Im Volltext enthalten sind ausgewählte Abstracts der Vorträge, die im Rahmen des 3. Dresdner Werkstoffsymposium gehalten wurden.

The fast and furious pace of change in computing technology has become an article of faith for many. The reliability of computer-based systems cru- cially depends on the correctness of its computing. Can man, who created the computer, be capable of preventing machine-made misfortune? The theory of static analysis strives to achieve this ambition. The analysis of numerical properties of programs has been an essential research topic for static analysis. These kinds of properties are commonly modeled and handled by the concept of numerical abstract domains. Unfor- tunately, lifting these domains to heap-manipulating programs is not obvious. On the other hand, points-to analyses have been intensively studied to an- alyze pointer behaviors and some scale to very large programs but without inferring any numerical properties. We propose a framework based on the theory of abstract interpretation that is able to combine existing numerical domains and points-to analyses in a modular way. The static numerical anal- ysis is prototyped using the SOOT framework for pointer analyses and the PPL library for numerical domains. The implementation is able to analyze large Java program within several minutes. The second part of this thesis consists of a theoretical study of the com- bination of the points-to analysis with another pointer analysis providing information called must-alias. Two pointer variables must alias at some pro- gram control point if they hold equal reference whenever the control point is reached. We have developed an algorithm of quadruple complexity that sharpens points-to analysis using must-alias information. The algorithm is proved correct following a semantics-based formalization and the concept of bisimulation borrowed from the game theory, model checking etc.

A domain-specific embedded language (DSEL) is a domain-specific programming language with no concrete syntax of its own. Defined as a set of combinators encapsulated in a module, it borrows the syntax and tools (such as type-checkers and compilers) of its host language; hence it is economical to design, introduce, and maintain. Unfortunately, this economy is counterbalanced by a lack of room for growth. DSELs cannot match sophisticated domain-specific languages that offer tools for domainspecific error-checking and optimisation. These tools are usually based on syntactic analyses, so they do not work on DSELs. Abstract interpretation is a technique ideally suited to the analysis of DSELs, due to its semantic, rather than syntactic, approach. It is based upon the observation that analysing a program is equivalent to evaluating it over an abstract semantic domain. The mathematical properties of the abstract domain are such that evaluation reduces to solving a mutually recursive set of equations. This dissertation shows how abstract interpretation can be applied to a DSEL by replacing it with an abstract implementation of the same interface; evaluating a program with the abstract implementation yields an analysis result, rather than an executable. The abstract interpretation of DSELs provides a foundation upon which to build sophisticated error-checking and optimisation tools. This is illustrated with three examples: an alphabet analyser for CSP, an ambiguity test for parser combinators, and a definedness test for attribute grammars. Of these, the ambiguity test for parser combinators is probably the most important example, due to the prominence of parser combinators and their rather conspicuous lack of support for the well-known LL(k) test. In this dissertation, DSELs and their signatures are encoded using the polymorphic lambda calculus. This allows the correctness of the abstract interpretation of DSELs to be proved using the parametricity theorem: safety is derived for free from the polymorphic type of a program. Crucially, parametricity also solves a problem commonly encountered by other analysis methods: it ensures the correctness of the approach in the presence of higher-order functions.

The human central auditory system can automatically extract abstract regularities from a variant auditory input. To this end, temporarily separated events need to be related. This study tested whether the timing between events, falling either within or outside the temporal window of integration (~350 ms), impacts the extraction of abstract feature relations. We utilized tone pairs for which tones within but not across pairs revealed a constant pitch relation (e.g., pitch of second tone of a pair higher than pitch of first tone, while absolute pitch values varied across pairs). We measured the mismatch negativity (MMN; the brain’s error signal to auditory regularity violations) to second tones that rarely violated the pitch relation (e.g., pitch of second tone lower). A Short condition in which tone duration (90 ms) and stimulus onset asynchrony between the tones of a pair were short (110 ms) was compared to two conditions, where this onset asynchrony was long (510 ms). In the Long Gap condition, the tone durations were identical to Short (90 ms), but the silent interval was prolonged by 400 ms. In Long Tone, the duration of the first tone was prolonged by 400 ms, while the silent interval was comparable to Short (20 ms). Results show a frontocentral MMN of comparable amplitude in all conditions. Thus, abstract pitch relations can be extracted even when the within-pair timing exceeds the integration period. Source analyses indicate MMN generators in the supratemporal cortex. Interestingly, they were located more anterior in Long Gap than in Short and Long Tone. Moreover, frontal generator activity was found for Long Gap and Long Tone. Thus, the way in which the system automatically registers irregular abstract pitch relations depends on the timing of the events to be linked. Pending that the current MMN data mirror established abstract rule representations coding the regular pitch relation, neural processes building these templates vary with timing.

