Os Procedimentos de Manifestação de Interesse (PMI) e Propostas Não Solicitadas (PNS) são mecanismos para a viabilização de projetos de Parcerias Público-Privadas (PPP/PFIs). Contudo, a ausência de critérios claros para avaliação e priorização de parceiros, de forma transparente e técnica, leva à seleção adversa de propostas, contribuindo para o insucesso dos projetos. O principal instrumento utilizado na avaliação PPP/PFIs é uma análise quantitativa baseada no Value For Money (VfM). O propósito deste trabalho foi investigar os problemas relacionados com os processos de avaliação das propostas dos PMI submetidas aos órgãos demandantes dos serviços. Iniciou-se com a análise do Fatores Críticos para o Sucesso (FCSs) das PPP/PFIs, dos PMI e PNS, identificadas na literatura. Os Fatores Críticos foram apresentados a um conjunto de 78 especialistas que responderam também sobre a processo de análise das propostas. No questionário os autores foram confrontados também com as propriedades da visão. O resultado indica que a análise com o VfM seria incompleta e insuficiente para capturar a complexidade do escopo destes projetos. A pesquisa possibilitou ainda organizar os fatores críticos em 7 dimensões a serem consideradas na avaliação destes projetos. Essas dimensões podem ser utilizadas para a criação de instrumentos de avaliação e um exemplo foi proposto, um indicador nomeado de índice \"T\", avaliado por um grupo de 12 especialistas em projetos de PPP/PFIs. A pesquisa contribui, portanto, para apoiar a criação de instrumentos de avaliação de PPP/PFIs que considerem uma análise mais ampla, considerando aspectos qualitativos e quantitativos. / Unsolicited Proposals (UNPs) are one of the main mechanisms for verifying the feasibility of Public-Private Partnerships (PPP/PFIs/PFIs) projects. However, the lack of clear criteria to evaluate and to prioritize the partner, in a transparent and technical way, has led to adverse selection of proposals increasing the level of failure of these projects. The main instrument used in the PPP/PFIs evaluation is a quantitative analysis based on Value for Money. The purpose of this research was to investigate the problems related with the processes of evaluation of the Unsolicited Proposals (UNPs) that are submitted to the respective government offices. The study compiles the analysis of Critical Success Factors (CSFs) for Public-Private Partnerships (PPP/PFI) and the Unsolicited Proposals (UNPs), identified in the literature, and synthetized by the evaluation of a group of 78 specialists, resulting in the organization of 7 (seven) dimensions to be considered in the evaluation of these projects. This study also demonstrated that, according to these experts, the analysis with VfM would be incomplete and insufficient to capture the complexity of the scope of these projects. As an alternative a set of 7 (seven) indicators where created related to each dimension and a general index that synthetize the performance in the dimensions, considering qualitative and quantitative aspects. The index was labeled as \"T\" index and evaluated by a group of 12 (twelve) experts on PPP/PFI projects. The results from the analyses indicate that the index allows a linear comparative analysis between the proposals received by the public entity requesting the Public- Private Partnership (PPP/PFI) services. The model of indicators decreases the subjectivity of the choices made by public managers, resulting in greater transparency for the process of proposal qualification and security to the market.

Referentiell kommunikation innebär att kunna delge andra information så att de förstår, samt att själv förstå när man lyssnar och även att vara medveten om när man själv inte har förstått (Sonnenschein & Whitehurst, 1984). Genom en referentiell kommunikationsuppgift har lyssnaregenskaper inom referentiell kommunikation undersökts hos ungdomar med lindrig till måttlig utvecklingsstörning, samt en mentalt åldersmatchad kontrollgrupp. Snabb benämning (Rapid Automatized Naming) av enstaviga och flerstaviga ord har också testats, och huruvida samband mellan resultat i kommunikationsuppgiften och snabb benämning föreligger. För att mäta referentiella lyssnarförmågor har olika typer av efterfrågningar studerats, det vill säga hur individen uttrycker sig för att efterfråga mer information. Resultatet visar att kontrollgruppen producerar signifikant fler av efterfrågningstyperna begäran av utökad information och kontrollfrågor. Båda grupperna var mycket heterogena i kommunikationsuppgiften vad gäller antalet efterfrågningar hos varje individ. I målgruppen återfanns ett signifikant samband mellan efterfrågningstypen aningen-eller-frågor och långsam benämningshastighet av enstaviga ord. En klinisk implikation av resultatet i studien är att en del individer i målgruppen skulle kunna gynnas av träning i referentiell kommunikation.

