Monografias em Ciência da Computação

2005

ABSTRACTS

Departmento de Informática 
Pontifícia Universidade Católica do Rio de Janeiro - PUC-Rio
Rio de Janeiro - Brazil


This file contains a list of the technical reports of the Departmento de Informática, Pontifícia Universidade Católica do Janeiro - PUC-Rio, Brazil, which are published in our series Monografias em Ciência da Computação (ISSN 0103-9741), edited by Prof. Carlos Lucena. Please note that the reports not available for download are available in their print format and can be obtained via the e-mail below.
For any questions, requests or suggestions, please contact:
Rosane Castilho bib-di@inf.puc-rio.br

Last update: 30/DECEMBER/2005

INDEX


[MCC01/05]
FILIPPO, D.; ENDLER, M.; FUKS, H. Colaboração móvel com realidade aumentada. [Title in English: Augmented reality with mobile collaboration] 33 p. Port. E-mail: endler@inf.puc-rio.br

Abstract: In close relationship with Virtual Reality, Augmented Reality is a recent but promising research field with application in several human activities as medicine, education and trading. Augmented Reality systems are those who offers a real-time interactive interface, in which real world objects perceived by the user are superimposed with virtual ones. When associated to Mobility and Collaboration, these applications became still more powerful. This works presents a general view of Augmented Reality and show the main features of research prototypes that includes Mobility and Collaboration.

[MCC02/05]
MEDEIROS, A.P.; SCHWABE, D.; FEIJÓ, B. Design rationale for model-based designs in software engineering. 18 p. Eng. E-mail: bruno@inf.puc-rio.br

Abstract: In this paper we present the Kuaba Ontology, a vocabulary for Design Rationale described in an ontology definition language that allows attributing semantics to recorded Design Rationale content, and defining rules that enable performing computable operations and inferences on this content. This vocabulary extends the argumentation structure of the Issue Based Information System (IBIS) enriching this argumentation structure by explicating the representation of the decisions made during design and their justifications, and the relations between the argumentation elements and generated artifacts. It consists also of interating this argumentation structure with descriptions of the produced artifact, and with information about the design history. In addition, we propose to support the design process through the use of the semantic descriptions defined by formal models of the artifacts. Representing Design Rationale using an ontology definition language and the artifacts formal models, enables a type of software reuse at the highest abstraction level, where rationales are re-employed in designing a new artifact. This kind of reuse is possible in knowledge domains where there are formal models describing the artifacts, in particular, in
the Software Design domain.

[MCC03/05]
BORNSTEIN, C.; LABER, E.S.; MAS, M.A.F. A logarithmic competitive mechanism for unit-demand combinatorial auctions. 3 p. Eng. E-mail: laber@inf.puc-rio.br

Abstract: This paper studies the unit-demand combinatorial auction problem in which each of the n consumers can purchase at most one item from a set of k items. We obtain a randomized truthful mechanism that attains an expected competitive ratio of O(log(min{k,n})) with high probability.

[MCC04/05]
HAZAN, C.; STAA, A.v. Análise e melhoria de um processo de estimativas de tamanho de projetos de software. [Title in English: Analysis and improvement of software projects size estimation] 31 p. Port. E-mail: arndt@inf.puc-rio.br

Abstract: Documented estimation constitutes the basis for the construction of trustworthy software project plans. The size estimation of software projects is used to derive the cost, effort and schedule estimations. Several models have been used by industry to estimate effort and cost given system or artifact size estimates, for example COCOMO. However, the methods to estimate the size of a project have gotten little attention. This work analyzes, from the perspective of CMMI, methods used to generate software projects size estimations using Function Points. It then proposes a new method for estimating size based on experiments. Finally, we describe a simplified process for generating and documenting estimations regarding size, effort, budget, schedule and required critical computer resources.

[MCC05/05]
SARDINHA, J.A.R.P.; NOYA, R.C.; SILVA, V.T.; MILIDIÚ, R.L.; LUCENA, J.C.P. A combined specification language and development framework for agent-based application engineering. 20 p. Eng. E-mail: milidiu@inf.puc-rio.br

Abstract: Software agents are used to solve several complex problems. To properly build agent systems, it is necessary to create software engineering processes and tools to support all the development phases. Indeed, many modeling languages and implementation frameworks are available for system engineers. However, there are few methods that properly combine agent-oriented design and implementation initiatives. In this paper, we propose a development method that goes from the requirements elicitation to the actual implementation of agent systems using a modeling language and an implementation framework. We also present a case study to illustrate the suitability of our approach.

