<p>The purpose of this experimental study is to examine the translation processes and the resulting translations of eight translators performing translation tasks from German into Swedish. The data about the translation processes were collected with the think-aloud technique, i.e. the translators were asked to verbalize as much of their thinking as possible while translating. The translators – four of whom had long working experience and the others substantially shorter experience – were asked to translate the same source text (a news item) for two different fictitious audiences: first for a morning newspaper and then for a children’s magazine.</p><p>With regards to the translation process, the results showed that the two categories of translators did not differ considerably from each other, neither in their dictionary use, nor in their overall accomplishment of the translation tasks. However, they did differ in their attitude. The less experienced translators were more involved in the process, which also led to quality improvements. At the same time, individual differences were substantial regarding all of the criteria examined.</p><p>The resulting translations (the products) were evaluated both by the researcher and by journalists from the newspapers or magazines in which the texts, according to the briefs, were to be published. The results showed small differences in text quality for the two categories of translators. According to the journalists, the less experienced translators even produced slightly better texts for the morning paper.</p>

The purpose of this experimental study is to examine the translation processes and the resulting translations of eight translators performing translation tasks from German into Swedish. The data about the translation processes were collected with the think-aloud technique, i.e. the translators were asked to verbalize as much of their thinking as possible while translating. The translators – four of whom had long working experience and the others substantially shorter experience – were asked to translate the same source text (a news item) for two different fictitious audiences: first for a morning newspaper and then for a children’s magazine. With regards to the translation process, the results showed that the two categories of translators did not differ considerably from each other, neither in their dictionary use, nor in their overall accomplishment of the translation tasks. However, they did differ in their attitude. The less experienced translators were more involved in the process, which also led to quality improvements. At the same time, individual differences were substantial regarding all of the criteria examined. The resulting translations (the products) were evaluated both by the researcher and by journalists from the newspapers or magazines in which the texts, according to the briefs, were to be published. The results showed small differences in text quality for the two categories of translators. According to the journalists, the less experienced translators even produced slightly better texts for the morning paper.

Viva la Diva är ett Svenskt makeup-varumärke som nyligen har lanserat sin nya webbplats. Denna webbplats har inte blivit testad på deras slutanvändare eller målgrupp. Genom att använda en heuristisk utvärdering och 12 Think aloud-tester har flera användbarhetsproblem upptäckts där designen behöver förbättringar. Senare i rapporten presenteras några förbättringsförslag om hur användarupplevelsen kan förbättras på webbplatsen, men också hur problemen som jag fann under utvärderingen och Think aloud-testerna kan lösas. Jag tillhandahåller också för Viva la Diva ett designförslag för en av deras sidor som inte är helt färdig i designen. Detta arbete ger insikt i hur man kan använda en heuristisk utvärdering och Think aloudtester för att upptäcka användbarhetsproblem på webbplatser. / Viva la Diva is a Swedish makeup-brand that has recently launched their new website. This website has not been tested with their end-users or target users. By using a heuristic evaluation and 12 Think aloud-tests several user problems have been discovered where the design needs improvements. Later in the article I present some suggestions on how to improve the user experience on the website, as well as fix the problems that I found during the evaluation and Think aloud-tests. I also provide Viva la Diva with a design suggestion for one of their pages that is not completely finished in the design. This work provides insights on how to use a heuristic evaluation and Think aloud-tests to discover usability problems on websites.

