Wikimedia CH/Grant apply/Cite Unseen in 2026

Infodata

Name of the project: Cite Unseen in 2026
Amount requested: 30,000-40,0000 CHF
Type of grantee: Individual
Name of the contact: Kevin Payravi (User:SuperHamster)
Contact: kevinpayravigmail.com

Biography

Kevin Payravi (User:SuperHamster) is a software engineer from the United States, and has been editing Wikimedia projects since 2007. In addition to day-to-day editing, Kevin has developed a number of tools to help wiki readers and editors, including Cite Unseen, Commons Gallery, View it! (2024 Coolest Tool Award winner!), and Indie Wiki Buddy. He also serves on the Board of Wikimedia DC and as an organizer for WikiPortraits and the Ohio Wikimedians User Group. Each year, he helps organize WikiConference North America and Wiki Loves Monuments in the United States. More information and a list of previous grants is available on Payravi's CV.

Quinn Gao (User:SuperGrey) is a Taiwanese software developer currently residing in California. He is a JRPG enthusiast and an active contributor to the WikiProject Video Games on Chinese Wikipedia. He has also developed several user scripts on Chinese Wikipedia that have been adopted by many of the platform's active editors. These gadgets include the Reaction gadget (allowing Wikipedia users to add emoji reactions to talk page messages) and the Voter gadget (allowing Wikipedia users to more easily vote in discussions). Quinn spearheaded the adoption of Cite Unseen as a gadget on the Chinese Wikipedia, the first language version of Wikipedia to do so. Quinn also implemented support for localization on Cite Unseen, making it a truly international tool.

The problem and the context

What is the problem you're trying to solve?

Wikipedia is built on citations, and a lengthy article can have hundreds of them. A standard Wikipedia citation will include basic metadata such as author, title, publication, and date. However, there is often little else to indicate the nature or reliability of the source. The English Wikipedia, for example, maintains an extensive list of perennially discussed sources and their reliability, but this data is not natively surfaced in the UI when viewing a list of citations. Similar lists exists on over a dozen other language versions of Wikipedia, including Farsi, French, and Swedish. There are other interesting and important indicators to consider for each source, such as whether a cited New York Times is a news article or opinion piece, or whether the cited article is sponsored by an external party. These sort of classifiers provide important context and can even help indicate the reliability of a source, but these details are not readily surfaced in our existing citation systems.

In 2025, Wikimedia CH funded the extension of Cite Unseen (see grant). That support allowed us to:

Recruit a second code maintainer and make the project more sustainable.
Migrate our codebase to Wikimedia GitLab, adopt a modular architecture, and create a deployment pipeline.
Move all categorization lists to Meta-Wiki for collaborative editing, with annotations and provenance support.
Add internationalization through TranslateWiki.
Make Cite Unseen run universally across all language versions of Wikipedia.
Add major new features:
- Filtering dashboard
- In-article category suggestion flow
- In-article settings UI
- Ability to apply categorizations by date, publisher, and author
Expand coverage by importing additional source lists from multiple WikiProjects and language versions of Wikipedia.

These 2025 outcomes have laid a strong technical and organizational foundation. The feedback has been positive, with some users commenting:

"Also, thanks ... for making me aware of Cite Unseen, this is actually going to help me a lot with the en:WP:AFC process."
"I love the new citeunseen update"
"I like the changes made to Cite Unseen. The header bar makes finding specific references a lot easier"
"I love it! It is really helpful for patrolling/reviewing."

Users have also recognized and appreciated Wikimedia CH's support for Cite Unseen. A user on Discord exclaimed, "Cite Unseen, now supported by Wikimedia 🇨🇭!"

In October 2025, the Chinese Wikipedia became the first language Wikipedia to formally adopt Cite Unseen as a gadget, enabling logged-in users to enable the tool with one-click in their settings.

With the infrastructure built out and early adoption spreading, the central challenge now is scale: building contributor capacity, expanding coverage in non-English languages, expanding and importing source lists, and socializing the tool widely across the Wikimedia movement. The project is promising and gaining traction, and we would like support to ensure Cite Unseen continues to sustainably grow and have the support it needs to continue to scale and become used across language versions of Wikipedia.

What is your solution to this problem (please explain the context and the solution)?

Building on the 2025 foundation, this 2026 project will continue to have software development, but will also have more focus towards people, process, and content scale. Our main goals will be to:

Recruit and support contributors across multiple languages, including German, French, Italian.
Systematically import and normalize source lists from across more language versions of Wikipedia and external partners.
- We are exploring working with and using domain categorizations vetted external organizations such as NewsGuard and the Trust Project, where licensing permits.
Feature development, particularly through feedback from the community.
Expand outreach through presentations and workshops at both Wikimedia and external conferences to demonstrate Cite Unseen, gather feedback, and recruit new contributors.

Project goals

Adoption & Localization
- Ensure Cite Unseen is positioned as a truly global tool. By the end of 2026, we ambitiously want to see the script translated into at least 20 different languages.
- Drive adoption by users in at least 50 language versions of Wikipedia, making Cite Unseen a visible and recommended tool across many communities. Our goal is that readers and editors everywhere, regardless of language, can benefit from consistent citation context.
- Establish Cite Unseen as a top-tier user script globally, with at least 1,000 active installs. Beyond numbers, the aim is for the tool to be recognized and recommended across communities as part of a regular editor's toolkit.