[MCC06/05]
AMARAL, F.N.; HAEUSLER, E.H. A logic-based model for (Meta)Heuristics. 15 p. Eng. E-mail: hermann@inf.puc-rio.br

Abstract: We present a structural model for (meta)heuristic search strategies for solving computational problems. The model is defined through the use of topos-theoretical tools and techniques which provide an appropriate internal logic (with the language of Local Set Theory) where objects of interest can be represented.

[MCC07/05]
CIARLINI, A.E.M.; POZZER, C.T.; FURTADO, A.L.; FEIJÓ, B. A logic-based tool for interactive generation and dramatization of stories. 18 p. Eng. E-mail: furtado@inf.puc-rio.br

Abstract: A key issue in interactive storytelling is how to generate stories which are, at the same time, interesting and coherent. On the one hand, it is desirable to provide means for the user to intervene in the story. On the other hand, it is necessary to guarantee that user intervention will not introduce events that violate the rules of the intended genre. This paper describes LOGTELL, a tool we have been developing for interactively generating and dramatizing stories. We focus on the specification of a formal logic model for events and characters' behaviour. Based on the model, the user can interact with the tool at various levels, obtaining a variety of stories agreable to individual tastes, within the imposed coherence requirements. The system alternates stages of goal inference, planning, user intervention and 3D visualization. Our experiments have shown that the system can be used not only for entertainment but also to help in the creation and adaptation of stories in conformity with a specified genre.

[MCC08/05]
MAIA, R.F.; RODRIGUES, V.J.S.; ENDLER, M. ORB para dispositivos móveis em redes sem fio. [Title in English: Object Request Brokers - ORB for mobile wireless network] 25 p. Port. E-mail: endler@inf.puc-rio.br

Abstract: This technical report describes and discusses the recent efforts towards standards and implementations of Object Request Brokers for wireless networks. In particular, it describes the Wireless CORBA standard, an implementation of this standard called MIwCO, as well as other CORBA middlewares to mobile computing.

[MCC09/05]
BRANDÃO, A.A.F.; SILVA, V.T.; LUCENA, C.J.P. A model driven approach to develop multi-agent systems. 14 p. Eng. E-mail: lucena@inf.puc-rio.br

Abstract: The Object Management Group's (OMG) Model Diven Architecture (MDA) initiative has been the focus of research in academia and industry because it raises a fast and consistent software development through the use of software models. In this context we propose a model driven approach to the development of multi-agent systems beginning with an ontology for the multi-agent systems (MAS) domain. The approach aims to combine these two emerging research areas in order to overcome the complexity associated with the development of MAS.

[MCC10/05]
BRANDÃO, A.A.F.; SILVA, V.T.; LUCENA, C.J.P. Multi-agent system design verification using knowledge-based reasoning. 10 p. Eng. E-mail: lucena@inf.puc-rio.br

Abstract: This paper presents a knowledge-based approach to the verification of the design consistency of multi-agent systems modeled by using MAS-ML, which is a multi-agent system modeling language. We provide a formal description of the MAS-ML diagrams as well as a reasoning engine that are used together to detect intra-model and inter-models inconsistencies. Intra-model inconsistencies are detected based on the specification of each diagram provided by the MAS-ML formal description. Inter-models inconsistencies are identified based on the interdependencies between the diagrams described in the reasoning engine.

[MCC11/05]
FEIJÓ, B.; BADARÓ,P. Proposta de um sistema brasileiro para produção de conteúdo digital. [Title in English: Proposal of a Brazilian digital content development  system] 17 p. Port. E-mail: bruno@inf.puc-rio.br

Abstract: This paper presents the concept of visualization and its relation with the strategic sector of digital content development, which involves audiovisual pieces and software. The technical aspects in Digital TV, films, and high definition are treated in an integrated and clear way. Two complementary models, called Model of Integrated Subchannels and Model of Integrated Regional Axes, are presented as the solution for a number of problems of the audiovisual production chain in Brazil, such as: interactivity; regionalization; boom of demand for digital content; Channel Brazil - Mercosul; trainning of labor force. Also this article warns about the serious mistake commited by the government in the Digital TV program when it does not identify needs before searching for solutions. Finally, the paper claims that a medium and long term industrial policy should be pursued aiming at the formation of industrial clusters in visualization.