La présente recherche a pour objectif d’examiner les stratégies de politesse utilisées par des apprenants coréens qui apprennent le français comme langue seconde lors des situations de demande et de mettre en évidence le rôle de la L1 et de la culture d’origine des apprenants quant à leurs choix des stratégies de politesse en français. Trois groupes de répondants ont participé à la recherche : 30 locuteurs natifs coréens qui habitent à Séoul, en Corée, 30 étudiants coréens apprenant le français comme langue seconde à Montréal, au Canada, et 30 locuteurs natifs francophones habitant Montréal. Les données recueillies auprès des locuteurs natifs ont servi de bases de données reflétant respectivement les normes de la L1 et celles de la langue cible. Pour la collecte de données, nous avons utilisé quatre outils : le Discourse Completion Test (DCT) écrit, des échelles d’évaluation des variables contextuelles, un Questionnaire à choix multiples (QCM) ainsi qu’une entrevue semi-structurée. Les réponses obtenues au DCT écrit ont été analysées à l’aide du schéma de codage élaboré dans le cadre du Cross-Cultural Speech Act Realization Project (CCSARP) (Blum-Kulka, House, & Kasper, 1989). Les résultats de notre recherche ont démontré que les apprenants coréens ont estimé les variables contextuelles, comme l’intimité et le pouvoir social ressentis vis-à-vis de l’interlocuteur dans les situations de demande de façon similaire aux locuteurs natifs francophones, alors qu’ils ont eu tendance à estimer une plus grande gravité d’imposition de la demande que les deux groupes de locuteurs natifs dans plusieurs situations de demande. Quant aux stratégies de demande, tandis que ce sont les stratégies indirectes qui ont été le plus fréquemment utilisées par les apprenants coréens et les locuteurs natifs francophones, les apprenants coréens ont montré une tendance au transfert pragmatique de la L1, en recourant aux stratégies directes plus fréquemment que les locuteurs natifs francophones. De plus, étant donné que les apprenants coréens ont utilisé beaucoup moins d’atténuateurs que les locuteurs natifs francophones, leurs formules de demande étaient plus directes et moins polies que celles employées par les locuteurs natifs francophones. En ce qui concerne les mouvements de soutien (supportive moves), les apprenants coréens se sont démarqués des deux groupes de locuteurs natifs, en recourant plus fréquemment à certaines catégories de mouvements de soutien que les autres répondants. Les analyses de corrélation entre l’estimation des variables contextuelles et le choix des stratégies de politesse nous ont permis de constater que les apprenants coréens ainsi que les locuteurs natifs francophones ont eu tendance à utiliser davantage d’atténuateurs lorsque le locuteur possède moins de pouvoir social. Les implications des résultats pour l’enseignement et l’apprentissage des aspects de la politesse dans la classe de français langue seconde ont été exposées. / The aim of this study is to investigate request strategies that Korean learners of French use in request situations. The role played by the learners’ first language and their culture of origin in making these choices will also be explored. Three groups of respondents participated in this study: 30 native Korean speakers living in Seoul, Korea, 30 Korean French learners in Montreal, Canada, and 30 French native speakers living in Montreal. Data collected from native speakers of both Korean and French were used to create baseline data reflecting first and second language norms respectively. Four data collecting methods were used: the Discourse Completion Test (DCT); an assessment questionnaire evaluating situational factors such as intimacy, the social power between interlocutors and the degree of imposition of requests; the multiple choice questionnaire (MCQ); and a semi-constructed interview. The responses obtained from the written DCT were analysed using the coding scheme elaborated in the Cross-Cultural Speech Act Realization Project (CCSARP) (Blum-Kulka, House, & Kasper, 1989). The results of the study showed that the Korean French learners estimated situational factors like intimacy and the relative social power that they feel with interlocutors in a manner similar to that of the French native speakers. However, in many situations, the Korean learners of French tended to estimate that a request represented a greater imposition than either group of native speakers. Further, while the Korean French learners frequently used indirect strategies as the French native speakers used them, they showed a tendency towards pragmatic transfer from their first language (L1), using direct strategies more frequently than their native French-speaking counterparts. Moreover, as the Korean French learners used downgraders far less frequently, their semantic formulas were more direct and less polite than the ones used by French native speakers. Concerning supportive moves, the Korean French learners showed a different tendency from the two groups of native speakers, in that they used some categories of supportive moves more frequently. The correlation analysis between the results of the assessment questionnaire about situational factors and the politeness strategies used by respondents revealed that the Korean learners and the French native speakers used more downgraders when they have less social power than the hearer. Finally, the implications of these findings as relates to the pedagogical and didactic aspects of politeness in French as a second language classroom are examined.

