SOFTWARE

Author: VILLALOBOS ABARCA MARCO ANDRES.
Year: 2003.
University: POLITÉCNICA DE MADRID [www.upm.es].
Place of defense: FACULTAD DE INFORMATICA.
Place of preparation: FACULTADA DE INFORMATICA.

Summary: This thesis has been defined a Model for Developing Sofwarre Education Curriculum (MDSWEC) based approach and a collaborative work environment that supports this model. Initially defines a generic collaborative model based on a set of theoretical approaches that essentially designing systems Work Collaborative Soportador Computer (Computer supported cooperative work [CSCW]). Basic components of the model are comceptos of Place Activity, Career Activities Pocedimiento Collaborative and the relations between ellos.El modelocolaborativo is specialized to the development of SWEC by defining a process and equipment development of this type of aplicaciones.El process and equipment were derived from the study of the foundations of enseañanza computer-assisted (Computer Assisted Learning [CAL]) and some proposals demodelos to develop SWEC.El generic model was compared with the related work of identifying ties and strengths to improve our model. The environment, called Synergy, a multidisciplinary team to develop SWEC can use dediversos communication mechanisms (such as videoconferencing or chat); manage the monitoring of actimvidades team and individual, in terms of the definition, visualization and execution model development process (step / task); get a group consciousness by the losintegrantes of the development team through annotations related to the actions being performed on devices built in a certain time, or through motificacinoes of occurrence of events in the development process of SWEC and record information of the development process comotambión of artifacts product and its history, in a basr data. Sinergí has been developing through an evolutionary prototyping approach to the Web and with a three-tiered structure, with technologies such as HTML.Sevelts, JSP, and abase relational data. In addition, each iterción prototype has been tested for functionality and usability that have allowed their mejora.Además of evidence the environment was compardo with traabajo realcinado identified weaknesses and strengths that enabled her mejora.Finalmente, entornpo underwent a summative evaluation to verify its practical usefulness and validity of the proposals sutentan estetrabajo research. The model and its emfoque collaborative underlying (incorporated into its various components) have proven ujn strong generic and novel mechanism coordinación.El model is the basis of architecture and dimámica system Synergy that, in addition to integrated communication mechanism, perception and memory group, and definition of process models an environment useful and effective pair support the development of collaborative software.

Author: SIERRA RODRÍGUEZ JOSÉ LUIS.
Year: 2003.
University: COMPLUTENSE DE MADRID [www.ucm.es].
Place of defense: FACULTAD DE INFORMÁTICA.
Place of preparation: FACULTAD DE INFORMÁTICA UCM.

Summary: This thesis defended the feasibility of producing applications through the automatic processing of documents, opening up, thus keeping the path to what is called a paradigm documentary construction applications. The development of an application, according to this paradigm begins by describing its most salient aspects through one or more documents. Next, those documents are marked by a descriptive markup that depends on the specific domain and the functionality of the application. The application itself is processing its documentation with a processor suitable for the language used in the marking of the same. The development of this thesis is conducted in five stages: * CONTEXT OF THE THESIS This review key aspects of descriptive markup languages, their processors, as well as the approach to software development based on specific languages domain. The role of descriptive markup languages in this thesis is to make explicit the logical structure of applications documentation. In this way, these languages allow processors to produce documentation of the applications. In this way, these languages allow processors to produce applications from such documentation. Finally, aspects relating to the approximation based on specific language lays the foundation for the development and operationalization of markup languages used in each application domain. * FORUMULACIÓN OF APPROXIMATION ADDS Approximation ADDS (Documentary Approach to Software Development) is a pragmatic realization paradigm documentary, which according to the language used to mark documents applications evolve according to the needs, both of which provide expert those documents, as the developers that process. * FORMULATION OF MODEL OPERACIONALIZACIÓN OADDS The evolving nature of markup languages in ADDS reflected, at the operational level, in a model of development-oriented processing incremental processors modulators from components that can be extended and combined as the language evolves. This model is called OADDS (Operationalization in ADDS). * MAKING OADDS AS A FRAMEWORK ORIENTADO TO OBJECTS OADDS is independent of specific implementation technologies. However, in this thesis that model is practical and makes a direct and compact as an object-oriented framework. * IMPLEMENTATION OF ADDS TO THE CONSTRUCTION OF PROCESSORS OADDS In building processors OADDS can apply the same paradigm documentary to describe procesdores through documents marked.

