Wikifunctions
Wikifunctions is a collaborative, multilingual wiki and free repository of computer functions launched by the Wikimedia Foundation on December 5, 2023, as the organization's first new project in over a decade.[1] It enables users worldwide to create, edit, and maintain a library of reusable code functions for tasks such as calculations, data processing, and natural language generation, supporting Wikimedia projects like Wikipedia and Wikidata while fostering accessible, machine-readable knowledge.[2] As the technical foundation for the broader Abstract Wikipedia initiative, Wikifunctions aims to centralize and democratize computational logic, allowing functions to be composed, executed in sandboxes, and integrated across more than 300 languages without duplicating efforts across language editions.[1]
The project was approved by the Wikimedia Foundation's Board of Trustees in 2020 and initially launched in read-only mode earlier in 2023 before opening for full public editing and contributions.[1] Funded in part by grants from Google.org, the Rockefeller Foundation, and the Wikimedia Endowment, it builds on the success of Wikidata by extending collaborative editing to executable code, including descriptions, parameters, test cases, and implementations in multiple programming languages.[2] Key features include support for pure functions with metadata like runtime estimates, binary input/output handling (e.g., for image analysis), and online execution in browsers or cloud environments, all while ensuring security through sandboxing and auditing.[3]
Wikifunctions complements other Wikimedia sites by providing a shared space for functions that can answer practical questions—such as computing distances between cities or days between dates—and generate content dynamically, ultimately aiming to enhance global knowledge equity by making computational tools editable and reusable like encyclopedia articles.[4] Development remains ongoing, with community involvement encouraged through editing, discussions, and events like Wikimania sessions, positioning it as a fourteenth core Wikimedia project that bridges human editing with machine intelligence.[5]
Overview
Purpose and Goals
Wikifunctions is a free, editable catalog of computer functions designed for reuse across Wikimedia projects and beyond, allowing collaborative creation and maintenance by a global community.[3] Its primary goals include supporting natural language generation (NLG) to produce content in multiple languages, representing abstract content for structured knowledge, and enabling multilingual knowledge creation that transcends language barriers.[2] By centralizing reusable functions with associated test cases and implementations, the project aims to democratize access to computational tools, much like Wikidata does for structured data.[3]
The vision of Wikifunctions is to empower volunteers worldwide to develop functions that address practical questions, such as calculating time differences between cities, distances between locations, or volumes of objects, making these capabilities available in over 300 Wikipedia language versions.[1] As of September 2025, the library contains over 3,000 functions.[6] It has seen initial integrations, such as the first roll-out in a Wikipedia article in 2025.[7] This collaborative approach emphasizes accessibility for non-coders, enabling both humans and machines to edit and contribute without requiring advanced programming skills, thereby fostering a community-led expansion of knowledge pathways.[1][2]
As a foundational component of the broader Abstract Wikipedia initiative, Wikifunctions seeks to integrate computational logic with encyclopedic content to generate dynamic, language-agnostic representations of knowledge.[2] Initial development has been supported by grants from Google.org, the Rockefeller Foundation, and the Wikimedia Endowment.[2][1]
Relation to Abstract Wikipedia
Abstract Wikipedia is an initiative by the Wikimedia Foundation to develop language-independent knowledge structures, enabling the creation and maintenance of Wikipedia articles that can be automatically translated into multiple languages using structured data from Wikidata and executable functions from Wikifunctions.[8] This project aims to address knowledge gaps in smaller Wikipedia language editions by allowing abstract content to be generated once and rendered dynamically across languages, thereby accelerating language equality and expanding access to information globally.
Wikifunctions serves as a core subsystem within Abstract Wikipedia, providing a collaborative library of reusable, executable functions that process and transform Wikidata's structured data into natural language summaries, computations, or other outputs.[2] These functions complement Wikidata's role in storing factual entities and relationships by adding computational capabilities, such as natural language generation, to derive meaningful content from raw data without manual translation efforts.[9] For instance, Wikifunctions enables the creation of renderers that convert abstract content—stored in Wikidata and represented using ZObjects from Wikifunctions—into coherent articles tailored to specific languages.[9]
In the broader ecosystem, Wikifunctions parallels Wikidata, which handles structured data, and Wikimedia Commons, which manages media files, to form a comprehensive knowledge infrastructure for the Wikimedia projects.[9] This integration supports dynamic content generation, where functions from Wikifunctions can invoke Wikidata queries to produce up-to-date, multilingual Wikipedia articles on demand.[2] The project originated from a 2019 proposal led by Wikimedia Foundation engineer Denny Vrandečić, which envisioned Wikifunctions (initially called Wikilambda) as the foundational component for realizing Abstract Wikipedia's vision of a unified, translatable knowledge base.
