SEAM
A shared-memory social application prototype for co-located communities, grounded in the concept of distributed cognition. Through contextual associative responses, it allows knowledge to find people proactively, and permits certain memories to act as instructions that alter system behavior.
Timeline (last 90 days)
SEAM is a community-oriented shared memory system that has been prototyped as a social application.
SEAM's design philosophy is to let knowledge actively find people, rather than making people struggle to find knowledge.
SEAM does not build a centralized knowledge base, but instead relies on social relationships to organize, activate and transmit distributed knowledge.
SEAM members can post any fragment of daily life (a sentence, a thought, an experience), and the system automatically analyzes the associations among these fragments and distills knowledge about the group.
SEAM's reply mechanism is contextualized and associative, distinct from conventional keyword retrieval or vector similarity matching.
SEAM's associative replies expand multi-hop along the chains of association between memory fragments, potentially bringing up people and topics the user has never encountered.
In SEAM, some memories can act as instructions that change the system's behavior, enabling members to program the system.
The research team deployed SEAM in a co-located community, where 46 members engaged in open-ended use for two weeks.
After the two-week deployment, the research team held follow-up workshops with 14 of the members.
SEAM's effectiveness in co-located community scenarios may partly depend on the support of offline relationships, and whether its conclusions generalize to online communities with dispersed members and loose ties still requires further research.
2 more timeline events
All Facts (12)
The two-week deployment scale is limited, and the long-term effectiveness of the SEAM model in larger communities remains to be verified.
50%UnverifiedSEAM's design philosophy is to let knowledge actively find people, rather than making people struggle to find knowledge.
50%UnverifiedSEAM does not build a centralized knowledge base, but instead relies on social relationships to organize, activate and transmit distributed knowledge.
50%UnverifiedSEAM members can post any fragment of daily life (a sentence, a thought, an experience), and the system automatically analyzes the associations among these fragments and distills knowledge about the group.
50%UnverifiedSEAM's reply mechanism is contextualized and associative, distinct from conventional keyword retrieval or vector similarity matching.
50%UnverifiedSEAM's associative replies expand multi-hop along the chains of association between memory fragments, potentially bringing up people and topics the user has never encountered.
50%UnverifiedIn SEAM, some memories can act as instructions that change the system's behavior, enabling members to program the system.
50%UnverifiedThe study adopted a two-week open-ended deployment rather than a controlled experiment, in order to observe spontaneous behavior and long-term interaction patterns in real use.
50%UnverifiedSEAM is a community-oriented shared memory system that has been prototyped as a social application.
50%UnverifiedSEAM's effectiveness in co-located community scenarios may partly depend on the support of offline relationships, and whether its conclusions generalize to online communities with dispersed members and loose ties still requires further research.
50%UnverifiedThe research team deployed SEAM in a co-located community, where 46 members engaged in open-ended use for two weeks.
50%UnverifiedAfter the two-week deployment, the research team held follow-up workshops with 14 of the members.
50%