Author: ESTEBAN ANGELES JUAN LUIS.
Year: 2003.
University: POLITÉCNICA DE CATALUÑA [www.upc.edu].
Place of defense: FIB.
Place of preparation: LLENGUATGES I SISTEMES INFORMATICS, CAMPUS NORD UPC.

Summary: Chapter 1 presents most of the concepts that will be used later. We begin by explaining the origin of the complexity of the demonstration. Then we define systems and demonstration of complexity measures considered in this work. Complexity also reseñamos results of the demonstration and list our own results. Chapter 2 deals with results on size for Resolution, R (k), and Driers Plans, which include lower levels of space and higher when they are related provide separations between different systems or between demonstration versions of a large tree and same system demonstration. Chapter 4 is short and it recapitulamos our results and suggest problems associated with our work that we consider interesting. It also appears a table of upper and lower bounds of size and space including our results. This table provides see what have been our contributions to the complexity of the demonstration. The results appeared in [BEGJ]. In this section we improve the separation between versions tree and general resolution and Flat Driers. To do this, we extend a lower bound for circuit sizes of Boolean monotonous Ran and McKenzie circuits real monotonous. This type of separation is interesting because many demonstrators automatic theorems are based on the version tree demonstration systems, and shows that the separation is not always a good idea to the restricted version tree. What we do is demonstrate exponential lower levels for Flat Driers tree for certain formulas using ownership of the interpolation monotonous feasible, which are in turn lower bounds for the Resolution tree. To obtain the separation showed higher levels polynomial in Resolution to the same formulas, which are in turn to higher heights Flat Driers. In fact, not only split Resolution Resolution tree also split Resolution tree certain restrictions Resolution Resolution as regular and negative. The separation between Resolution and Resolution tree was subsequently enhanced by BenSasson, Impagliazzo and Wigderson. The third section shows a relationship between R (2) and refutations in Resolution Principle Pigeonhole. The result is not published. We demonstrate that the lower levels of R (2) for the Top Pigeonhole with n palomares yn ^ (1.5) doves provides lower bounds for Resolution Principle Pigeonhole with palomares and n * n pigeons. It was a new attempt to resolve an issue open for a long time. Of course we do not know if this approach would have made it easier demonstration, but as Raz and Razborov solved the problem independently while he worked in abandoning such an approximation. The fourth section is called "Sixe and Treelike (R) k. The results were published in [EGM]. Was known that R (2) was more potent than Resolution R (2) tree more potent than Resolution tree therefore a natural question was to find out if you could set aside future levels of R (k) or R (k) tree. show exponential separations between successive levels from which we can draw the hierarchy R (k) tree. This means that there formulas require rebuttals in R (k) in size exponentially while rebuttals have size polynomial in R (k +41). This is also true separation between R (2) tree to tree and Resolution Related to this result, Segerlind, Biss and Impaglilazzo showed that R8k) is a hierarchy with respect to the size, extended our result. also demonstrated that R81) dominates R8k) tree. is a particular case of a simulation of Krajicek, but we show that an increase by a factor of 2, independent k enough.

Author: BALLIS DEMIS.
Year: 2004.
University: POLITÉCNICA DE VALENCIA [www.upv.es].
Place of defense: DEPTO. DE MATEMÁTICA E INFORMÁTICA.
Place of preparation: DEPTO. SISTEMAS INFORMÁTICOS Y COMPUTACIÓN.

