Boundary Object

From P2P Foundation
Jump to navigation Jump to search

Boundary Object = An object that can attract shared understanding and engagement from different communities.


"Artifacts, Documents and perhaps even vocabulary that can help people from different communities build a shared understanding. Boundary objects will be interpreted differently by the different communinities, and it is an acknowledgement and discussion of these differences that enables a shared understanding to be formed." (from the enTWIne project


Matt Moore links Social Objects to Boundaries:

"Social objects remind me of the "boundary objects" than Denham Grey outlines. Boundary objects are narrower in scope as they allow different groups to communicate with each other where as Jyri's social objects could mediate between those who know each other well."

Denham Grey quotes:

"Boundary object (BO), originally introduced by Starr (1989), is a concept to refer to objects that serve an interface between different communities of practice. Boundary objects are an entity shared by several different communities but viewed or used differently by each of them. As Star points out, boundary objects in an organization work because they necessarily contain sufficient detail to be understandable by both parties, however, neither party is required to understand the full context of use by the other - boundary objects serve as point of mediation and negotiation around intent.

Boundary objects are flexible enough to adapt to local needs and have different distinct identities in different communities, but at the same time robust enough to maintain a common identity across the boundaries to be a place for shared work. Boundary objects are not necessarily physical artifacts such as a map between two people: they can be a set of information, conversations, interests, rules, plans, contracts, or even persons."


From an article on Object-oriented Sociality at

By Susan Leigh Star:

The structure of ill-structured solutions: boundary objects and heterogeneous distributed problem solving

The concept of boundary objects as presented below thus is simultaneously metaphor, model, and high-level requirement for a distributed artificial intelligence system p.44

Ø Real world information systems are distributed and decentralized, they evolve continuously, embody different viewpoints, and have arms-length relationships between actors requiring negotiation p.45

Ø The information in an open system is thus heterogeneous, that is, different locales have different knowledge sources, viewpoints, and means of accomplishing tasks based on local contingencies and constraints

Ø Boundary objects are objects that are both plastic enough to adapt to local needs and constraints of the several parties employing them, yet robust enough to maintain a common identity across sites p.46

Ø There are different types of boundary objects depending on the characteristics of the heterogeneous information being joined to create them

§ Study of neurophysiologists – the case is the unit of analysis for clinicians, whereas for basic researchers it is analytic generalizations about classes of events p.47

§ Study of biologists – the specimen itself is the unit of analysis for the amateur, whereas for the professional biologist the specimens form part of an abstract generalization about ecology, evolution, or the distribution of species p.48

four types

Found four types of boundary objects p.48-51 (not an exhaustive list)

Ø Repositories – ordered piles of objects that are indexed in a standard fashion

§ Built to deal with problems of heterogeneity caused by differences in unit of analysis

§ Have the advantage of modularity

Ø Platonic Object (Ideal Type) – A map or atlas that does not accurately describe the details of any one location

§ Abstracted from all domains – fairly vague – a means of communicating and cooperating symbolically

§ Have the advantage of adaptability

Ø Terrain with coincident boundaries – common objects with the same boundaries but different internal contents

§ Arise in the presence of different means of aggregating data and when work is distributed over a large-scale geographic area

§ Have the advantage of resolution of different goals

Ø Forms and labels – devised as methods of common communications across dispersed work groups

§ Standardized forms to fill out at highly distributed sites

§ Result in standardized indexes and immutable mobiles (convey unchanging information over long distances)

§ Advantages of such objects are that local uncertainties are deleted

§ May or may not come to be part of repositories



Summary and Conclusions p.51

Ø The creation of boundary objects both respects local contingencies and allows for cross-site translation

Ø Problem solving in the contexts described above produces workable solutions that are not … well-structured. Rather, they are ill-structured: they are inconsistent, ambiguous, and often illogical. Yet they are functional and serve to solve many tough problems in distributed artificial intelligence." (

Key Book to Read

Peter Morville. Ambient Findability. O'Reilly, 2005

quote: "As we build our Internet of objects, the permutations of sociosemantic metadata will create new avenues of findability. Where has this object been? Which objects were in close proximity to this object? Who touched my object? Where are they now? The era of ambient findability will overflow with metadata, as every object and location sprouts tags: social and semantic, embedded and unembedded, controlled and uncontrollable. Imagine the sensory overload of a walk in the park. Every path shimmers with the flow of humanity. Every person drips with the scent of information: experience, opinion, karma, contacts. Every tree has a story: taxonomies and ontologies form bright lattices of logic. Desire lines flicker with unthinkable complexity in this consensual hallucination of space and non-space, a delicious yet overwhelming sociosemantic experience."