[MCC12/05 - Classified]
SOARES, L.F.G.; RODRIGUES, R.F. Levantamento do estado da arte em sincronismo de mídias. [Title in English: State of the art on media synchronization] 110 p. Port. E-mail: lfgs@inf.puc-rio.br

Abstract: This report presents a state of the art on temporal and spatial synchronization in multimedia/hypermedia presentations, focusing on the interactive digital television scenario. The main temporal and spatial sychronization paradigms are presented, as well as the conceptual data models for the applications and the main existent alternatives for synchronization specification. We conclude by presenting a proposal to the Brazilian Digital TV System (SBTVD).


[MCC13/05]
BORNSTEIN, C.; LABER, E.S.; MAS, M.A.F. On behalf the seller and society: a bicriteria mechanism for unit demand combinatorial auctions. 23 p. Eng. E-mail: laber@inf.puc-rio.br

Abstract: This work focuses on obtaining truthful mechanisms that aim at maximizing both the revenue and the economic efficiency (social welfare) for the unit-demand combinatorial auction problem (UDCAP), in which a set of k items is auctioned to a set of n consumers. Although each consumer bids on all items, no consumer can purcharse more tham one item in the UDCAP. We present a framework for devising poly-time randomized competitive truthful
mechanisms that can be used to either favor economic efficiency or revenue.


[MCC14/05]
FIGUEIREDO, E.M.L.; STAA, A.v. Avaliação de um modelo de qualidade para implementações orientadas a objetos e orientadas a aspectos. [Title in English: Object-oriented and aspect-oriented implementations quality model assessment] 29 p. Port. E-mail: arndt@inf.puc-rio.br

Abstract: This paper aims at assessing, identifying weaknesses and proposing enhancements to the quality model used by Sant'Anna to compare properties of object-oriented and aspect-oriented software systems. The assessment includes an empirical study and a qualitative analysis of the metrics used by Sant'Anna's in his model. We present a revised quality model attempting at reducing the weaknesses of the original one. The new model contains a minimal set of metrics, some of which complement the original model. All metrics are defined by means of a process that could be automated.


[MCC15/05]
RODRIGUES, P.S.L.; POZZER, C.T.; FEIJÓ, B.; VELHO, L.; CIARLINI, A.E.M.; FURTADO, A.L. An expressive talking head narrator for an interactive storytelling system. 18 p. Eng. E-mail: bruno@inf.puc-rio.br

Abstract: This paper addresses the problem of telling - or, in a broader sense, presenting and exhibiting - stories based on sequences of events previously computed in an interactive fashion. A scheme is proposed which, in addition to the participating 3D virtual actors, makes use of an expressive talking head system, represented by an avatar with facial expressions, who acts as an animated Narrator of the story to the attending spectators. The implemented system integrates modules for managing plot generation, user interaction, visualization and narration. Scene
rendering as well as the Narrator's speech and facial expressions are real-time synchronized during plot dramatization. The resulting environment may be used in the development of applications involving chidlike worlds, interactive TV, games, and distance learning.


[MCC16/05]
HAZAN, C.; FUKS, H.; LUCENA, C.J.P. Avaliação do tamanho funcional de ferramentas de E-learning. [Title in English: Functional size assessment of e-learning tools] 28 p. Port. E-mail: hugo@inf.puc-rio.br

Abstract: E-learning tools are key factors to the success of Web-based Education. There has been substantial research evaluating such tools on criteria regarding the quality of the finished product. However, little has been said so far about the quality and productivity of the tools' development process. This paper proposes the Function Point Analysis (FPA) method as a means to functional size assessment of e-learning tools. It also presents how the method was used and the results collected in the evaluation of the AulaNet learningware.


