Throughout the duration of the conference, we are hosting a hackathon for attendees to work on developing solutions for problems in the credibility space (on Wikimedia platforms and beyond).
Thanks to our partner Credibility Coalition, we will also have funding available to provide microgrants for groups that seek to continue working on credibility-related projects coming out of the event.
- On Friday, please go to Museum of Fine Arts to register. Starting at 10:30, we will move to a separate room at MFA. (Address TBA)
- If you have any questions, please email Stephen LaPorte at email@example.com.
Saturday - Sunday
- On Saturday - Monday, we will be at the MIT Stata Center (room TBA)
We've developed the following challenges for individuals or groups to rally around to ideate and/or create solutions for. These are simply ideas to help inspire you; by no means do you have to stick to one of these!
If you have your own challenge ideas, you are more than welcome to expand the list!
Challenge 1: Replacing unsuitable sources
The English Wikipedia maintains a list of perennial sources that are frequently discussed, and includes a list of sources that are deprecated and that editors are cautioned against using. Even so, many of these sources are still used in articles.
Can we develop a solution to make the replacement of some of these sources easier? If a statement is sourced to a deprecated source, how can we make it easier for editors to find suitable, more credible replacements?
Challenge 2: Making sites citation-friendly
Wikipedia uses Citoid to automatically generate references from just a given URL. However, data extraction is not always successful; for example, of the top 90 most cited news sources, the date is extracted correctly only 60% of the time, and authors 35% of the time.
How can we increase the accuracy of Citoid? Can we build a standard or list of recommendations for news organizations to use to make metadata extraction easier?
Challenge 2b: Help site owners see if their site is citation-friendly
Provide a tool that lets people input a link and see the structuted data extracted by Citoid from it. Take the API call (https://en.wikipedia.org/api/rest_v1/data/citation/mediawiki/…?action=query&format=json) and display the response in a way that they can see what and how well the metadata has been extracted.
- Andrew Lih
Challenge 3: First MediaWiki contribution
MediaWiki is the software used by Wikipedia, Wikidata and friends, written in PHP. This challenge is for people to make their first contribution improving the MediaWiki software, and set up a testing version of the software in order to test that change. A list of suggested things to work on is available here.
Challenge 4: PaceTrack for campaigns!
Let's build PaceTrack, a quick and easy tool for Wikipedians to track their improvement campaigns.
Challenge 5: Automated user behaviour documentation tool
Let's build a tool to gather objective data for complaints of harassment, vandalism and paid editing that involve large numbers of pages and/or users. Requires determination of useful variables and recommendations for reviewer guidelines.
Working on a project? List it below!
- Scholia has a diverse range of issues
- We’re working on a platform where you can filter the information you receive using the judgement of the people you trust! If you’re interested in the project, you can contact me at firstname.lastname@example.org.
- Wikispore and Event Spore + Fact Spore
- Using Wikimedia SUL to log into Wikispore (and other wikis) - T110460
With the support of Credibility Coalition, we will be offering microgrants for groups that seek to continue working on credibility-related projects coming out of the event. Grants will range from between $500 to $10,000, depending on the scope and nature of the project.
Applications will open near the end or after the conference.