Summary: The increasing complexity of software systems has led to the development of methodologies for formal verification and correction of data and programs. Generally, if a program set behaves according to the original intentions of the developer or control the consistency and correctness of large data sets are not trivial tasks, as attested by numerous case studies that found in the literature. In this thesis, we address two issues open for verification and correction. Specifically, verification and correction of declarative programs and the verification and correction of Web sites (ie data sets semi). First, it has defined a general outline for the automatic correction program based declarative rules, which operates a combination of inductive learning techniques top-down and bottom-up. Our methodology is able to lower hybrid corrections that are difficult, if not impossible, to achieve with a system simpler automatic learning purely top-down or bottom-up. It has also been particularized general outline two declarative programming paradigms well known: the functional paradigm logical and functional paradigm. Secondly, it has formalized a framework for the verification of autonomous Web sites, which can be used to specify terms of integrity about them, and then automatically check if these conditions are met. On the one hand, we have defined a specification language based on rules, which allow you to define both syntactic and semantic properties of a website. Moreover, it has formalized a verification technique that detects patterns incorrect / prohibited and lack of information, ie Web pages incomplete or absent. During the verification process, contains useful information that can be used for the repair of the portal. Therefore, after the verification phase, it is also possible to infer some possible fixes for the semi-automatically fix the website wrong. Our methodology is based on a new technique based on rewriting (partial rewriting), which replaces the traditional mechanism of pattern matching with a technical adjustment more convenient (tree simulation), which facilitates the recognition of patterns in a semi.

Author: FUENTES FERNÁNDEZ RUBEN.
Year: 2004.
University: COMPLUTENSE DE MADRID [www.ucm.es].
Place of defense: FACULTAD DE INFORMÁTICA.
Place of preparation: FACULTAD DE INFORMÁTICA.

Summary: The methodologies of software engineering Oriented Agents (ISOA) have addressed social issues and intentional Systems Multi-Agente (GHS) using traditional methods of development, from other paradigms, or with specific technical focus on isolated plots of these aspects. However, such solutions are not satisfactory. The first option is not entirely suited to this kind of things, outside the traditional artifacts of the Software Engineering. The second has led to a problem working with social and intentional properties affecting many facets of GHS. To address the problems in dealing with these properties, this thesis proposes to use an integrative framework that addresses the social aspects and intentional of SMAs as a whole, both individually and in community. This framework is inspired by the theory of Activity (CA). This is a proposal conceptual and methodological which is used for the analysis of individuals and societies in the social disciplines. The TA is studying properties similar to those focused our interest in the ISOA but in human societies. Starting with the knowledge and techniques of the TA have been defined modeling languages and structures to represent knowledge of the TA, as well as methods to collect information and resolving inconsistencies in the specifications of SMAs. These methods are designed as tools for existing methodologies in the ISOA. The development cycle methodology is a permanent job translation of the specifications to the concepts inherent in the TA, capture new knowledge and / or verification focused on the social aspects and intentional GHS and translation back to the methodology agents. In this way developers can work on the system with the techniques they wish: the methodology of the ISOA or those of the TA. At the same time enjoy the support of a body of knowledge based on the social sciences to work consistently with problems of the paradigm of agents, such as motivation, organization, leadership development, cognitive abilities, learning, laws or trust.

Author: PEREZ CAMAÑO JUAN LUIS.
Year: 2004.
University: POLITÉCNICA DE MADRID [www.upm.es].
Place of defense: FACULTAD DE INFORMATICA.
Place of preparation: FACULTAD DE INFORMATICA . UNIVERSIDAD POLITECNICA DE MADRID..