[MCC17/05]
FURTADO, A.L.; CIARLINI, A.E.M.; FEIJÓ, B.; POZZER, C.T. Conceptual modelling for storytelling (with a case study). 42 p. Eng. E-mail: furtado@inf.puc-rio.br

Abstract: How to characterize a literary genre is a much debated problem, which can be approached with useful results by combining models drawn from both Literary Theory and Computer Science. Once a genre is specified with some rigour in a constructive way, it becomes possible not only to determine whether a given plot is a legitimate representative of the genre, but also to generate such plots, an ability of obvious relevance to Storytelling theory and  practice. A conceptual modelling method with this purpose is presented, based on a plan recognition/plan generation paradigm. The method leads to the formulation of static, dynamic and behavioural schemas, expressed in temporal logic, and to multi-stage interactive plot generation, supported by a prototype tool. The tool is of special help for the composition of plots through the adaptation of plots fully or partially generated in automatic mode. A case study, involving a simples Swords and Dragons genre, illustrates the discussion.


[MCC18/05 - classified]
SOARES, L.F.G.; RODRIGUES, R.F. Nested Context Model 3.0 - Part 1: NCM Core. 32 p. Eng. E-mail: lfgs@inf.puc-rio.br

Abstract: This technical report describes the basic entities of Nested Contex Model (NCM) version 3.0. NCM is a conceptual model focused on the representation and handling of hypermedia documents.


[MCC19/05]
LEAL, M.A.; IERUSALIMSCHY, R. A formal semantics for finalizers. 15 p. Eng. E-mail: roberto@inf.puc-rio.br

Abstract: Automatic finalization is a common but inherently complex language facility that makes the garbage collection process semantically visible to client programs. With finalizers, memory management becomes more flexible, and garbage collectors can be used to recycle other resources in addition to memory. Formal language models usually ignore garbage collection, and therefore are unable to properly describe finalization. In this paper we use an operational approach to develop a new abstract model that explicitly represents memory management actions in a garbage-collected programming language based on the alpha-calculus. We formally state and prove several important properties related to memory management, and employ the model to describe and explore a semantics for finalizers.


[MCC20/05]
SAYÃO, M.; LEITE, J.C.S.P. Rastreabilidade de requisitos. [Title in English: Requirements traceability] 26 p. Port. E-mail: julio@inf.puc-rio.br

Abstract: The Requirements Management, which is associated to the quality of software development process, is based on requirements traceability. In this work we deal with requirements traceability, aiming to present some associated aspected to its use in the software development and evolution process. We also seek to make the reader aware of the implications of traceability to the product quality, associating traceability to both technical and managerial aspects. We present two traceability reference models, and heuristics that can guide the development team in a traceability model definition process to the system to be developed.


[MCC21/05]
BASTOS, T.A.; RAPOSO, A.B.; GATTASS, M. Um framework para o desenvolvimento de aplicações de realidade virtual baseadas em componentes gráficos. [Title in English: A framework for the development of virtual reality applications based on graphic components] 14 p. Port. E-mail: mgattass@inf.puc-rio.br

Abstract: Virtual Reality (VR) applications differ from other applications for the usage of special devices and the concern with the user's sensation of immersion. The use of frameworks is important to such applications due to the high complexity and costs involved in their development. This work presents a graphical component-based VR framework designed to be simple, flexible, and to allow the creation of easy-to-operate applications.


[MCC22/05]
RUBINSZTEJN, H.K.; ENDLER, M.; RODRIGUEZ, N. A framework for building customized adaptation proxies. 15 p. Eng. E-mail: endler@inf.puc-rio.br

Abstract: This article presents a framework for the development of adaptive proxies. The framework is in charge of collecting clients current context (device and network), and trigger the appropriate adaptations. MoCA's ProxyFramework offers mechanisms for cache management, as wel as for adaptation management. It is possible to
specify priorities and associate selectors to each adaptation. It is also possible to start an action at the moment of a context change. Developers need only to dreate their application-specific adaptations (developing adapters modules) and define trigger conditions, priorities and selectors, in the form of rules in XML format. In addition to the adapters, another extension point is the caching policy to be used.


[MCC23/05]
PAULA, M.G.; BARBOSA, S.D.J.; LUCENA, C.J.P. Conveying human-computer interaction concerns to software engineers through an interaction model. 15 p. Eng. E-mail: simone@inf.puc-rio.br

Abstract: This paper addresses the challenge of efficiently representing and communicating decisions about human-computer interaction to software engineers. It describes and illustrates in a case study how an interaction model may be used to derive a skeleton of certain UML diagrams, namely: use case, class, and sequence diagrams. Our goal is to provide a clear representation of the interactive exchanges that may take place, in order to prevent human-computer interaction decisions to be lost or inadvertently overruled when designing the system architecture and internal functional behavior.