No description available.

Multipath propagation and mobility in wireless communication systems give rise to variations in the amplitude and phase of the transmitted signal, commonly referred to as fading. Many wireless applications are affected by slowly varying fading, where the channel is non-ergodic, leading to non-reliable transmission during bad channel realizations. These communication scenarios are well modeled by the block-fading channel, where the reliability is quantatively characterized by the outage probability. This thesis focuses on the analysis and design of adaptive transmission schemes to improve the outage performance of both single- and multiple-antenna transmission over the block-fading channel, especially for the cases where discrete input constellations are used. Firstly, a new lower bound on the outage probability of non-adaptive transmission is proposed, providing an efficient tool for evaluating the performance of non-adaptive transmission. The lower bound, together with its asymptotic analysis, is essential for efficiently designing the adaptive transmission schemes considered in the thesis. Secondly, new power allocation rules are derived to minimize the outage probability of fixed-rate transmission over block-fading channels. Asymptotic outage analysis for the resulting schemes is performed, revealing important system design criteria. Furthermore, the thesis proposes novel suboptimal power allocation rules, which enjoy low-complexity while suffering minimal losses as compared to the optimal solution. Thus, these schemes facilitate power adaptation in low-cost devices. Thirdly, the thesis considers incremental-redundancy automatic-repeat-request (INR-ARQ) strategies, which perform adaptive transmission based on receiver feedback. In particular, the thesis concentrates on multi-bit feedback, which has been shown to yield significant gains in performance compared to conventional single-bit ARQ schemes. The thesis proposes a new information-theoretic framework for multi-bit feedback INR-ARQ, whereby the receiver feeds back a quantized version of the accumulated mutual information. Within this framework, the thesis presents an asymptotic analysis which yields the large gains in outage performance offered by multi-bit feedback. Furthermore, the thesis proposes practical design rules, which further illustrates the benefits of multi-bit feedback in INR-ARQ systems. In short, the thesis studies the outage performance of transmission over block-fading channels. Outage analysis is performed for non-adaptive and adaptive transmission. Improvements for the existing adaptive schemes are also proposed, leading to either lower complexity requirements or better outage performance. Still, further research is needed to bring the benefits offered by adaptive transmission into practical systems. / Thesis (PhD)--University of South Australia, 2010

Multipath propagation and mobility in wireless communication systems give rise to variations in the amplitude and phase of the transmitted signal, commonly referred to as fading. Many wireless applications are affected by slowly varying fading, where the channel is non-ergodic, leading to non-reliable transmission during bad channel realizations. These communication scenarios are well modeled by the block-fading channel, where the reliability is quantatively characterized by the outage probability. This thesis focuses on the analysis and design of adaptive transmission schemes to improve the outage performance of both single- and multiple-antenna transmission over the block-fading channel, especially for the cases where discrete input constellations are used. Firstly, a new lower bound on the outage probability of non-adaptive transmission is proposed, providing an efficient tool for evaluating the performance of non-adaptive transmission. The lower bound, together with its asymptotic analysis, is essential for efficiently designing the adaptive transmission schemes considered in the thesis. Secondly, new power allocation rules are derived to minimize the outage probability of fixed-rate transmission over block-fading channels. Asymptotic outage analysis for the resulting schemes is performed, revealing important system design criteria. Furthermore, the thesis proposes novel suboptimal power allocation rules, which enjoy low-complexity while suffering minimal losses as compared to the optimal solution. Thus, these schemes facilitate power adaptation in low-cost devices. Thirdly, the thesis considers incremental-redundancy automatic-repeat-request (INR-ARQ) strategies, which perform adaptive transmission based on receiver feedback. In particular, the thesis concentrates on multi-bit feedback, which has been shown to yield significant gains in performance compared to conventional single-bit ARQ schemes. The thesis proposes a new information-theoretic framework for multi-bit feedback INR-ARQ, whereby the receiver feeds back a quantized version of the accumulated mutual information. Within