Summary: This work Doctoral thesis describes the development of a computer system that allows robust and efficient way of monitoring and forecasting of air pollution for different sources. The primary motivation for its implementation has been trying to respond to the increasing demands on issues of air quality in the environmental authorities (EU, National, Community, municipal, etc.) to build a system with the capacity to make predictions quality air and evaluate the impact of various sources, in an operational mode and real time, which helps to bring effective strategies to reduce pollution. The complexity and high non-linearity of the physical, chemical, biological, etc.presentes in atmospheric dynamics requires the use of numerical models with high computational requirements. In this paper we have integrated model MM5 (PSU / NCAR, USA) EMMO (Emission Model, UPM, Spain) and CMAQ (Community Multiscale Air Quality modeling system, EPA, USA). The process d eintegración has required research and development of a series of modules to adapt, usually aimed at generating consistently in each of the input data required that the research and development of a series of modules adaptation, directed normally to generate consistently in each of genetic data entry required for different models. For the interpretation of the information obtained has developed a sophisticated information system, Analysis and Post-procesado that allows access to the wealth of information gathered, which has used recent advances in Information Technology. The developed tool has been used in various real-world applications where it has proven its usefulness and efficiency, in addition to check the robustness, reliability and robustness through processes of validation and calibration.

Author: ECHAGUE GUARDIOLA JUAN JESUS.
Year: 2004.
University: JAUME I DE CASTELLON [www.uji.es].
Place of defense: E.S. TECNOLOGIA Y CIENCIAS EXPERIMENTALES.
Place of preparation: E.S. TECNOLOGIA Y CIENCIAS EXPERIMENTALES.

Summary: The increase in demand for transmission of multimedia applications across packet-switched networks, such as the Internet, has led to a line of research in the analysis of the conditions that can offer Quality of Service (QoS) to such applications. The concept is associated QoS guarantee end-to-end delay, packet loss percentage, garanizar bandwidth, and so on. The QoS in packet switching networks depends mainly on the network topology, the algorithm planning packages, and the flow of data coming into the system. The algorithm planning packet FIFO is the most used in packet-switched networks because of its simplicity and ease of implementation. Unfortunately, the protocol FIFO, even when the limit inflows can lead to situations in which one can not restrict or end-to-end delay and size of buffers for storing packages (unstable situation in the network); imposibilitando the ability to ensure the maximum transmission time nor the percentage of pédida package. This thesis presents contributions to enhance the state of the art related anális stability packet switching networks in the protocol that is used for planning packet FIFO. The system model is common and intended to reflect its application to the architecture of the Internet and integrated services digital network.

Author: LOPEZ PEÑA PEDRO.
Year: 2004.
University: POLITÉCNICA DE MADRID [www.upm.es].
Place of defense: FACULTAD DE INFORMÁTICA.
Place of preparation: FACULTAD DE INFORMÁTICA.

Summary: The software is becoming an increasing part of our daily lives, including critical issues. There are techniques that seek to obtain a free software errors or attempting a software bug could cause catastrophic consequences. The routine use of these requires a taxonomy of software failures. There are classifications in different domains such as space systems or nuclear weapons, however there is a taxonomy of software bugs that covers all types of software failures accidental. The objective of this thesis is to define a clear and comprehensive taxonomy that will help to better understand its origins and software failures, establecientes the foundation for a comprehensive and systematic analysis of software and solutions for better fault-tolerant software looking for a more reliable and secure. This thesis proposes a taxonomy of software failures, based on an analysis of software development based on the classifications of existing software failures. This taxonomy of software failures (TSF) provides 4 views and judgments classified under 8 dimensions.

Author: MONTERO MORENO SUSANA.
Year: 2004.
University: CARLOS III DE MADRID [www.uc3m.es].
Place of defense: ESCUELA POLITECNICA SUPERIOR.
Place of preparation: UNIVERSIDAD CARLOS III DE MADRID.

Summary: The methods of design hypermedia attending the designer through a series of phases and products. However, a designer, whether it is not as skilled, could take an inadequate solution to a particular design problem that can be harmful to the user. The design patterns are such a mechanism that can help relieve both embarked on new developments such as the difficulty in making decisions for the improvement of existing systems, as long as they can efficiently integrate with the design methods and are accessible effectively to the various members of the development team. To address these shortcomings, the first objective of this thesis has been organizing awareness of the pdh, providing criteria for classification, a catalog of patterns, patterns of a language and a framework for integration with the design methods, in which a set step guide an approximation based on patterns that transform allows the specification of requirements entities conceptual design modeling, thanks to solutions captured by the employers. In addition to automating this toolkit has proposed a model of semantic representation based on ontologies that keeps linked the textual representation of the pattern through endorsements, making patterns mutually comprehensible to designers and programs.