[MCC24/05]
ENDLER, M.; RUBINSZTEJN, H.; ROCHA, R.C.A.; SACRAMENTO, V. Proxy-based adaptation for mobile computing. 22 p. Eng. E-mail: endler@inf.puc-rio.br

Abstract: In mobile computing and wireless communications, proxies are mainly used to overcome the three major problems of these networks: throughput and latency differences between the wired and the wireless links, host mobility, and limited resources of mobile hosts (MH). This report aims to present a general classification of proxy-based approaches, describe the most frequent functions assigned to proxies, discuss some of the underlying techniques used for implementing these functions, and present concrete examples of the most successful and well-known systems.


[MCC25/05]
BRANDÃO, A.A.F.; SILVA, V.T.; LUCENA, C.J.P. An ontology-based method for structuring multi-agent systems formal specifications. 13 p. Eng. E-mail: lucena@inf.puc-rio.br

Abstract: Agents are becoming a popular technology for the development of distributed, heterogeneous and always available systems. The application of agent technologies requires extensions to the existing object-oriented modeling languages to accommodate agent-related abstractions such as roles, organizations and environments. If it is difficult to analyze and establish the well-formedness of a set of diagrams of a UML-like object-oriented modeling language, it gets far more complex when the language is extended to add a set of agency related abstractions. This paper presents an ontology-based method for structuring MAS specifications. The goal of the method is to facilitate the analysis of such systems. The method proposes the analyses of MAS designs based on three phases that cover different sets of design properties. Focusing the analysis on related properties grouped into three different phases facilitates the design activity, the automatic detection of inconsistencies and the improvement of the design.
 

[MCC26/05 - classified]
SOARES, L.F.G.; RODRIGUES, R.F. Nested Context Model 3.0 - Part 5: NCL (Nested Context Language).  108 p. Eng. E-mail: lfgs@inf.puc-rio.br

Abstract: This technical report describes the basic elements and attributes of the of Nested Contex Language (NCL), version 2.2. NCL is a XML declarative language based on the NCM (Nested Context Model) conceptual model for hypermedia document specification.


[MCC27/05]
LIMA, L.S.; GOMES, A.T.A.; ZIVIANI, A.; ENDLER, M.; SOARES, L.F.G.; SCHULZE, B. Peer-to-peer resource discovery in mobile grids. 13 p. Eng. E-mail: endler@inf.puc-rio.br

Abstract: Grids are likely to be the mainstay of distributed computing. Grid was firstly applied as a concept for sharing computing resources among wired nodes, but there has been a growing interest in extending this concept to mobile environments. Although significant work has been done towards mobile grids, much of it has followed centralized approaches based on infrastructured wireless networks. We belive that a less restrictive, decentralized approach that supports mobile collaboration in ad hoc wireless networks can cater for novel grid applications. To address this issue, we propose a middleware architecture called MoGrid. MoGrid orchestrates the distribution of grid tasks among mobile devices in a peer-to-peer (P2P) fashion. In this paper, we focus on the P2P Discovery Protocol (P2PDP), which is a central element of our architecture. P2PDP aids in distributing tasks among the most resourceful devices, while mitigating the overhead of control messages exchanged among them. We describe a prototype implementation of our architecture and discuss some issues related to the adoption of P2PDP as an ad hoc resource discovery mechanism in mobile grids.


[MCC28/05]
CONDACK, J.F.; SCHWABE, D. Swell, SwellOnt and SwellQL - a software engineering environment for searching semantic web services. 15 p. Eng. E-mail: dschwabe@inf.puc-rio.br

Abstract: In this report, we describe the Swell environments. It is a software engineering tool for searching semantic Web services. In order to accomplish this, we developed a specialized ontology and a query language, named SwellOnt and SwellQL recpectively. The environment is available as an extension of Eclipse.