this framework, the thesis presents an asymptotic analysis which yields the large gains in outage performance offered by multi-bit feedback. Furthermore, the thesis proposes practical design rules, which further illustrates the benefits of multi-bit feedback in INR-ARQ systems. In short, the thesis studies the outage performance of transmission over block-fading channels. Outage analysis is performed for non-adaptive and adaptive transmission. Improvements for the existing adaptive schemes are also proposed, leading to either lower complexity requirements or better outage performance. Still, further research is needed to bring the benefits offered by adaptive transmission into practical systems. / Thesis (PhD)--University of South Australia, 2010

Made available in DSpace on 2016-06-02T19:06:00Z (GMT). No. of bitstreams: 1 4732.pdf: 3415662 bytes, checksum: 0b118a8a41e64b0a67c7b0956df6d970 (MD5) Previous issue date: 2012-09-12 / Universidade Federal de Sao Carlos / Immersive, interactive, and collaborative distributed applications, due to their real time and multi-platform characteristics, require a monitoring structure of their functionalities during the development period, which is not fully available in the current monitoring systems, which are usually specific of the languages and platforms in which they are made available. Monitoring systems that require synchronism of distributed objects during the processing is a complex task. Especially in distributed systems from multi-projection where timing is critical to its operation. It is often difficult to check the correct execution of the system. It might be useful for developers and users of such systems to have the ability of identifying, for example, if the reason for the system delay is due to the abusive use of a resource or due to programming mistakes; the identification of which object has crashed, compromising the system as a whole. Distributed object systems, also known as SOA (Service Oriented Architecture), generally use middleware of the broker type in their infrastructure and can have a dynamic number of connected clients or servers when running. There are, in the literature, several tools for the monitoring of performance, resources, debugging, and support, among others, of such systems. Nevertheless, the tools available in the literature are limited to programming languages or platforms that are able to process the monitoring agents related to them. In this work, a tool for the monitoring of distributed object systems is proposed, based on communication described in XML documents, which allows language and platform independence in the such systems development. A structure for the design of tools for the monitoring of distributed object systems is proposed, based on communication described in XML documents, which allows language and platform independence in the such systems development. In addition to the structure that is theoretically founded, the conception of a tool is structured in the manner described and proven in monitoring data from a case study. / Aplicações Distribuídas Imersivas, Interativas e Colaborativas, dadas suas características de tempo real e multiplataformas, demandam uma estrutura de monitoramento de suas funcionalidades, não plenamente disponíveis nos sistemas atuais de monitoramento, em geral específicos das linguagens e plataformas em que são disponibilizadas. Monitorar sistemas que necessitam de sincronismo no processamento de objetos distribuídos é uma tarefa complexa. Principalmente em sistemas distribuídos e de multiprojeção, onde o sincronismo é fundamental para seu funcionamento. Muitas vezes, é difícil apurar o correto funcionamento dos mesmos. Pode ser útil a desenvolvedores e usuários de tais sistemas a capacidade de se identificar, por exemplo, se a lentidão do sistema está sendo ocasionada por: uso abusivo de um recurso; por um erro na programação de um objeto ou; a identificação de qual objeto travou, comprometendo o sistema como um todo. Sistemas de objetos distribuídos, também conhecidos como SOA (Service Oriented Archtecture), geralmente utilizam mediadores do tipo broker em sua infraestrutura e podem ter um número dinâmico de clientes ou servidores conectados quando estão em execução. Existem na literatura diversas ferramentas destinadas ao monitoramento de performance, recursos, depuração, auxílio, manutenção de tais sistemas, etc. Porém, as ferramentas encontradas na literatura são restritas a linguagens ou plataformas que possam processar os agentes de monitoramento vinculados às mesmas. Neste trabalho, é proposta uma estrutura para a concepção de ferramentas para o monitoramento de sistemas de objetos distribuídos baseados em comunicações descritas em documentos XML, o que permite independência de linguagem ou plataforma na concepção de tais sistemas. Além da estrutura que é teoricamente fundamentada, a concepção de uma ferramenta nos moldes estruturados é descrita e provada no monitoramento de um estudo de caso.