History
Development and Approval
The development of Wikifunctions originated as a key component of the Abstract Wikipedia initiative, proposed in early 2020 by Denny Vrandečić, the founder of Wikidata, to create a language-independent knowledge base supported by a reusable library of computational functions (initially termed Wikilambda).[10] This proposal built on years of prior discussions within the Wikimedia community about addressing linguistic disparities in content creation across Wikipedia language editions.[11] The project aligned with the broader Wikimedia movement's goals of expanding access to knowledge globally by enabling structured, translatable content generation.[12]
Key milestones included unanimous approval by the Wikimedia Foundation Board of Trustees on May 22, 2020, marking it as the Foundation's first major new project in over a decade.[13] Following the approval, Vrandečić joined the Foundation in July 2020 to lead the effort, with an official announcement of the project on July 2, 2020.[11] Initial development was funded through the Wikimedia Endowment, which awarded $1 million in 2022–2023 specifically for Abstract Wikipedia and Wikifunctions to support prototype building and technical infrastructure.[14]
The development team was primarily based in the Wikimedia Foundation's research and technology departments, under Vrandečić's direction as Head of Special Projects, with significant contributions from volunteer developers and community testers.[15] Early prototypes focused on function schemas and evaluators, culminating in a read-only beta site launched on July 26, 2023, at wikifunctions.org to allow public testing of core features without editing access.[16] Additional funding bolstered this phase, including a $1 million grant from the Rockefeller Foundation in July 2023 and a $3 million grant from Google.org later that month, both earmarked for advancing Wikifunctions' implementation and integration capabilities.[17]
Launch and Early Progress
Wikifunctions officially launched to the public on December 5, 2023, at wikifunctions.org, marking the first new Wikimedia project in over a decade.[1]
In the ensuing months, the project experienced swift initial adoption, with the community creating over 900 functions by late March 2024; these were supported by more than 1,300 implementations and 2,300 tests to ensure reliability.[18] By September 2025, this growth had continued, surpassing 3,000 functions.[6]
A pivotal early development was the rollout of the natural numbers type in March 2024, which facilitated the rapid development of over 100 functions for numerical operations, including addition (Z13521) and the nth Fibonacci number (Z13835).[18][19][20]
Community involvement gained momentum through the initiation of monthly Volunteers' Corner meetings in 2024, conducted via Google Meet to discuss progress and contributions, alongside newsletter subscriptions managed on Meta-Wiki for ongoing updates.[18] These efforts continued into 2025, including regular Natural Language Generation Special Interest Group meetings.[21]
As early as March 2024, the platform encompassed natural language generation functions across an expanded set of languages, such as Breton (with over 60 text-creation functions), Croatian, Igbo, Rohingya, Esperanto, and Japanese, enhancing its utility for multilingual applications.[18][22]
Technical Architecture
ZObjects and Function Schema
ZObjects are the fundamental data structures used in Wikifunctions to represent functions and related entities in a serialized, JSON-like format, analogous to items and properties in Wikidata.[23] Each ZObject consists of key-value pairs where both keys and values are themselves ZObjects, enabling a hierarchical, tree-like representation that supports persistence and querying.[24] These structures are stored in Unicode Normalization Form C to ensure consistent serialization and are identified by unique ZIDs (ZObject IDs) for persistent objects across the system.[24]
At the core of the function schema are ZFunctions, which serve as the primary ZObjects for defining executable functions. A ZFunction includes essential metadata such as labels (primary names), descriptions (human-readable explanations), and aliases (alternative names) available in multiple languages to facilitate global accessibility.[23] Additional metadata encompasses the function's ZID and persistence status, ensuring traceability and version control within the wiki environment. Key components of a ZFunction include argument declarations, which specify inputs with types and identifiers (e.g., global keys like Z13521K1 for the "left" argument); return types, defined as instances of Z4/Type (such as Z10/natural number); and constraints for validation, enforced through type-specific validators to maintain data integrity without side effects.[24]
ZObjects, including ZFunctions, are hosted on wikifunctions.org as part of the WikiLambda MediaWiki extension, with their JSON representations stored directly on wiki pages for collaborative editing and retrieval.[25] This storage integrates seamlessly with Wikidata's query service, allowing ZFunctions to reference and process Wikidata items via SPARQL queries for enhanced knowledge processing.[25]
For illustration, consider the ZFunction for adding two natural numbers (Z13521/add), which declares two arguments of type Z10/natural number and specifies a return type of Z10/natural number. Its structure in normal form might appear as:
{
"Z1K1": "Z8", // type: Function
"Z8K1": { // label
"Z6K1": "en", // language: English
"Z6": "add" // value: "add"
},
"Z8K2": { // description