[MCC29/05]
CARVALHO, G.R.; PAES, R.B.; LUCENA, C.J.P.; CHOREN, R. Governing the interactions of an agent-based open supply chain management system. 27 p. Eng. E-mail: lucena@inf.puc-rio.br

Abstract: In open multi-agent systems, in which components are autonomous and heterogeneous, trust is crucial. This paper presents a law-governed mechanism to ensure trust and augment reliability in open systems. The mechanism is  based on governing the interactions within the system. This is a non-intrusive method, which allows free development of agents for open systems - they must only follow the protocols specified for the system. This paper gives a complete example of the law government approach, using a real open system environment for supply chain management. In this paper, the XMLaw description language was extended to specify the characteristics of the Trading Agent Competition - Supply Chain Management (TAC SCM) domain and these new features have been mapped to a mechanism that interprets the descriptions and analyzes the compliance of the software agents that inhabit the open software system to its laws. Using this specification, a supply chain management application based on TAC-SCM's description was developed.


[MCC30/05]
PAES, R.B.; LUCENA, C.J.P.; ALENCAR, P. Governing agent interaction in open multi-agent systems. 35 p. Eng. E-mail: lucena@inf.puc-rio.br

Abstract: Open multi-agent systems are increasingly gaining importance in both academy and industry. These systems are composed of autonomous distributed agents that can be developed by independent teams. Furthermore, agents can have different architectures, and even conflicting goals.  However, in some applications domains such as electronic commerce, the system a whole should inspire confidence in their users and also in the parts involved in the business. Confidence means basically that the system will act according to what is expected, that is, it is behaving according to specifications while maintaining the quality that its services are being provided. However, in an open multi-agent system scenario, the code of agents are frequently inaccessible, or even the agents themselves are unknown beforehand. Achieve higher degrees of confidence claims for mechanism to monitor system execution and verify if the actual behavior is compatible with what is specified. This paper uses a law-enforcement approach and presents an object-oriented framework to monitor and verify conformity of agents' behavior. First it is used a model to specify agent's interaction. This model is extended to increase its expressivity. Then, based on this extended model the object-oriented framework is built. The main goal of this paper is to share ideias about how a law governing approach can be implemented in a way that it could serve as 'how-to-do-it' models for future work in the same field.


[MCC31/05]
DE MARIA, B.A.; SILVA, V.T.; LUCENA, C.J.P. An MDA-based approach for developing multi-agent systems. 16 p. Eng. E-mail: lucena@inf.puc-rio.br

Abstract: In this paper, we propose an MDA-based approach for developing multi-agent systems (MAS). MDA specifies a structured software development process divided in modeling stages. In the PIM stage, where platform independent models are specified, we propose to use MAS-ML, a MAS modeling language that does not restrict or specify implementation platforms. In the PSM stage, where platform specific models are defined, we propose to transform MAS-ML models into UML models based on an object-oriented framework for implementing MAS. In the last stage, the application code is generated from UML models. The MDA stages and related transformations are detailed in the paper.

[MCC32/05]
FELICISSIMO, C.H.; SARDINHA, J.A.P.R.; LUCENA, C.J.P. Sweetening regulated open multi-agent systems with support for agents to reason about laws.  14 p. Eng. E-mail: lucena@inf.puc-rio.br

Abstract: Inspired by the Tim Berners-Lee's vision of a Semantic Web, this work aims to present an approach to regulate open Multi-Agent Systems (MAS) based on ontologies plus layers of logic and rules for data inference. In our approach, a top-down modeling of laws is designed for regulating an open MAS. We propose to regulate agents' actions based on four levels of abstractions: Environment laws, organization laws, role laws and interaction laws. These levels of regulations are represented by a domain independent normative ontology, which has the six related main concepts: Environment, Organization, Role, Action, Norm and Penalty. The layer of logic from our solution is composed of a combination of Description Logic, Deontic Logic and some ideas from Defeasible Logic. The rule layer is composed of user defined rules and a rule-based inference engine. Some advantages of this work are: automatic composition of laws by using rules and inference, facility to regulate agents' actions considering all the designed law levels, consistency check for pre-defined laws, and support for black-boxes agents to reason about laws in open MAS.



[MCC33/05]
CARVALHO, G.R.; LUCENA, C.J.P. Governing the interactions of an agent-based open supply chain management system.  21 p. Eng. E-mail: lucena@inf.puc-rio.br

Abstract: Governance means that specifications are enforced dynamically at application run-time. Governance framework is a technique to design open systems for extensions. We based this proposal on object-oriented framework concepts and adapted them for distributed agents and interactions. A governance framework reuses general specifications and some infrastructural services provided by open MAS. Governance frameworks structure the extensions of open system instances as variations in components, defined as roles, and variations in interactions among agents, defined as templates. Templates are used to gather a core implementation and extension points. Extension points are "hooks" that will be customized to implement an instance of the governance framework. During framework instantiation, roles are bound to external agents and templates are refined to concrete interaction specification. As a proof of concept experiment, in this paper we propose a framework for instantiating supply chain management applications as open systems.