As the use of mobile payment applications (apps) and electronic wallets (e-wallets)increases, so does the demand for a improved user experience when interactingwith these apps. The field of Human-Computer interaction (HCI) focuses onthe design, evaluation, and implementation of interactive computing systems forhuman use. One aspect of HCI is usability, i.e., the quality of the interactions witha product or system. This thesis investigates how an e-wallet can be designed to provide a high level ofusability by conforming to best HCI practices and by formative evaluation using aset of usability evaluation methods.The research process consisted of an initial literature study and developmentof a prototype, which was evaluated iteratively through Thinking-aloud-protocol(TAP) and a combination of performance measurements and questionnaire by achosen test group. By each iteration, the results of the performance measurements, as well as theverbal data improved. The most complex or difficult task, for the test subjectsto perform, was, according to the results, Pay via manual input. All goals wereachieved for all tasks except for the performance goal of a percentage of errorsbelow 5%. To conclude, it was clear that the test subjects had more trouble understandingthe concept of the e-wallet rather than navigating and completing tasks. Thedifficulties lay in understanding how currencies were stored and how transactionshappened. When developing this e-wallet we noticed that the most importantissue was to make new functions and concepts familiar to the user through relatingit to recognizable ideas. / I och med att användning av mobila betalningslösningar (appar) och elektroniskaplånböcker (e-plånböcker) ökar, ökar även efterfrågan på en förbättradanvändarupplevelse vid interaktion med dessa appar. Området människadatorinteraktion(MDI) fokuserar på design, utvärdering och implementeringav interaktiva datorsystem för mänsklig användning. En aspekt av MDI äranvändbarhet, dvs kvaliteten på interaktionerna med en produkt eller ett system.Detta kandidatexamensarbete undersöker hur en e-plånbok kan utformas föratt ge en hög användbarhet genom att anpassas till MDI praxis och formativutvärdering av designen med hjälp av en uppsättning utvärderingsmetoder föranvändbarhet. Forskningsprocessen bestod av en litteraturstudie och utveckling av en prototoyp,som utvärderades iterativt genom Thinking-aloud Protocol (TAP) samt enkombination av prestationsmätningar och frågeformulär av en vald testgrupp.Efter varje iteration förbättrades resultaten av prestationsmätningarna, såvälsom för den verbala datan. Den mest komplexa eller svåra uppgiften,för testpersonerna att utföra, var, enligt resultaten, Betalning via manuelinmatning. Alla mål uppnåddes för alla uppgifter förutom prestationsmålet fören procentandel av fel under 5%. Avslutningsvis var det tydligt att testpersonerna fann det svårare att förståkonceptet av e-plånboken än att navigera och slutföra uppgifterna. Svårigheternalåg i att förstå hur valutor lagrades och hur transaktioner gick till. När viutvecklade den här e-plånboken märkte vi att den viktigaste uppgiften var att göranya funktioner och koncept förståerliga för användaren genom att koppla dem tilligenkännliga idéer.

One important phase in software development process is to create a design model of the system which follows all the architectural rules. Often the architectural rules are defined by the system architect and the system model is designed by the system designer. The architect defines the rules in a text file where no standard or pattern is followed. Therefore, there is always the risk of violating the architectural rules by the designer. So manual reviews on the system model should be done by the architect to ensure the system model is valid.In order to remove this manual checking which can be erroneous and time consuming ArCon (Architecture Conformance Checker) was developed by Combitech AB. ArCon is a tool which lets the architect define the architectural rules in the format of UML (Unified Modeling Language) models where the elements of the model have different meaning than the standard UML. ArCon can read this model and extract architectural rules from it and check the system model against those rules and then print all the rule violations.ArCon is an open source tool i.e. free for everyone to download and use. Currently, it supports Papyrus as the UML modeling tool. Papyrus is integrated to Eclipse platform and is a general purpose modeling tool. It supports users with all types of UML diagrams and elements.The idea for this thesis work was to implement a new feature for ArCon in order to facilitate the design process for system designers. The feature should provide the system designers only those types of elements which they are permitted to add to a specific fraction of the system model. The list of permitted element types should be extracted from the architecture model where all the architectural rules are defined in advance. This new support in ArCon was named Domain Specific Modeling (DSM) support.To evaluate the effect of DSM support on the system designers performance a few test sessions, called usability tests, were performed. The participants in the test sessions were a representative sample of software designers. After analyzing the data collected from the test sessions, the pros and cons of the new support were discovered. Furthermore, a few new ideas for enhancing DSM support were generated.

This diploma thesis discusses a modern concept of information retrieval and search engines for libraries and other academic institutions. The concept named "web scale discovery" comprises of search engines, whose main characteristics lie in simplicity and user friendliness for the end users while maintaining all functional qualities of traditional research databases. Users can search in a wide variety of international research databases, and also in local sources of an institution that are combined within a large central index. The theoretical section presents definitions of web scale discovery, which conceptually set the new method of information retrieval within the field of information and library science. A graphic scheme of the basic functionality of a web scale discovery system is presented. Also discussed are requirements for a modern discovery system, an overview of the current situation in the Czech Republic, and a short characteristic of commercially available discovery systems. The theoretical part concludes with a literature review of selected foreign research, studying user satisfaction with the new solution, the impact on electronic and print resources in libraries, and usability testing. The practical part presents an original research study - usability testing of EBSCO Discovery...