Vortragsfolge: J. Anders Ein H261-Player in Java D. Thie Wavelet- und fraktale Kompression von Videodaten A. Karwath Audiokomprimierung J. Hüttmann Chemnitzer Absolventennetz M. Ehrig Betrachtungen zu verteilten Filesystemen Th. Müller Migration und Datensicherung im MR-AFS M. Clauß Management von Nutzeranforderungen im URZ M. Tautenhahn Systemmanagement auf der Basis von LDAP und SSL S. Brose Zero-Administration-Client mit Ignite-UX und CFENGINE Chr. Ziegler Backupdienst - Stand und Perspektiven Th. Müller BOF: Sicherheit im Campusnetz A. Breiler Server Side Applets N. Schulze Java Beans C. Ranfeld Softwaremanagement RPM vs. OSD G. Heide Neue PD-Software für graphische Benutzeroberflächen J. Wagner Untersuchung zur Neugestaltung der WWW-Seiten zur Softwareausleihe B. Auerbach Digitale Signaturen in MONARCH W. Riedel Lyx - das Wysisyg-LaTeX? W. Riedel BOF: StarOffice - Die vertane Chance? G. Fischer Skalierbarkeit von Serverleistung und Level-4-Switching H. Trapp aktuelle Entwicklungen bei: PGP 5.0, TLS, SSH, Kerberos, vim-5.0 R. Sontag Kommunikationsserver in Schulen: Stand der Vernetzung U. Hübner Perspektiven im Netzbereich

Depuis la fin du XXème siècle, le nombre d'enquêtes et de procédures judiciaires civiles et criminelles internationales a considérablement augmenté à cause de la mondialisation de l'économie et de la facilité de transport de personnes et de biens. Toutefois, les normes de coopération juridique internationale, y compris celles existantes entre la France et le Brésil, n'ont pas accompagné cette évolution. En réalité, l'assistance juridique internationale a toujours faiblement fonctionné à cause, notamment, de la défense rigoureuse de la souveraineté étatique, de l'absence de confiance mutuelle et de coordination entre les pays. La distribution de pouvoirs dans la communauté internationale est faite horizontalement, de sorte que les normes d'assistance entre les États sont marquées par la décentralisation et le particularisme. Dans la plupart des régions du monde, l'accomplissement des demandes d'entraide judiciaire est lent et bureaucratisé, en dépit de la signature de nombreux traités bilatéraux et de conventions régionales et mondiales. L'extradition est encore réalisée presque selon la même formule inventée par les égyptiens et les hittites en 1280 av. J.-C .. La commission rogatoire, dont les racines se retrouvent dans les systèmes de la litterœ publicœ romaine et de la lettre de pareatis française, est encore le principal outil d'entraide civile et pénale. Quelques normes qui peuvent être classées comme de première génération prévoient encore l'utilisation de la voie diplomatique indirecte. Les règles de deuxième génération, créées, notamment, à partir de la fin du XXème siècle, ont amélioré la transmission de demandes, à travers les autorités centrales, mais ne résolvent pas tous les problèmes. En fait, seules les normes de l'Union européenne sont véritablement de troisième génération, parce qu'elles privilégient l'envoi direct de demandes d'entraide civile et pénale, la rapide exécution du mandat d'arrêt européen et l'accomplissement d'une grande quantité de jugements civils sans exequatur ou autre procédure intermédiaire. Les normes de coopération juridique franco-brésiliennes, qui découlent des conventions signées en 1996, sont aujourd'hui dépassées et anachroniques. Les règles de droit interne des deux pays qui visent l'assistance internationale sont franchement plus généreuses et plus efficaces que les règles d'origine conventionnelle. L'élaboration de nouvelles conventions surgit ainsi comme une solution envisageable, malgré les énormes défis politiques et juridiques. Évidemment, les conventions éventuellement parrainées par les Nations Unies ou par la Conférence de La Haye qui traitent de la coopération juridique directe semblent plus appropriées à une résolution plus ample des principales questions. Mais rien n'empêche la conclusion de textes régionaux, inter-régionaux ou même bilatéraux, mais leur étendue serait plus restreinte. / Since the end of the twentieth century, the number of international investigations and legal proceedings in civil and criminal matters has grown considerably due to the globalization of the economy and the ease of transportation of people and goods. Nevertheless, the international legal cooperation rules and regulations, including those between France and Brazil, have not kept up with this evolution. In fact, the international legal cooperation has always been flawed, mostly as a result of the rigorous defense of state sovereignty