Reid Priedhorsky – Publications

Table of contents

• First-author peer-reviewed papers
• Middle-author peer-reviewed papers
• Abbreviations

First-author peer-reviewed papers

1. Reid Priedhorsky, Jilin Chen, Shyong (Tony) K. Lam, Katherine Panciera, Loren Terveen, John Riedl. “Creating, Destroying, and Restoring Value in Wikipedia”. In Proc. GROUP 2007.

   • Acceptance rate: 28%
   • Fulltext: PDF, ACM master. (Note: our PDF contains one minor correction not in the ACM version.)
   • Key contributions:
     • This work was the first to compute the value of Wikipedia edits and the impact of damage, using reader-based measures rather than author-based ones, which created a much firmer foundation for wiki research than was available previously.
     • Discover who produces Wikipedia’s value: 1/10th of 1% of editors contribute nearly half of its value, measured by words read.
     • Quantitatively categorize the different types of wiki damage.
     • Measure the impact of damage (e.g., wiki vandalism): 42% of Wikipedia vandalism is repaired essentially immediately (within one estimated view).

2. Reid Priedhorsky, Benjamin Jordan, Loren Terveen. “How a Personalized Geowiki Can Help Bicyclists Share Information More Effectively”. Short paper. In Proc. WikiSym 2007.

   • Acceptance rate: 50%
   • Fulltext: PDF, ACM master
   • Key contributions:
     • Formalize the notion of geowiki (an online map which allows editing of all geographic features displayed and supports standard wiki monitoring features) and introduce the notion of personalized geowiki (a geowiki which can customize itself according to individuals’ needs).
     • Analyze the information needs of bicyclists and show that they need an information resource which is comprehensive, up-to-date, and personalized — properties met by a personalized geowiki.
     • Show that the geowiki model is plausible: for example, 73% of subjects were willing to spend at least five minutes correcting a map error if the result would be available to other cyclists immediately, but only 44% if availability would take six months.

3. Reid Priedhorsky and Loren Terveen. “The Computational Geowiki: What, Why, and How”. In Proc. CSCW 2008.

   • Honorable Mention in the conference Best Paper awards.
   • Acceptance rate: 23%
   • Fulltext: PDF, ACM master
   • Key contributions:
     • Invent the notion of computational geowiki, a geowiki where user contributions feed into an algorithm (in our case, route finding).
     • Present the design and implementation of a computational geowiki (early Cyclopath), producing significant design innovations and solving important implementation challenges.
     • Show quantitatively that geowiki and computational geowiki features form a powerful knowledge sharing tool in a representative domain, bicycling, suggesting the general utility of geowikis and computational geowikis.

4. Reid Priedhorsky, Mikhil Masli, and Loren Terveen. “Eliciting and Focusing Geographic Volunteer Work”. In Proc. CSCW 2010, forthcoming.

   • Acceptance rate: 20%
   • Fulltext: PDF
   • Key contributions:
     • Devise two techniques to elicit and focus user work in a geowiki or similar system, one using familiarity to direct users to work opportunities and another visually highlighting them.
     • Show quantitatively the success of the techniques in eliciting work and the nature of the distribution of work across users.
     • Explain factors that contributed to the success of the techniques, for example that certain types of work require users to be familiar with the location of the work but others do not.
     • Show that user work meaningfully benefits the community — in particular, user work since Cyclopath went live has shortened the average route in Cyclopath by 1 kilometer.

Middle-author peer-reviewed papers

1. Pam Ludford, Reid Priedhorsky, Ken Reily, Loren Terveen. “Capturing, Sharing, and Using Local Place Information”. In Proc. CHI 2007.

   • Acceptance rate: 25%
   • Fulltext: PDF, ACM master
   • Key contributions:
     • Show that location-based reminders — i.e., task reminders delivered when the user is near places relevant to the task — yield new local knowledge about those and related places.
     • Identify heuristics people use when deciding what place-related information to share; for example, people are more likely to share content they believe would be useful to others or about places they recommend, and unlikely to share content containing names or about private places like homes or workplaces.
     • Offer design guidelines for local knowledge sharing systems; for example, such systems could compute and display usefulness metrics or detect and offer to suppress names.
     • Identify new uses of shared place information, most notably opportunistic errand planning.

2. Michael Ludwig, Reid Priedhorsky, and Loren Terveen. “Path selection: A novel interaction technique for mapping applications”. In Proc. CHI 2009.

   • Acceptance rate: 25%
   • Fulltext: ACM master
   • Key contributions:
     • Present a novel technique for user selection of routes within a graph, called path selection, based on dynamic shortest-path computation and continuously updated visual feedback.
     • Show quantitatively the benefits of path selection in comparison to state-of-the-art techniques: faster route selection, fewer errors, and greater user satisfaction.
     • Discover the properties of routes that make path selection particularly beneficial: long straight segments and a close correspondence between algorithmically optimal and visually attractive nodes.

3. Katherine Panciera, Reid Priedhorsky, Tom Erickson, and Loren Terveen. “Lurking? Cyclopaths? A Quantitative Lifecycle Analysis of User Behavior in a Geowiki”. In Proc. CHI 2010, forthcoming.

   • Acceptance rate: 22%
   • Key contributions:
     • Quantitatively analyze the lifecycle of users in an open-content system, including portions of the lifecycle which occur before account registration.
     • In particular, we show that users who eventually register do so soon after first encountering the site; i.e., users generally do little or no “educational lurking”.
     • Show that geographic locations viewed (private traces) are associated with locations edited (public traces), and this association grows stronger as a user's edit count increases. This has potentially interesting consequences for privacy and also for work elicitation techniques.

Abbreviations

• CHI — ACM International Conference on Human Factors in Computing Systems. This is the premier global conference in the field of human-computer interaction.
• CSCW — ACM Conference on Computer Supported Cooperative Work. This is the premier global conference in the field of computer-supported human collaboration.
• GROUP — ACM International Conference on Supporting Group Work.
• WikiSym — ACM International Symposium on Wikis and Open Collaboration.