[MCC34/05]
CACHO, N.A.A.; FIGUEIREDO, E.M.L.; SANT'ANNA, C.N.; GARCIA, A.F.; BATISTA, T.V.; LUCENA, C.J.P. Aspect-oriented composition of design patterns: a quantitative assessment.  25 p. Eng. E-mail: lucena@inf.puc-rio.br

Abstract: Pattern composition has been shown as a challenge to applying design patterns in real software systems. One of the main problems is that the implementations of multiple design patterns in a system are not limited to affect the application classes. They also crosscut each other in multiple  heterogeneous ways so that their separation and composition are far from being trivial. In this context, it is of paramount importance to systematically verify whether aspect-oriented programming (AOP) supports improved composability of design patterns. This paper presents a systematic investigation on how AOP scales up to deal with modularization of pattern-specific concerns in the presence of pattern interactions. We have made both qualitative and quantitative assessments of 62 pair-wise compositions taken from 3 medium-sized systems implemented in Java and AspectJ programming languages. Our analysis has also included the evaluation of compositions involving more than two patterns. The assessment was based on four fundamental software attributes, namely separation of concerns, coupling, cohesion, and conciseness.
 

[MCC35/05]
MILANÉS BARRIENTOS, A.Y.; RODRIGUEZ, N.L.R.; SCHULZE, B. Managing jobs with an interpreted language for dynamic adaptation. 10 p. Eng. E-mail: noemi@inf.puc-rio.br

Abstract: In this paper we explore the advantages of using interpreted languages for building submitting engines for the grid. In particular, we discuss an example engine, developed using the interpreted language Lua, for submitting jobs in a cluster, which can be extended to a grid environment. We claim that the  exibility oered by interpreted languages justies the problems related to the intrinsic loss of e-ciency associated with this kind of languages. The focus of this work is on adaptation and ease of use.


[MCC36/05]
SARDINHA, J.A.R.P.; MOLINARO, M.S.; PARANHOS, P.M.; CUNHA, P.M.; MILIDÍU, R.L.; LUCENA, C.J.P.  A multi-agent architecture for  a dynamic supply-chain management. 13 p. Eng. E-mail: lucena@inf.puc-rio.br

Abstract: Supply chain management (SCM) is a very challenging problem that is leveraging the e-commerce explosion. Today's supply chains are essentially static, because they rely on long-term relationship among key trading partners. Dynamic practices are vital because they offer better matches between suppliers and customers as market conditions change. This paper presents a flexible architecture for dealing with the next generation SCM problems, based on a distributed multi-agent architecture of a dynamic supply chain.  We define intelligent agent roles that tackle sub problems of a dynamic SCM, and present a solution for combining this results in a distributed environment. The agents' typical tasks are bid planning, production scheduling, supplier negotiation, among others. We also present an implementation of this achitecture used in the international test bed for SMC solutions, the Trading Agent SCM competition, as well as some experimental results.


[MCC37/05]
ARAÚJO, A.P.F.; URRUTIA, S.A.; BOERES, M.C.S.; REBELLO, E.F.V.; RIBEIRO, C.C. Hybrid parallel metaheuristics for computational grids - a case study for the mirrored traveling tournament problem.  19 p. Eng. E-mail: celso@inf.puc-rio.br

Abstract: Metaheuristics are general high-level procedures that coordinate simple heuristics and rules to find good approximate solutions to computationally difficult combinatorial optimization problems. Parallel implementations of metaheuristics appear quite naturally as an effective approach to speedup the search for approximate solutions. Besides the accelerations obtained, parallelization also allows solving larger problems or finding better solutions. We present in this work four slightly differing strategies for the parallelization of an extended GRASP with ILS heuristic for the mirrored traveling tournament problem, with the objective of harnessing the benefits of grid computing. Computational experiments on a dedicated cluster illustrate the effectiveness and the scalability of the proposed strategies. In particular, we show that the parallel strategy implementing cooperation through a pool of elite solutions scales better than the others and is able to find solutions that cannot be reached by the others. Computational grids are distributed high latency environments which offer significantly more computing power than traditional clusters. The best parallel strategy was also implemented and tested using a true grid platform. We report original results from pioneer computational experiments on a shared computational grid formed by 82 machines distributed over four clusters in three cities, illustrating the potential of the application of computational grids in the fields of metaheuristics and combinatorial optimization.