"Z6K1": "en",
"Z6": "Returns the sum of two natural numbers"
},
"Z8K3": [ // arguments
{
"Z17K1": "Z13521K1", // key: left
"Z17K2": "Z10" // type: natural number
},
{
"Z17K1": "Z13521K2", // key: right
"Z17K2": "Z10" // type: natural number
}
],
"Z8K4": "Z10" // return type: natural number
}
{
"Z1K1": "Z8", // type: Function
"Z8K1": { // label
"Z6K1": "en", // language: English
"Z6": "add" // value: "add"
},
"Z8K2": { // description
"Z6K1": "en",
"Z6": "Returns the sum of two natural numbers"
},
"Z8K3": [ // arguments
{
"Z17K1": "Z13521K1", // key: left
"Z17K2": "Z10" // type: natural number
},
{
"Z17K1": "Z13521K2", // key: right
"Z17K2": "Z10" // type: natural number
}
],
"Z8K4": "Z10" // return type: natural number
}
This schema ensures the function adheres to functional programming principles, with constraints validating that inputs are valid natural numbers before computation.[24]
Implementations, Evaluators, and Testing
In Wikifunctions, implementations provide the executable code for ZFunctions, typically written in programming languages such as JavaScript, Python, or PHP, and stored as separate ZObjects (specifically Z16/Code objects) that are linked to the function's schema.[24] These implementations can include direct code snippets, compositions of other functions, or built-in primitives, allowing multiple versions in different languages to be associated with a single function for cross-verification.[24] For instance, the nth Fibonacci number function (Z13835) features implementations in Python using recursive methods and Binet's formula, demonstrating how code handles sequence generation.[20]
Evaluators serve as secure execution engines that interpret and run these implementations, producing Z22/Evaluation results as ZObjects. The primary evaluator, hosted in a dedicated repository, operates as a RESTful service that processes function calls via post-order traversal of the call tree, supporting lazy evaluation while prohibiting side effects to ensure pure functions.[26] It employs sandboxed environments using Docker containers, AppArmor profiles, and Seccomp filters to restrict system calls, file access, and network operations, thereby preventing malicious code from compromising servers.[26] Computational limits, such as timeouts and resource caps, are enforced to avoid overload, with monitoring handled by the Wikifunctions development team, often referred to as Functioneers.[27]
Testing in Wikifunctions relies on Z20/Tester ZObjects, which define input-output pairs and validators to verify implementation correctness across versions. These testers enable cross-testing, where multiple implementations are run against the same inputs to detect inconsistencies, fostering reliability in the function library.[24] As of January 2025, Wikifunctions hosts over 6,000 such tests across its functions.[28] For the Fibonacci function (Z13835), a representative test case with input n=5 yields output 5, confirming the sequence value (where F(0)=0, F(1)=1, and subsequent terms sum the priors); this test passes for available implementations, illustrating evaluator output as a ZObject result.[20]
Features and Functionality
Creating and Editing Functions
Creating and editing functions in Wikifunctions occurs through a wiki-based interface hosted on wikifunctions.org, where users collaboratively build ZFunctions by defining schemas, adding implementations, and developing tests.[29] The platform employs MediaWiki software, enabling straightforward editing similar to other Wikimedia projects, with options to create new functions via the Special:CreateObject tool by specifying a name, inputs, outputs, and aliases in multiple languages.[29] Multilingual support is integrated throughout, allowing labels, descriptions, and aliases to be added in various languages to facilitate global contributions.[29]
The typical workflow starts with proposing a function idea on the dedicated suggestion page, where community input helps refine its scope before formal creation.[30] Once proposed, users define the function schema, outlining required inputs and expected outputs, then proceed to implement it either by writing code in supported languages such as Python or JavaScript, or through a visual composition mode that assembles existing functions without coding.[31] For instance, a string concatenation function can be implemented visually by chaining primitive operations like addition and spacing.[31] Tests are then created to verify implementations, specifying input values and asserting expected outputs, such as equality checks for strings or booleans, with an integrated testing panel to run and validate them automatically.[29]
To enhance accessibility for non-coders, the interface incorporates guided prompts during schema definition and visual builders for compositions, reducing the need for programming expertise while encouraging broader participation.[31] Edits to existing functions are made directly on their pages by selecting "Edit source," allowing modifications to schemas, implementations, or tests, followed by preview and publication.[29] Versioning mirrors Wikipedia's model, with a complete edit history accessible for each object, enabling users to review changes, discuss proposals on associated talk pages, and revert edits if needed to maintain quality. Community review is integral, as proposed changes can be flagged for feedback before final connection to the function.[29]
