Grants:IdeaLab/Teahouse question query tool
Project idea[edit]
What is the problem you're trying to solve?[edit]
Thousands of new editors register accounts on English Wikipedia every day. About 100 of these new editors are invited to the Teahouse by a bot every day, and others find their way they by other means, but the Teahouse can't easily be scaled to support every new editor.
A 2012 survey of Teahouse participants indicate that new editors find the Teahouse Q&A board to be a valuable resource. Recent research has provided the strongest evidence to date that the Teahouse has a positive overall impact on new editor retention.
In order to scale this impact, we need to make valuable features of the Teahouse accessible to more editors. Particular groups who are not well served by the Q&A board include:
- New editors who do not meet the 10-edits-in-first-24-hour threshold for receiving an automated Teahouse invitation
- New editors who are hesitant to ask questions at the Teahouse (for whatever reason)
- New editors who are likely to lose patience or interest in editing because they are unable to find the information they need immediately
- New editors who prefer to learn independently
- New editors who edit primarily on mobile devices (for which the Teahouse Q&A board is not optimized)
What is your solution?[edit]
Leverage the 12000+ (and growing) Teahouse question archive and the ElasticSearch API to allow people to type in their questions, and see a list of related questions pulled from the archive. The search box could be implemented as a Gadget that is turned on by default for all new users, and/or through a stand-alone web application hosted on Labs.
The basic scenario we want to support is kind of like how new programmers use sites like StackOverflow. You think of a question, you type it into a search box, find that 10 other new programmers have asked similar questions, and you scan through the results to find the best answer. We know that the questions in the Teahouse archive have high-quality, detailed answers, they are phrased the way newcomers would phrase them, and are more likely to be up to date and easy to understand than Wikipedia's help documentation.
The Teahouse Q&A data would probably have to be scraped from the wiki and cleaned up a little bit to facilitate better searching. Relevant metadata such as the question date, the number of answers, could be extracted from the question to facilitate search filtering. Links to policy/help pages could be extracted and presented separately as supplemental support material.
Another important design consideration is support for mobile-first editors. More and more new editors are editing primarily from mobile devices. It is difficult (at best) to participate in the Teahouse Q&A forum from a mobile device. A mobile-optimized self-service application may provide a better experience for mobile-first users who seek answers to common newbie questions.
Goals[edit]
- allow more new editors to benefit from Teahouse Q&A without overwhelming the Teahouse with more questions than the volunteers (and the Q&A page itself) can support.
- experiment with new use cases for ElasticSearch within Wikimedia.
- provide a more lightweight, mobile-friendly way of getting help for users who are unable or uninclined to ask questions at the Teahouse.
Get Involved[edit]
Participants[edit]
- Advisor I can advise on this project. Jmorgan (WMF) (talk) 18:27, 7 December 2015 (UTC)
- Advisor I can do technical advisory stuff! YuviPanda (talk) 22:31, 7 December 2015 (UTC)
Endorsements[edit]
- Allowing editors to search for their answer directly would prove to be pretty valuable here. The Teahouse has been open for business for quite sometime, and similar questions do routinely surface there. With this tool, new editors could immediately obtain useful answers to their questions. I JethroBT (talk) 21:10, 9 December 2015 (UTC)
- FAQ needs a refresh; need a friendly way to answer common questions; should not have to be hard to search tl;dr Slowking4 (talk) 14:52, 29 February 2016 (UTC)
Expand your idea[edit]
Would a grant from the Wikimedia Foundation help make your idea happen? You can expand this idea into a grant proposal.