Author: ROSSAINZ LOPEZ MARIO.
Year: 2004.
University: GRANADA [www.ugr.es].
Place of defense: E.T.S. INGENIERIA INFORMATICA..
Place of preparation: E.T.S INGENIERIA INFORMATICA.

Summary: Under an atmosphere of Parallel Objects, an approach Programming Paradigm of Object Orientation, it is proposed a method of programming based on the High-Level Parallel Compositions or CPANs through classes. They are used modes of communication synchronous, asynchronous and future asynchronous model Objects Parallels, patterns of communication / interaction of predefined structured approach, the encapsulation and abstraction of Object Orientation, to provide reusability these patterns, along with a restrictions predefined set of synchronization between processes (maxpar, mutex and sync). It explains the implementation of the communication patterns most commonly used by the application of the method, which make up a class library, which might be used in applications within the programming environment of C + + + and the POSIX standard programming with threads. This research focuses on methods Structured Parallel Programming, proposing a new implementation with C + + + threads library and POSIX programming approach based on the model of the High-Level Parallel Compositions or CPSNs of Corradi and colleagues, which based on the paradigm of Object Orientation to solve problems paralelizables using a class of objects competing assets. This paper proposes a class library that provides the programmer patterns of communication / oteracción most commonly used in parallel programming, in particular the pattern of pipeline, called the pattern and the pattern tree farm in the technical design of algorithms , known as divide and conquer.

Author: ARJONA FERNÁNDEZ JOSÉ LUIS.
Year: 2004.
University: SEVILLA [www.us.es].
Place of defense: ESCUELA TÉCNICA SUPERIOR DE INGENIERÍA INFORMÁTICA.
Place of preparation: ESCUELA TÉCNICA SUPERIOR DE INGENIERÍA INFORMÁTICA.

Summary: In recent years the website has established itself as one of the most important respiratory information. A big challenge for the actors software has been dealing with that amount, shortly manageable data to extract information with meaning. This process is difficult for the following reasons: First, the information on the website is aimed at consumption by humans and does not contain a description of their semantics, which would help agents understand. Secondly, the Web changes constantly, which generally has an impact on the representation of information, but not in its semantics. Finally, there is a huge repository with 4200 terabytes of information ready to be consumed. Members of The Distributed Group have been working in distributed systems since 1997. Specifically, they have worked on models of interaction multipartite providing the programmer appropriate mechanisms for describing complex interactions from a conceptual point of view. The results have materialized publications in leading journals and dissertations. The research in this report opened a new line of investigation in the group. Its aim is to facilitate the design and implementation of software agents. Currently, this line investigaicón is reinforced by the proposition Joaquin Pena, in which mechanisms are being developed to describe the complex interactions abstractly in societies multi-agentes. In this report we present a new framework for the collection of information with syntactical meaning of the web today. Its main advantages are: semantics associated with the extracted information, improve interoperability Agent: traffic changes on the Web, enhancing resilience: in addition, establishes a separation of responsibilities in the task of extracting, automating the development of distributed knowledge extractors . Finally, the detailed study of the labor involved shows that our proposal represents an original contribution.

Author: VICENTE CHICOTE CRISTINA.
Year: 2005.
University: POLITÉCNICA DE CARTAGENA [www.upct.es].
Place of defense: ESCUELA TÉCNICA SUPERIOR DE INGENIERIA DE TELECOMUNICACIÓN.
Place of preparation: ESCUELA TÉCNICA SUPERIOR DE INGENIERÍA DE TELECOMUNICACIÓN.