Functioneers, a designated user group established upon the project's 2023 launch, play a key role in overseeing edits by approving and deactivating implementations and tests to ensure performance and safety.[32] Under the 2023 policy, these users—numbering 66 as of November 2025—monitor contributions for efficiency, auto-patrol routine edits, and handle connections between components, preventing unverified code from affecting function evaluations.[32] Functions may reference Wikidata items for inputs, such as entity identifiers, to leverage structured data in computations.[2]
Supported Types and Examples
Wikifunctions supports a variety of data types for function arguments and outputs, enabling diverse computations and generations. Core types include natural numbers, which represent non-negative integers and support operations such as equality checks, addition, subtraction, Fibonacci sequence generation, and conversions to Roman numerals.[33][34] Text types handle string manipulation, allowing functions to process and transform textual data, such as concatenating strings or extracting substrings.[33][35] In 2025, new types including Wikidata quantity, geo-coordinate, and time were introduced to better support structured data computations.[36]
Natural Language Generation (NLG) types form a key category, facilitating the creation of human-readable text in specific languages. For instance, functions generate descriptive sentences in Breton, with over 60 such NLG functions developed by 2024 to support text creation in that language.[18] Similar NLG capabilities exist for other languages, including Japanese, where functions handle character-specific manipulations and sentence construction.[22]
Representative examples illustrate practical applications across types. A distance calculator function takes geographic coordinates as input and outputs the distance in kilometers, leveraging natural number and text types for computation and formatting.[35] Volume computation functions for geometric shapes, such as spheres or cylinders, accept dimensions as natural numbers and return calculated volumes.[34] Time difference functions compute intervals between events, using calendar-related types to produce results in days or other units.[37]
As of November 2025, Wikifunctions hosts 3,278 functions, including hundreds of numeric functions covering arithmetic, sequences, and conversions, and over 60 NLG functions in total across languages.[38][18] These functions are reusable in bots or templates, enabling dynamic content generation for Wikimedia projects like automated summaries or infobox calculations.[35]
Linkage with Wikidata
Wikifunctions integrates with Wikidata through a structured data linkage that allows ZFunctions to access and utilize Wikidata's knowledge base for enhanced computational capabilities. ZFunctions and other ZObjects are stored on Wikifunctions with unique ZIDs, but they reference Wikidata items via their QIDs, enabling direct incorporation of encyclopedic and lexical data. This integration supports built-in fetch functions in Wikifunctions for retrieving Wikidata content such as items, statements, and lexemes on-the-fly, with SPARQL querying available on Wikidata's query service for underlying data.[39][40]
The shared schema between the projects relies on Wikidata's property system to define function arguments and parameters consistently. For instance, properties like P585 (point in time) are used to specify dates or temporal data within ZFunction inputs, ensuring interoperability and reuse of standardized qualifiers and references. Bidirectional links further strengthen this connection: Wikifunctions employs reference types, such as the Wikidata item reference (Z6091), to link to QIDs, while ZFunctions can pull Wikidata entities as inputs for processing; conversely, Wikidata items can reference ZFunctions through proposed or emerging properties linking to ZIDs.[39][41]
This linkage provides significant benefits by allowing computations on real-world data from Wikidata, such as using population statistics (e.g., property P1082) to calculate growth rates over time. Since Wikifunctions' launch in 2023, implementation has included embeddable function calls within Wikimedia wiki contexts, facilitating dynamic queries and transformations without duplicating data stores.[39][2]
Applications in Wikipedia and Beyond
Wikifunctions integrates with Wikipedia primarily through Lua modules and templates that invoke ZFunctions, enabling dynamic content generation such as calculations for infoboxes based on Wikidata properties.[42] This approach allows editors to call reusable functions directly within articles, reducing duplication of code across language editions and facilitating updates to complex elements like citations or statistical displays.[42]
Proposed examples include auto-generating article summaries from structured data, where ZFunctions process Wikidata to produce concise prose overviews, and bot-assisted edits that automate computations such as updating election results or demographic figures in real-time.[42] These applications enhance efficiency for editors, particularly in smaller Wikipedias, by leveraging centralized logic to handle repetitive tasks without requiring local scripting expertise. As of 2025, integrations have expanded to include notifications for Wikipedia contributors about relevant Wikifunctions changes (January 2025), live deployment on wikis like Dagbani Wikipedia (April 2025), and embedded function calls accessing Wikidata items (August 2025).[42][5]