and the lack of mutual trust and coordination among countries. The authority distribution in the international community is made horizontally, thus the norms of assistance among States are marked by decentralization and particularism. In most regions of the world, the enforcement of legal cooperation claims is slow and bureaucratized, despite numerous treaties and bilateral, regional or international conventions. Extradition is still carried out in accordance with practically the same formula invented by the Egyptians and Hittites in 1280 B.C .. The letter of request, which springs from the Roman litterœ publicœ and the French lettre de pareatis, is still the major tool of civil and criminal legal cooperation. Some standards, regarded as first-generation, stipulate the use of indirect diplomatic ways. The second-generation norms established mostly at the end of the twentieth century have improved the transmission of demands through central authorit1es but have not solved all the problems. In fact, only the European Union norms are truly third-generation, since they favor the straight submission of civil and criminal legal cooperation claims, the swift execution of the European arrest warrant and the execution of a large number of civil judgments, with no exequatur or any other intermediate procedure. The French-Brazilian legal cooperation rules, which stem from conventions signed in 1996, are outdated and anachronistic. The rules of internal law of the countries that pursue international cooperation are utterly more generous and effective than those of conventional origin. The elaboration of new conventions thus arises as a significant solution, despite the enormous political and juridical challenges. Naturally, prospective conventions sponsored by the United Nations or by The Hague Convention concerning: direct legal cooperation seem more suitable to a comprehensive solution of the main issues. However, nothing precludes the conclusion of regional, interregional and bilateral texts, even though these will have a smaller reach.

Notre thèse a pour objectif la construction d’un système omni-langage de recherche de mots dans les documents numérisés. Nous nous plaçons dans le contexte où le contenu du document est homogène (ce qui est le cas pour les documents anciens où l’écriture est souvent bien soignée et mono-scripteur) et la connaissance préalable du document (le langage, le scripteur, le type d’écriture, le tampon, etc.) n’est pas connue. Grâce à ce système, l'utilisateur peut composer librement et intuitivement sa requête et il peut rechercher des mots dans des documents homogènes de n’importe quel langage, sans détecter préalablement une occurrence du mot à rechercher. Le point clé du système que nous proposons est les invariants, qui sont les formes les plus fréquentes dans la collection de documents. Pour le requêtage, l’utilisateur pourra créer le mot à rechercher en utilisant les invariants (la composition des requêtes), grâce à une interface visuelle. Pour la recherche des mots, les invariants peuvent servir à construire des signatures structurelles pour représenter les images de mots. Nous présentons dans cette thèse la méthode pour extraire automatiquement les invariants à partir de la collection de documents, la méthode pour évaluer la qualité des invariants ainsi que les applications des invariants à la recherche de mots et à la composition des requêtes. / The objective of our thesis is to build an omni-language word retrieval system for scanned documents. We place ourselves in the context where the content of documents is homogenous and the prior knowledge about the document (the language, the writer, the writing style, etc.) is not known. Due to this system, user can freely and intuitively compose his/her query. With the query created by the user, he/she can retrieve words in homogenous documents of any language, without finding an occurrence of the word to search. The key of our proposed system is the invariants, which are writing pieces that frequently appeared in the collection of documents. The invariants can be used in query making process in which the user selects and composes appropriate invariants to make the query. They can be also used as structural descriptor to characterize word images in the retrieval process. We introduce in this thesis our method for automatically extracting invariants from document collection, our evaluation method for evaluating the quality of invariants and invariant’s applications in the query making process as well as in the retrieval process.