Master of Philosophy / Grace Crowley was one of the leading innovators of geometric abstraction in Australia. When she returned to Australia in 1930 she had thoroughly mastered the complex mathematics and geometry of the golden section and dynamic symmetry that had become one of the frameworks for modernism. Crowley, Anne Dangar and Dorrit Black all studied under the foremost teacher of modernism in Paris, André Lhote. Crowley not only taught the golden section and dynamic symmetry to Rah Fizelle, Ralph Balson and students of the Crowley-Fizelle Art School, but used it to develop her own abstract art during the 1940s and 1950s, well in advance of the arrival of colour-field painting to Australia in the 1960s. Through her teaching at the most progressive modern art school in Sydney in the 1930s Crowley taught the basic compositional techniques as she had learnt them from Lhote. When the art school closed in 1937 she worked in partnership with fellow artist, Ralph Balson as they developed their art into constructive, abstract paintings. Balson has been credited with being the most influential painter in the development of geometric abstraction in Australia for a younger generation of artists. This is largely due to Crowley’s insistence that Balson was the major innovator who led her into abstraction. She consistently refused to take credit for her own role in their artistic partnership. My research indicates that there were a number of factors that strongly influenced Crowley to support Balson and deny her own role. Her archives contain sensitive records of the breakup of her partnership with Rah Fizelle and the closure of the Crowley-Fizelle Art School. These, and other archival material, indicate that Fizelle’s inability to master and teach the golden section and dynamic symmetry, and Crowley’s greater popularity as a teacher, was the real cause of the closure of the School. Crowley left notes in her Archives that she still felt deeply distressed, even forty years after the events, and did not wish the circumstances of the closure known in her lifetime. With the closure of the Art School and her close friend Dangar living in France, her friendship with Balson offered a way forward. This thesis argues that Crowley chose to conceal her considerable mathematical and geometric ability, rather than risk losing another friend and artistic partner in a similar way to the breakup of the partnership with Fizelle. With the death of her father in this period, she needed to spend much time caring for her mother and that left her little time for painting. She later also said she felt that a man had a better chance of gaining acceptance as an artist, but it is equally true that, without Dangar, she had no-one to give her support or encourage her as an artist. By supporting Balson she was able to provide him with a place to work in her studio and had a friend with whom she could share her own passion for art, as she had done with Dangar. During her long friendship with Balson, she painted with him and gave him opportunities to develop his talents, which he could not have accessed without her. She taught him, by discreet practical demonstration the principles she had learnt from Lhote about composition. He had only attended the sketch club associated with the Crowley- Fizelle Art School. Together they discussed and planned their paintings from the late 1930s and worked together on abstract paintings until the mid-1950s when, in his retirement from house-painting, she provided him with a quiet, secluded place in which to paint and experiment with new techniques. With her own artistic contacts in France, she gained him international recognition as an abstract painter and his own solo exhibition in a leading Paris art gallery. After his death in 1964, she continued to promote his art to curators and researchers, recording his life and art for posterity. The artist with whom she studied modernism in Paris, Anne Dangar, also received her lifelong support and promotion. In the last decade of her life Crowley provided detailed information to curators and art historians on the lives of both her friends, Dangar and Balson, meticulously keeping accurate records of theirs and her own life devoted to art. In her latter years she arranged to deposit these records in public institutions, thus becoming a contributor to Australian art history. As a result of this foresight, the stories of both her friends, Balson and Dangar, have since become a record of Australian art history. (PLEASE NOTE: Some illustrations in this thesis have been removed due to copyright restrictions, but may be consulted in the print version held in the Fisher Library, University of Sydney. APPENDIX 1 gratefully supplied from the Grace Crowley Archives, Art Gallery of New South Wales Research Library)