Beyond Wikimedia projects, Wikifunctions offers an open API that enables third-party applications to access and execute its functions, supporting uses in education tools for interactive simulations and research platforms for data processing.[43] For instance, functions like distance calculations could integrate into mapping applications, allowing developers to build custom geographic analyses atop Wikimedia's knowledge base.[43] In 2024, user studies involving interviews with contributors from ten language communities, including underrepresented ones like Bengali and Hausa, explored adoption pilots for Wikipedia integration, revealing opportunities to address template inconsistencies across wikis.[42]
The platform's scalability stems from its support for natural language generation (NLG) functions, which enable content creation in approximately 300 languages by composing multilingual templates and logic.[44][45] This multilingual capability ensures that applications can adapt to diverse linguistic needs, promoting equitable knowledge access in global contexts.[44]
Community and Governance
Editing and Collaboration
Wikifunctions draws volunteers from the global Wikimedia community, fostering a diverse group of contributors who collaborate on building and maintaining the project's library of functions. Monthly meetings, known as Volunteers' Corner, have been held via Google Meet since 2024 to discuss progress, share ideas, and coordinate efforts among participants.[18][37]
Collaboration occurs through established Wikimedia tools, including talk pages for project-wide discussions on policies, proposals, and technical issues.[46] Contributors also use Phabricator for tracking tasks and bugs related to development and editing.[47] Regular newsletters provide updates on recent changes, new functions, and community announcements, keeping participants informed and engaged.
New contributors are supported through onboarding resources such as tutorials available on wikifunctions.org, including guides on working with Wikidata content and introductory videos like "Wikifunctions in 7 minutes." Outreach efforts emphasize accessibility for non-coders, highlighting how the project enables collaboration without requiring programming expertise.[48][49][50]
Within the community, specialized roles include Functioneers, users granted permissions to approve or deactivate implementations and test cases for quality oversight. Editing remains open to all registered users, with reputation built through the longevity and impact of contributions, akin to established Wikimedia practices.[32][47]
As of November 2025, the community has created over 3,200 functions by contributors from various linguistic and cultural backgrounds, demonstrating the project's growing collaborative scope. Recent initiatives include a naming contest for Abstract Wikipedia ongoing into late 2025 and the onboarding of new volunteers. The permissive licensing under CC0 for definitions and documentation, alongside Apache 2.0 for code implementations, further supports this open collaboration by allowing free reuse and modification.[37][5]
Policies and Licensing
Wikifunctions operates under a licensing framework that dedicates all function definitions, labels, and documentation in the main namespace to the public domain via the Creative Commons CC0 1.0 Universal Public Domain Dedication, allowing unrestricted reuse and modification without attribution requirements, while code implementations are available under the Apache 2.0 License.[37][51][52] This approach aligns with the project's goal of fostering a freely accessible library of functions, waiving all copyrights to maximize interoperability with other Wikimedia projects and external tools.[53]
Content policies in Wikifunctions emphasize neutrality and verifiability, particularly for function descriptions and associated metadata, drawing from established Wikimedia principles to ensure reliable and unbiased representations.[54] Neutrality requires that descriptions present function purposes and behaviors in a balanced manner, avoiding promotional language or subjective interpretations, while verifiability mandates that claims about function outputs or applications be supported by testable implementations or references to source data, such as Wikidata entries.[54] These policies extend to the determinism guideline, which stipulates that functions must produce consistent outputs for the same inputs to enable caching and predictable behavior, thereby supporting verifiability by minimizing variability from external factors like time or randomness unless parameterized.[55]
The Functioneer role, formalized in community discussions during 2023, grants select trusted users the authority to approve, deactivate, or manage implementations and test cases to maintain function quality and system performance.[56][32] This policy aids load management by allowing Functioneers to intervene in potentially resource-intensive functions, such as those involving heavy computations or external data fetches, ensuring scalability as the project grows.[55][18]
Governance of Wikifunctions is overseen by the Wikimedia Foundation, which provides technical infrastructure and legal framework, while community-drafted guidelines are developed and hosted on Meta-Wiki for cross-project coordination.[2][57] These guidelines, including protections against abuse like page locking for high-risk edits, are enforced through administrator tools similar to other Wikimedia sites.[58]
Dispute resolution follows a consensus-based model akin to Wikipedia's, where editors discuss issues on project talk pages to reach agreement through compromise, with escalation to administrators for blocks in cases of vandalism or policy violations.[59] This process prioritizes civil dialogue and documentation of decisions to uphold community standards.