Summary: The wide variety of systems currently managed by software triggered the complexity of the applications you need to be developed. Moreover, given the speed with which technological advances occur, especially in areas such as electronics or communications, it is essential to provide the flexibility of the applications they need to assume power, with enough agility, the constant changes imposed by the market. Moreover, it is equally essential to the achievement of this increased flexibility is not at the expense at the expense of harming other requirements such as those related to the efficiency, cost or time of application development. In response to these new demands, in the past have presented new paradigms of software development among which may include Object Oriented paradigm, the Oriented aspects, Directed by models, based Components, based in Product Lines , or paradigms Programming Generativa and Visual Programming. However, for various reasons, these proposals have not found an excessive acceptance in the area of systems development mixed hardware / software (HW / Sw). This has helped the lack of consensus on the methodology and notation to be used, as well as the lack of tools that support, in a holistic manner, the entire life cycle of these products. In fact, at present these systems are often developed in a very focused on the hardware, software being relegated to the background leading to the collection of designs very efficient but usually little or no flexibility or reusable. This thesis proposes a new approach to systems development Hw / Sw from a perspective more focused on the software and from which seeks to improve the flexibility and the degree of reuse of these applications without neglecting other aspects such as eficienci , reliability, or development time. To do so, since a mixed systems software is the most meleable, seeks to analyze how the incorporation of emerging trends in the area of software engineering can help improve both the process of building such systems as the results that it is derived. Seeking to draw on the experience accumulated by the Survey Group, which has conducted this Doctoral Thesis in the development of several types of mixed systems Hw / Sw, and in particular Systems Visual Information Processing (VIPS) The work presented here is centará in this application domain, although the results are, to some extent, directly extrapolabies to other domains such as the control systems in real time or the systems telecomuncación. The report describes some of the preliminary work in the field of VIPS who have served and support for the development of this thesis. It also describes the usual procedure for the construction of these systems and improvements to the existing paradigms of software development can bring to their design and implementation. On that basis proposes a new methodology for developing VIPS in which embodies many of the paradigms previously described and which seeks to resolve, to a large extent, the problems and limitations of 8 proceeded 3c3 tion today. The work was completed with the description of the tool IP-CoDER which is designed to support the new methodology which allows, among other things, to integrate a number of tools and libraries of image processing currently on the market.

Author: HAYA COLL PABLO ALFONSO.
Year: 2005.
University: AUTÓNOMA DE MADRID [www.uam.es].
Place of defense: ESCUELA POLITÉCNICA SUPERIOR.
Place of preparation: ESCUELA POLITÉCNICA SUPERIOR.

Summary: In the late 80 surge Ubiquitous Computing when Mark Weiser proposes moving the computing capacity of the stiff and bulky personal computers to thousands of devices spread throughout the environment, so that computers are built with the environment until it becomes invisible to the user . In Ubiquitous Computing interaction persona-ordenador spreads throughout space, giving rise to environments capable of independently acquire information and use it to suit the needs of the occupants. This interaction between the environment and the user is benefit when considering contextual information such as the location, the job being done by the user, other resources in the vicinity, the environmental conditions of environment, and so on. This implicit understanding of the situation makes it possible for the environment responds with a degree of proactivity, in a way that frees users' attention when it is not essential. The combination of contextual information with Intelligent Environments has a number of complications that hinder the development of context-sensitive applications. We have to integrate and manage a heterogeneous mix of technologies that allow capture, distribute and modify the context taking into account that this contextual information comes from sources of varying nature, and that can be described at different levels of abstraction. To this must be added that the configuration of the environment is dynamic, so they can come and go new components of the environment. The result is that the application developer context-sensitive faces a casuistry unwieldy due to the large number of possible configurations. The proposal for this thesis lies in a layer of context that serves as glue to get the necessary synergy between the elements that constitute the ubiquitous Intelligent Environment. It provides: (a) a standard representation of the context that is independent of their nature and level of abstraction. It provides a unified model of the world by reducing the complexity in access to the contextual information. (B) A flexible mechanism for storing and distributing information that facilitates the reuse and configuration of the various components of the environment. The implementation of the layer of context resides in a data structure called the overall slate. This board is a model of the world, where it is stored whole context generated by the components of the environment and the same environment. The board stores the information in the form of a graph, where each vertex represents an entity such as a user, an application, a device, room, etc.. The context is represented by a common language, so that coexist on the blackboard information from the sensors --- as a state of physical devices with context --- deducted from the sensory information, such as the number people who are in the environment, or the task being undertaken by one person at any given moment. Thanks to this architecture provides a weak coupling between various components at three levels: temporal, spatial and functional. This anonymity allows the connection or disconnection of a component in the environment is done in a transparent manner to the rest. The layer of context that is proposed has been tested in a real environment Intelligent equipped as a conventional living room of a house, and partly as an office. This prototype includes devices of different technologies, multiple applications context-sensitive user interfaces and two different nature which are generated automatically 8 minds 2a9 based on the model of the environment stored on the blackboard.