Coverage
- Expand coverage of sources beyond English and Chinese so that at least 10 languages have substantial categorization lists, including major media outlets and commonly cited sources in those regions.
- Build sustainable systems to regularly update and maintain these lists.
- Explore partnerships with external organizations and datasets (such as NewsGuard or academic institutions) to enrich our categorization lists from authoritative sources.

Community & Sustainability
- Continue to establish a broad and active contributor base that can maintain categorization lists in multiple languages.
- Foster strong community awareness and trust in Cite Unseen. By presenting at major Wikimedia events and engaging directly with contributors, we aim to make Cite Unseen a visible and collaboratively maintained part of the Wikimedia ecosystem.

Technical & User Experience
- Continue improving the user experience based on feedback.
- Ensure all new features are fully localized and accessible.

Project impact

How will you know if you have met your goals?

Cite Unseen becomes a recognizable and trusted tool across many language Wikipedias, seen as part of the normal workflow for editors and patrollers who work with citations.
Editors and readers feel more empowered to understand the reliability and nature of sources at a glance, helping identify potentially unreliable citations more quickly.
Communities in multiple languages take ownership of maintaining and expanding source categorizations, ensuring that the project reflects global communities and isn't dependent on only a few contributors.
The user experience feels seamless.
Third parties utilize Cite Unseen's categorization data, showing that the work has value beyond the script itself.

Do you have any goals or metrics around participation or content?

Installations & reach: ≥ 1,000 active installs globally; adoption in ≥ 50 language Wikipedias.
Coverage: At least 6 language communities with major source coverage (hundreds of categorized local sources for each). This is possible to be larger with more funding.
List growth & quality: Dozens if not hundreds of categorization suggestions and additions per month across numerous languages.
Contributor recruitment: ≥ 75 unique contributors suggesting and adding categorization lists across languages.
Events & outreach: Presentations and workshops at Wikimania, WikiConference North America, and potentially other Wikimedia and external events.

Project plan

Activities

Building on the technical foundation from 2025, our 2026 work will focus on scaling adoption, expanding source coverage, and continuing to invest in our sustainability, all on an international scale. The work will unfold in three major workstreams:

1. Contributor recruitment and training

Onboard more volunteers across various language Wikipedias (German, French, Italian, and others) to help maintain source lists, and even the code.
- Points-of-contact per language can help onboard new contributors and be a point-of-contact for their language communities.
Expand documentation in multiple languages to better show users how to edit source lists, contribute new categorizations, and use the in-article features.
Outreach and onboarding at conferences, such as Wikimania, WikiConference North America, and potentially others.
Work with Wikimedia affiliates to localize outreach materials and recruit contributors.
Publish blog posts on Diff to highlight progress and invite participation.

2. Source list imports and normalization

Explore and import new source reliability and classification tables from at least 5 additional language Wikipedias.
- As part of this, we may be able to help facilitate more languages developing their own local source lists.
Normalize these datasets into a unified schema (publisher, reliability rating, type, etc.).
Explore and integrate vetted external datasets such as NewsGuard or university journalism indexes.
- Using external datasets may require collaboration, as some datasets may be protected behind paywalls or copyright.

3. Software maintenance and UX improvement

Continued joint maintenance by our two active developers.
Implement community-requested refinements.
Continued support for all the various language versions of Wikipedia as adoption grows internationally.
Ensure full localization of all UI strings via TranslateWiki.

Budget

Total: 30,000-40,000 CHF

The table below suggests a budget given a 30,000 CHF grant. Given a major focus of the grant is scaling and increasing adoption, the grant can be easily adjusted (with greater funding meaning more investment in internationalization, adoption, source list expansion, etc.).

Category	Amount (CHF)	Description & Deliverables
1. Software development and maintenance	12,000	Continued part-time compensation for at least two maintainers to handle feature development, code review, deployment, and infrastructure. Features will be developed primarily based on feedback from users across language Wikipedias.
2. Contributor coordination & training stipends	6,000	Support for organizing workshops, onboarding sessions, and mentorship. Includes honoraria for community trainers, small participation incentives, and production+translation of tutorial documentation (and potentially video).
3. Categorization micro-grants	6,000	Small grants/bounties to support editors creating and reviewing categorization lists in underrepresented languages. Encourages local expertise and diversity of data.
4. Partnerships & dataset evaluation	3,000	Pilot integration works with external datasets (e.g., NewsGuard, academic media-bias lists). Covers time to facilitate partnerships, potential legal work, and gather/clean data.
5. Administration / fiscal sponsorship	3,000	Fiscal sponsor processing, accounting, and reporting (≈ 10% overhead) plus domain / infrastructure hosting and minor administrative expenses.
Total requested: 30,000 CHF

Community engagement

Where: Meta-Wiki, Village Pumps, WikiProjects, various Wikimedia mailing lists, and Wikimedia conferences.
How: Presentations and workshops at Wikimania, WikiConference North America, and other regional language-focused Wikimedia events.
Who: Various language versions of Wikipdia, including German, French, and Italian communities.