Inclusivity policies promote multilingual contributions by requiring that project guidelines be drafted in English first—as the authoritative version—before translation into other languages using the Translate extension, enabling global participation across Wikimedia's supported languages.[60] Accessibility is integrated through adherence to MediaWiki's internationalization features and Universal Language Selector, ensuring interface and content usability for diverse users, including those with disabilities, in line with broader Wikimedia efforts.[60][61]
Reception and Future Developments
Impact and Challenges
Wikifunctions has significantly enhanced computational knowledge within the Wikimedia ecosystem by establishing a collaborative repository of reusable functions, enabling automated computations and content generation across projects. As the first new Wikimedia project launched in over a decade, it has fostered innovation in natural language generation (NLG) by powering Abstract Wikipedia's ability to produce multilingual content from structured data.[1][62]
The project has received positive reception for its accessibility, empowering volunteer editors to contribute functions without advanced programming expertise through a wiki-based interface. Coverage of the 2023 launch highlighted its potential to democratize code creation and address knowledge gaps in underrepresented languages. Community growth has been steady, with the catalog expanding to over 3,000 functions by September 2025, reflecting active contributions from editors worldwide.[1]
Despite these achievements, Wikifunctions faces challenges from high computational demands, which have caused server stability issues, including overloads from web crawler traffic in 2024. These were mitigated through enhanced monitoring and bot restrictions by the development team. Adoption remains low among non-technical users due to the steep learning curve and unfamiliarity with function-based workflows.[63][42]
Criticisms center on code reliability in an open-editing model, where unvetted contributions could introduce errors, and integration hurdles with Wikipedia's legacy template systems, which complicate seamless uptake. 2024 progress reports indicate successful language expansions, such as integration with the Dagbani Wikipedia, but reveal slow overall adoption in major Wikipedias due to entrenched editing practices.[42][63]
Planned Enhancements
Wikifunctions is set to expand its natural language generation (NLG) capabilities through the development of new function types, including constructors and renderers for transforming abstract content into target languages. This includes creating monolingual generation functions for grammatical entities such as nouns and verbs, as well as regular inflection generation for Wikidata Lexemes. Plans also encompass renderers for English as a baseline, followed by support for a second major language, a language from a different family, and underserved languages to broaden accessibility.[9]
Deeper integration with Wikidata is a key priority, aiming for completion by 2026, with enhancements allowing Wikifunctions to support over 200 new functions tied to Wikidata entities and the import of statements including qualifiers, potentially covering approximately 50% of encyclopedic content. Integrations with Wiktionaries and low-traffic Wikipedias began in 2025 and continue with expansions to most Wiktionaries.[64][65][66] Improved evaluators and renderers will extend support to additional languages, addressing current limitations in multilingual processing. AI-assisted editing tools are planned to aid content creation while ensuring community control and transparency in generated outputs.[66]
Community outreach efforts include volunteer recruitment for user research and engagement through demonstrations of semantic functions to boost contributions, alongside socialization of specific functions like conjugation tables with relevant communities. A standalone wiki for hosting abstract articles built on Wikifunctions is slated for initial deployment by Q4 2026, providing platform capabilities for scalable, community-driven content. As of November 2025, ongoing developments include a backend rewrite using Rust and the second round of voting for naming the wiki hosting abstract articles.[67][66][52] Policy updates will focus on governance to handle increased scale, drawing from ongoing discussions on licensing and enforcement.[66]
The long-term vision centers on realizing the full Abstract Wikipedia, enabling automated article generation from abstract content to produce multilingual encyclopedia entries at scale. This includes planned launch of a prototype in Q3 2026 and validating platform performance for rich, multilingual outputs by Q4 2026, ultimately closing knowledge gaps across languages through Wikifunctions-powered NLG. Updates on these plans are tracked on the Meta-Wiki Abstract Wikipedia page as of 2025.[66]