[MCC38/05]
GARCIA, A.F.; BATISTA, T.V.; RASHID, A.; SANT'ANNA, C.N. Driving and managing architectural decisions with aspects.  13 p. Eng. E-mail: lucena@inf.puc-rio.br

Abstract: Software architects face decisions every day which have a broadly-scoped impact on the software architecture. These decisions are the core of the architecting process as they typically have implications in a multitude of architectural elements and views. Without an explicit representation and management of those crucial choices, architects can not properly communicate and reason about them and their crosscutting effects. The result is a number of architectural breakdowns, such as decreased evolvability, time-consuming trade-off analysis, and unmanageable traceability. Aspects are a natural way to capture widely-scoped architectural decisions and promote software architectures with superior modularity.


[MCC39/05]
FILIPPO, D.; FUKS, H.; LUCENA, C.J.P. AulaNetM: extension of the AulaNet environment to PDAs.  9 p. Eng. E-mail: hugo@inf.puc-rio.br

Abstract: The widespread use of mobile equipment and wireless networks offers a huge potential for educational applications, since these technologies enhance traditional applications and bring about new ones. Interactive multimedia portable devices can be regarded as a powerful learning tool, even though there are restrictions such as small screen sizes, limited memory, unstable connectivity and difficulties imposed by mobility. In this paper, we present the AulaNetM, an extension of the AulaNet teaching-learning environment for PDA users, explaining, how context information is important in order to enrich the learning activities carried out and, at the same time, select the content and the activities presented to the learner.


[MCC40/05]
FURTADO, A.L. Lattice operators over entity classes.  29 p. Eng. E-mail: furtado@inf.puc-rio.br

Abstract: The static conceptual specification of information systems, adopting the Entity-Relationship (ER) model, has been, for a long time already, extended with is_a and part_of links, whereby specialization/generalization and decomposition/aggregation of entity classes can be introduced. Usually these links impose tree-like hierarchies, although more general cases of partial ordering are quite often encountered in practice. Here we shall consider Lattice hierarchies, on which two dual operators are defined. Using these operators, we show one way to classify entity instances on the basis of their currently known properties, in turn organized as lists of property: value pairs (frames).
A broad overview of the facilities offered by our method is presented, including queries involving structured requirements, also expressed by frames, which look for approximately matching instances.


[MCC41/05 - classified]
SOARES, L.F.G. Recomendações para modelos de referência do Sistema Brasileiro de TV Digital.  20 p. Port.  E-mail: lfgs@inf.puc-rio.br

Abstract: This report is related to the Brazilian Digital TV System recommendations - SBTVD - , regarding to the SBTVD declarative middleware. In particular, it deals with the  problem of media synchronization.


[MCC42/05]
ALMENDRA, V.S.; SCHWABE, D.; CASANOVA, M.A. Toward real-world trust policies. 17 p. Port.  E-mail: dschwabe@inf.puc-rio.br

Abstract: The increasing reliance on information gathered from the Web and other Internet technologies (P2P networks, e-mails, blogs, wikis, etc.) raises the issue of trust. Trust policies are needed to filter out untrustworthy information. This filtering task can be leveraged by the increasing availability of Semantic Web metadata describing retrieved information. It is necessary, however, to adequately model the concept of trustworthiness; otherwise one may end up with operational trust measures that lack a clear meaning. It is also important to have a path from one’s trust requirements to concrete trust policies through Semantic Web technologies. This paper proposes a classical logic model for trust policies, grounded on real-world models of trust, and some guidelines to map trust requirements to trust policies. It also presents the implementation of a case study using Prolog.


[MCC43/05]
COSTA, T.S.  Métodos lineares de passo k para a solução numérica de equações diferenciais  ordinárias. 114 p. Port. E-mail: tcosta@inf.puc-rio.br

Abstract: The main objective of this text is to present a family of formulas for the numerical solution of ordinary differential equations with initial conditions, the k-step K Linear Methods.