision and Digital Rights Management.” Working Paper, December 2005.

Download: PDF

by Rick Wash

In these notes I explain symmetric key additive keystream ciphers, using as an example the cipher RC4. I discuss a number of attack models for this class of ciphers, using attacks on RC4 as examples. I cover a number of attacks on RC4, some of which are effective against implementations of RC4 used in the real world.

Download: PDF, PS