Question and Answer tools

From [[ CPsquare]], the community of practice on communities of practice.
Revision as of 11:35, 23 January 2012 by Smithjd (Talk | contribs)

Jump to: navigation, search
... part of the technology for communities project,
started off by the authors of [Digital Habitats], Etienne Wenger, Nancy White, and John D. Smith.


This tool allows participants to ask questions and receive answers. Unlike a listserv, it manages the process quite explicitly by routing questions to "experts," and either routing answers back to the questioner or shared with the community.

Uses in communities of practice

For many communities, an e-mail list will do when a member has a question. But for large and complex communities, or when communities make their expertise accessible to outsiders, it becomes useful to have a system that can manage the process more efficiently. Examples: Askme, Google (it's is a subset of google called..???), ask jeeves.


  • Together/apart, Synch/Asynch:
  • Interaction/publication:
  • Individual/group:

Key features

How the system recognizes questions

  • Communities/topics There is an explicit taxonomy of topics or communities from which a questioner must choose
    • When expertise is segmented and the breakdown into sub-specialties is clear enough to make it easier to direct questions to appropriate experts.
  • Parsing questions The system parses the question and does syntactic/semantic analysis.
    • More sophisticated parsing offers better matching chances with FAQ's and routing to experts.
  • Keyword recognition The system uses keywords to interpret questions.
    • Keyword recognition is often good enough.
  • Recognize previous questions The system recognizes questions that have been answered before and offers stored answers first.
    • Important when experts are really busy and would become discouraged if they have to answer the same questions too many times.

How experts are selected

  • System can select expert(s The system can propose some potential expert(s) or even direct the question to the most likely person(s) to respond well.
    • Again, good matches save experts' time. Questions have a better chance of being answered in a timely fashion when experts know that a question has been directed to them with high precision.
  • User can choose expert The user can choose an expert on a list according to some criteria (e.g., expected quality or speed).
    • Useful when one needs to reward experts.
  • Rating answer quality The user is asked to rate the quality of the answer(s) according to some scale.
    • Useful when one needs to reward experts and give them feedback, and also for ranking experts for further selection when directing questions.
  • Rating answer speedThe speed of the response is noted.
    • Same as above. Sometimes a questioner wants an answer right away, even if it is not the best. In such cases, knowing the avarage speed of response becomes an important criterion for selecting an expert.
  • Self-declared expert: People can declare themselves as experts on topics.
    • Useful to achieve an initial pool of experts.
  • Build expert profile The system builds a profile of experts according to the ratings they get on their responses.
    • Important to be able to target questions with increasing precision over time.
  • Moderation: Certain questions can be routed to moderators.
    • Checks and balances will improve overall performance of the system

What the system does with answers

  • Follow-up on questions: The system monitors questions that have not been answered and takes action.
    • People whose questions are not answered within a certain time may become discouraged with the process.
  • Store/archives/organizes answers: The system stores, archives, and organizes answers for later use or perusal.
    • Can build a knowledge base, but usually requires some human intervention

Related material