Computing Reviews
Today's Issue Hot Topics Search Browse Recommended My Account Log In
Review Help
Search
Promise theory : principles and applications
Burgess M., Bergstra J., CreateSpace Independent Publishing Platform, 2014. 218 pp. Type: Book (978-1-495437-77-9)
Date Reviewed: Aug 28 2014

Today, computers routinely do many jobs that people used to do, and thus need to interact with human stakeholders. To facilitate these interactions, researchers have devoted considerable attention to formalizing the basic mechanisms by which people interact with one another, and embodying these in computer agents. Examples of this research include formal studies of intention, commitment, negotiation, social norms, and trust. In each case, the research agenda starts with a concept that has an intuitive meaning in interactions among people, and then seeks to formalize it so that computers can behave in a way that is consistent with those intuitions.

Bergstra and Burgess offer a detailed analysis of one such metaphor, the notion of “promise,” which they define as an autonomous declaration of intended but unverified behavior. In Part 1, they describe promises informally, and then offer a formalism that distinguishes them from impositions (messages intended to induce voluntary cooperation in other agents) and obligations. They describe common patterns of promises, explore how the fulfillment of a promise may be assessed, and explore their mechanics and dynamics. This section also discusses how to reason about promises, what valuations may be attached to them, and how promises relate to the notion of trust.

In Part 2, they show how promises can support practical applications, including workflow management and transport networks (including physical and digital systems). They suggest how promises can support the emergence of organization, their relation to componentization and modularity, and their role in game theory. The book concludes with a description of CFEngine, a widely used open-source computer configuration system based on the notion of a promise.

The main value of the book is suggested by this last chapter. As an exposition of the mechanisms that underlie a system used by hundreds or thousands of networks, the volume will help users of CFEngine understand and apply the tool more effectively, and will be a valuable guide to members of the open-source development community in extending the tool consistently. But the book does not position itself as a guide to CFEngine. It aspires to offer a new metaphor for human–computer interaction, an aspiration that is frustrated by internal inconsistencies and lack of external context.

Internally, though the book attempts to develop a mathematical formalism of promises and promise structures, the notation is opaque and the discussion often inconsistent. For example, a promise is defined as a declaration of intention, but page 71 suggests that an agent can “internally promise” something that it does not disclose, a contradiction of the definition. Again: a commitment is defined (page 51) as “a promise to which one is committed,” a circularity that yields little insight. Again: a deception is defined as the combination of “a documented intention (that is, a promiser) and a nondocumented commitment,” (page 52), yet a commitment has already been defined as a promise, and thus necessarily documented. Until we have consistent definitions of intention, promise, commitment, and how these relate to one another, we can hardly claim to have a theory of promises.

These internal tensions might be greatly reduced if the authors interacted more carefully with the considerable literature that already exists in modeling intentions, commitment, and trust. This literature is mentioned only in passing, with two references that do not capture the rich body of insight and formalism that already exists. A good starting point is the foundational work of Phil Cohen and Hector Levesque 25 years ago on intention, choice, and commitment [1]. The authors’ discussion of related literature on trust is more extensive than of the literature on commitment, but is still confined to a single endnote.

In spite of these shortcomings, the book is useful in proposing a focus on the intuition of a promise as a central building block in multiagent interaction. A revision of this volume that engaged the existing literature more carefully would be an important contribution to the field.

More reviews about this item: Amazon

Reviewer:  H. Van Dyke Parunak Review #: CR142668 (1412-1033)
1) Cohen, P.; Levesque, H. Intention is choice with commitment. Artificial Intelligence 42, (1990), 213–261.
Bookmark and Share
  Featured Reviewer  
 
Multiagent Systems (I.2.11 ... )
 
 
Network Management (C.2.3 ... )
 
Would you recommend this review?
yes
no
Other reviews under "Multiagent Systems": Date
Engineering intelligent hybrid multi-agent systems
Khosla R., Dillon T. (ed), Kluwer Academic Publishers, Norwell, MA, 1998. Type: Book (9780792399827)
Aug 1 1998
Linguistic geometry: from search to construction
Stilman B., Kluwer Academic Publishers, Norwell, MA, 2000.  395, Type: Book (9780792377382)
Jan 1 2001
 Transactional agents: towards a robust multi-agent system
Nagi K., Springer-Verlag New York, Inc., New York, NY, 2002.  205, Type: Book (9783540430469)
Apr 13 2004
more...

E-Mail This Printer-Friendly
Send Your Comments
Contact Us
Reproduction in whole or in part without permission is prohibited.   Copyright 1999-2024 ThinkLoud®
Terms of Use
| Privacy Policy