Doctor of Philosophy / The rapid increase in available information has lead to many attempts to automatically locate patterns in large, abstract, multi-attributed information spaces. These techniques are often called data mining and have met with varying degrees of success. An alternative approach to automatic pattern detection is to keep the user in the exploration loop by developing displays for perceptual data mining. This approach allows a domain expert to search the data for useful relationships and can be effective when automated rules are hard to define. However, designing models of the abstract data and defining appropriate displays are critical tasks in building a useful system. Designing displays of abstract data is especially difficult when multi-sensory interaction is considered. New technology, such as Virtual Environments, enables such multi-sensory interaction. For example, interfaces can be designed that immerse the user in a 3D space and provide visual, auditory and haptic (tactile) feedback. It has been a goal of Virtual Environments to use multi-sensory interaction in an attempt to increase the human-to-computer bandwidth. This approach may assist the user to understand large information spaces and find patterns in them. However, while the motivation is simple enough, actually designing appropriate mappings between the abstract information and the human sensory channels is quite difficult. Designing intuitive multi-sensory displays of abstract data is complex and needs to carefully consider human perceptual capabilities, yet we interact with the real world everyday in a multi-sensory way. Metaphors can describe mappings between the natural world and an abstract information space. This thesis develops a division of the multi-sensory design space called the MS-Taxonomy. The MS-Taxonomy provides a concept map of the design space based on temporal, spatial and direct metaphors. The detailed concepts within the taxonomy allow for discussion of low level design issues. Furthermore the concepts abstract to higher levels, allowing general design issues to be compared and discussed across the different senses. The MS-Taxonomy provides a categorisation of multi-sensory design options. However, to design effective multi-sensory displays requires more than a thorough understanding of design options. It is also useful to have guidelines to follow, and a process to describe the design steps. This thesis uses the structure of the MS-Taxonomy to develop the MS-Guidelines and the MS-Process. The MS-Guidelines capture design recommendations and the problems associated with different design choices. The MS-Process integrates the MS-Guidelines into a methodology for developing and evaluating multi-sensory displays. A detailed case study is used to validate the MS-Taxonomy, the MS-Guidelines and the MS-Process. The case study explores the design of multi-sensory displays within a domain where users wish to explore abstract data for patterns. This area is called Technical Analysis and involves the interpretation of patterns in stock market data. Following the MS-Process and using the MS-Guidelines some new multi-sensory displays are designed for pattern detection in stock market data. The outcome from the case study includes some novel haptic-visual and auditory-visual designs that are prototyped and evaluated.

Doctor of Philosophy / The rapid increase in available information has lead to many attempts to automatically locate patterns in large, abstract, multi-attributed information spaces. These techniques are often called data mining and have met with varying degrees of success. An alternative approach to automatic pattern detection is to keep the user in the exploration loop by developing displays for perceptual data mining. This approach allows a domain expert to search the data for useful relationships and can be effective when automated rules are hard to define. However, designing models of the abstract data and defining appropriate displays are critical tasks in building a useful system. Designing displays of abstract data is especially difficult when multi-sensory interaction is considered. New technology, such as Virtual Environments, enables such multi-sensory interaction. For example, interfaces can be designed that immerse the user in a 3D space and provide visual, auditory and haptic (tactile) feedback. It has been a goal of Virtual Environments to use multi-sensory interaction in an attempt to increase the human-to-computer bandwidth. This approach may assist the user to understand large information spaces and find patterns in them. However, while the motivation is simple enough, actually designing appropriate mappings between the abstract information and the human sensory channels is quite difficult. Designing intuitive multi-sensory displays of abstract data is complex and needs to carefully consider human perceptual capabilities, yet we interact with the real world everyday in a multi-sensory way. Metaphors can describe mappings between the natural world and an abstract information space. This thesis develops a division of the multi-sensory design space called the MS-Taxonomy. The MS-Taxonomy provides a concept map of the design space based on temporal, spatial and direct metaphors. The detailed concepts within the taxonomy allow for discussion of low level design issues. Furthermore the concepts abstract to higher levels, allowing general design issues to be compared and discussed across the different senses. The MS-Taxonomy provides a categorisation of multi-sensory design options. However, to design effective multi-sensory displays requires more than a thorough understanding of design options. It is also useful to have guidelines to follow, and a process to describe the design steps. This thesis uses the structure of the MS-Taxonomy to develop the MS-Guidelines and the MS-Process. The MS-Guidelines capture design recommendations and the problems associated with different design choices. The MS-Process integrates the MS-Guidelines into a methodology for developing and evaluating multi-sensory displays. A detailed case study is used to validate the MS-Taxonomy, the MS-Guidelines and the MS-Process. The case study explores the design of multi-sensory displays within a domain where users wish to explore abstract data for patterns. This area is called Technical Analysis and involves the interpretation of patterns in stock market data. Following the MS-Process and using the MS-Guidelines some new multi-sensory displays are designed for pattern detection in stock market data. The outcome from the case study includes some novel haptic-visual and auditory-visual designs that are prototyped and evaluated.

Thesis (Ph. D.)--University of Sydney, 2003. / Title from title screen (viewed April 6, 2009). Submitted in fulfilment of the requirements for the degree of Doctor of Philosophy to the School of Information Technologies, Faculty of Science. Includes bibliographical references. Also available in print form.