Author: MORAGA DE LA RUBIA Ma. ANGELES.
Year: 2006.
University: CASTILLA-LA MANCHA [www.uclm.es].
Place of defense: E.S.INFORMATICA.
Place of preparation: ESCUELA SUPERIOR DE INFORMATICA CIUDAD REAL.

Summary: The fact that a portal is getting quality is critical because it is the only mechanism to ensure that users have access to it again (Offutt, 2002). It should be borne in mind that the quality has been a topic of research for a long time and in different fields. But despite this, quality from the point of view of the Web portals, has barely been addressed. Therefore, our objective is to fill this gap by creating a quality model for portals. To achieve this quality model, will be dealt with the quality of websites from two points of view: the quality external and internal quality. From the point of view of quality external objective will be to develop a model of quality for portals that allow on the one hand, to determine the level of quality of a particular site, depending on the quality characteristics affecting it. And secondly, to identify the weaknesses of the site in order to define corrective actions to improve its quality. On the other hand, from the point of view of quality, we will use internal websites developed by aggregating portlets. Specifically, within the scope of this thesis we will focus on the quality of each of the portlets that make up such portals, excluding the interaction between portlets that make up the portal. Therefore, it will develop a model of quality of portlets that allow on the one hand, to establish the level of quality of a particular portlet. And on the other hand, will establish what are the weaknesses Portlet with the objective to be able to define corrective actions for achieving better quality of the portlet. Finally, we develop a model for the selection of portlets to help developers websites to select the best Portlet among a set of portlets that have a similar set of functions for specific tasks and objectives of the users.

Author: GUENAGA GÓMEZ MARÍA LUZ.
Year: 2006.
University: DEUSTO [www.deusto.es].
Place of defense: FACULTAD DE INGENIERIA.
Place of preparation: FACULTAD DE INGENIERÍA.

Summary: The importance of accessibility and usability on the Internet is measured in terms of social, economic and personal for all those involved in the definition, development, operation and use of services and products. A few years ago, the Web was an optional service, which some users could benefit. Today it is essential for various daily activities. In this new context, people with disabilities represent a large group of users, with specific characteristics to be taken into account when designing technology products. Their role is changing from users receiving information to managers and administrators. In estepunto where guidelines for accessibility and usability do not reach to cover the needs of design tools as a Digital Resource Center, from the point of view of the content management system. This user profile has a strong interaction with the interface, not usual in the current developments Web, and handles a large amount of information, complejay strongly interrelated, whose accessibility has not been addressed in the work done so far. This paper describes design patterns accessible and usable interfaces for Digital Resource Center, but may be extended to other contexts with similar characteristics of interaction and information. These guidelines complement existing businesses and are aimed at users with visual impairments. It also defines a hierarchy of types of digital resources beyond Web pages, whose accessibility is necessary to study and improve. This is associated with each type of resource a set of metadata, or attributes of accessibility, which aim to replace the digital resource when it is not accessible, provide certain information when replacing or supplementing it is not possible to improve their understanding by the user with visual impairments. The results and contributions of this thesis are the result of a work in close collaboration with users with visual impairments, through a Web questionnaire, telephone contact, e-mail and tests with users. These methods have confirmed to be very valuable to know the reality of this group, the way they interact with the interface and to use technology to make direrentes tasks.