Today, beyond language preferences, there are few ways for users to express fine-grained content preferences to Web servers for them to adapt, personalize, or customize content. Without user-preference tools, sites like simple.wikipedia.org resort to providing separate URLs instead of providing different views of content for the primary URL.
It would both benefit users and provide websites with new opportunities to better meet users' needs if users could express a greater number of content-customization preferences such as reading level, language fluency, and background knowledge. Artificial intelligence is making it easier for websites to customize content, creating new opportunities to meet finer-grained preferences.
The mission of this group is to explore and discuss mechanisms for users to express content-customization preferences. This group will explore and discuss topics including artificial intelligence, adaptive hypermedia, adaptive explanation, adaptive learning, adaptive instructional systems, and user modeling.
Note: Community Groups are proposed and run by the community. Although W3C hosts these
conversations, the groups do not necessarily represent the views of the W3C Membership or staff.
Chairs, when logged in, may publish draft and final reports. Please see report requirements.
The Adaptation and Personalization Community Group intends to explore and discuss technologies including those with which to enable adaptive science news.
Science news connects society to scientific and technological progress and achievement. It simultaneously benefits multiple audience demographics: students, the general public, and scholars and scientists.
For students, science news can promote scientific literacy, inspire curiosity, and promote critical-thinking skills. Science news can encourage students’ enthusiasm about science, technology, engineering, and mathematics.
For the general public, science news can inform, mitigate misinformation, and facilitate the appreciation of science as a shared human endeavor.
For scholars and scientists, science news can be read to remain abreast of unfolding discoveries and developments.
Adaptation and personalization technologies for science news websites, then, would greatly benefit society and, therefore, in addition to adaptive encyclopedias, should be considered a use case of interest to our group.
Instead of content authors having to choose one intended audience to write their articles expressly for, content authors could utilize artificial-intelligence-enhanced adaptation and personalization technologies to adapt their articles’ contents, the essences or substances of their articles, into multiple variant articles for multiple intended audiences.
To browse and participate in our discussion area, please visit here.
The Adaptation and Personalization Community Group intends to explore and discuss technologies including those with which to enable adaptive encyclopedias.
With adaptive encyclopedias, end-users would be able to browse and navigate vast repositories of knowledge, finding that each article they viewed was adapted and personalized to their individual settings and configurations (or to proximate user-modeling stereotypes).
These settings and configurations would, as considered, include end-users’ reading levels, language fluencies, and background knowledge.
With respect to encyclopedia article authoring and editing, as envisioned, audience-independent inputs would be authored for software components to use to create prompts or agentic workflows for artificial-intelligence systems to use to generate audience-specific article variants.
These approaches would be entirely compatible with features and functionalities for collaborative, crowdsourced article editing. Articles’ editors would simply modify those ingredients which audience-specific article variants were to be generated from, perhaps while being able to view selected output article variants on-screen, in near-real-time, as they edited.
While, today, some of these features can be delivered for individual websites using existing and emerging techniques, it is envisioned that end-users would want to be able to set, configure, adjust, and calibrate their settings and configurations and to be able to choose to have these simply work across websites, across encyclopedias, technical documentation, science news articles, educational resources, and many other compatible sites across the Web. That is, it is envisioned that end-users would want to store and control their user-model data.
To browse and participate in our discussion area, please visit here.
For whom should large-scale repositories of knowledge, e.g., Wikipedia, phrase their content: article subject-matter experts, laypeople, or students? Similarly, what about technical documentation? Digital textbooks?
What if content authors could provide multiple intended audiences – differing with respect to their reading levels, language fluencies, and background knowledge – with multiple interrelated variations of Web resources?
What if content authors could provide software with styled content outlines and this software would formulate prompts for and interactions with artificial-intelligence systems to generate natural-language content for multiple intended audiences?
What if end-users could dynamically adjust their fine-grained content-related preferences by adjusting one or more “adaptation parameters” to maximize the subjective readability and comprehensibility of Web resources for themselves?
The Adaptation and Personalization Community Group intends to explore and to discuss these and many more related questions!
Technical topics of interest to our group include artificial intelligence, adaptive hypermedia, adaptive explanation, adaptive learning, adaptive instructional systems, and user modeling.
To browse and participate in our discussion area, please visit here.
Today, beyond language preferences, there are few ways for users to express fine-grained content preferences to Web servers for them to adapt, personalize, or customize content. Without user-preference tools, sites like simple.wikipedia.org resort to providing separate URLs instead of providing different views of content for the primary URL.
It would both benefit users and provide websites with new opportunities to better meet users’ needs if users could express a greater number of content-customization preferences such as reading level, language fluency, and background knowledge. Artificial intelligence is making it easier for websites to customize content, creating new opportunities to meet finer-grained preferences.
The mission of this group is to explore and discuss mechanisms for users to express content-customization preferences. This group will explore and discuss topics including artificial intelligence, adaptive hypermedia, adaptive explanation, adaptive learning, adaptive instructional systems, and user modeling.
This is a community initiative. This group was originally proposed on 2026-03-25 by
Adam Sobieski. The following people supported its creation: Aldo Gangemi, Adam Sobieski, Milton Ponson, Frances Gillis-Webber and Wolfgang Wimmer.
W3C’s hosting of this group does not imply endorsement of the activities.
w3c-cg/adaptation