Cette thèse revisite de deux manières le domaine abstrait des polyèdres utilisé pour l'analyse statique de programmes.D'abord, elle montre comment utiliser l'assistant à la preuve Coq pour apporter des garanties sur la correction des opérations sur les polyèdres sans compromettre l'efficacité de l'outil VP Lissu de ces travaux.L'outil est fondé sur le principe de la vérification de résultats :un oracle, auquel on ne fait pas confiance, fait les calculs,puis les résultats sont vérifiés par un validateur dont la correction est prouvée avec Coq. De plus, l'oracle fournit des témoins de la correction des résultats afin d'accélérer la vérification.L'autre caractéristique de VPL est l' utilsation de la seule représentation par contraintes des polyèdres,par opposition à l'approche habituelle qui consiste à utiliser à la fois des contraintes et des générateurs.Malgré ce choix inhabituel,les performances de VPL s'avèrent compétitives.Comme on pouvait le prévoir,l'opérateur "join",qui calcule l'enveloppe convexe de deux polyèdres,est le plus coûteux.Puisqu'il nécessite un grand nombre de projections,cette thèse explore plusieurs nouvelles approches de l'opérateur de projection,basées sur la programmation linéaire paramétrique.Elle propose une synthèse des variantes et des combinaisons possibles.La thèse se termine sur les éléments clés d'un nouvel algorithme de résolution tirant parti des spécificités de l'encodage afin d'obtenir de bonnes performances. / The work reported in this thesis revisits in two waysthe abstract domain of polyhedraused for static analysis of programs.First, strong guarantees are provided on the soundness of the operationson polyhedra,by using of the Coq proof assistant to check the soundness proofs.The means used to ensure correctnessdon't hinder the performance of the resultingVerimag Polyhedra Library (VPL).It is built on the principle of result verification:computations are performed by an untrusted oracleand their results are verified by a checkerwhose correctness is proved in Coq.In order to make verification cheap,the oracle computes soundness witnesses along with the results.The other distinguishing feature of VPL is thatit relies only on the constraint representation of polyhedra,as opposed to the common practice of using both constraints and generators.Despite this unusual choice,VPL turns out to be a competitive abstract domain of polyhedra,performance-wise.As expected, the join operator of VPL,which performs the convex hull of two polyhedra,is the costliest operator.Since it builds on the projection operator,this thesis also investigates a new approach toperforming projections,based on parametric linear programming.A new understanding of projection encoded asa parametric linear problem is presented.The thesis closes on a progress report in the design of a new solvingalgorithm,tailored to the specifics of the encodingso as to achieve good performance.

Detta kandidatarbete har som syfte att belysa rollen som glitch kan spela i skapande. Jag ställer mig frågan vad som händer när vi för in ett medvetet brott i en kreativ handling, hur vi behandlar det vi uppfattar som fel eller misstag, och vad som kan hända när vi utforskar det som blivit förkastat som trasigt och obsolet. Dessa frågor har uppstått dels ur en nyfikenhet för det okända och ett ständigt sökande efter nya uttrycksformer, men även ur kritik. Den teknologiska utvecklingen min generation befinner sig i för oss allt hastigare framåt, och jag ifrågasätter hur vi förhåller oss till värdet av medietekniska artefakter när konsumtionshetsen gör sig alltmer påmind. Genom att ägna mig åt studier av glitch och konstformer som uppstått därifrån, second hand och dess kultur samt mediearkeologi har jag bildat mig en grund att stå på som sedan lett mig genom mitt gestaltningsarbete. Med ett kritiskt förhållningssätt har jag använt mig av icke-konventionella metoder och program för att redigera och manipulera media, och reflekterat över resultaten med stöd av forskning och diskussion med andra människor. Jag har utforskat digital glitch i medieformer som jag under mina tre år av studier har bekantat mig med, men även vandrat över fysiska kretskort och undersökt vad som händer när ett konkret brott introduceras i deras förväntade flöde. Resultatet har blivit ett nytt förhållningssätt till utdaterad teknik, en ny vinkel att betrakta kreativitet och skapande från och förhoppningsvis en ökad respekt för vår fortsatta tekniska utveckling. / This bachelor thesis is intended to highlight the role that glitch can have in creation. I pose the question of what happens when we introduce a deliberate disruption in a creative act? How do we treat what we perceive as flaws and mistakes, and what can happen when we explore what has once been rejected as broken and obsolete? These questions have arisen partly from the curiosity about the unknown and a constant search for new forms of expression, but also from criticism. The technological development my generation is experiencing is moving us forward at an ever increasing pace, and it leaves me inquisitive about how we relate to the value of media artifacts when consumerism is becoming more and more prevalent. By studying glitches and their resulting art forms, second-hand and its culture, as well as media archeology, I have formed a foundation on which to base my work. With a critical approach, I have used non-conventional methods and programs to edit and manipulate media, and reflected upon the results supported by both research and discussion with other people. I have explored digital glitches in media that I've worked with during my three years of study, but also traversed physical circuit boards and investigated what happens when a break is introduced in their expected flow. The result is a new approach to outdated technology, a new angle from which to view creativity and creation, and​—​hopefully​—​a shift in respect for our continued technological development.