Web Information Systems and Technologies

Unlike traditional single-PC applications, which have access to directly attached computational resources (CPUs, memory, and I/O devices), web applications have to deal with the resources scattered across the network. Besides, web applications are intended to be accessed by multiple users simultaneously. That not only requires a more sophisticated infrastructure but also brings new challenges to web application developers.

The webAppOS platform is an operating system analog for web applications. It factors out the network and provides the illusion of a single computer, the “web computer”. That illusion allows web application developers to focus on business logic and create web applications faster. Besides, webAppOS standardizes many aspects of web applications and has the potential to become a universal environment for them.

The quantify-everything trend has reached the automotive sector while digitalization is a still the major driver of innovation. New digital services based on vehicle usage data are being created for different actors and purposes, e.g. for individual drivers who want to know about their own driving style and behavior or for fleet managers who want to find out about their fleet. As a side effect, a growing number of ICT start-ups from outside Europe have entered the automotive market to work on innovative use cases. Their digital services are based on the availability of vehicle data on a large scale. To better understand and capture this ongoing digital change in the automotive sector, we present an extended version of the Vehicle Data Value Chain (VDVC) originally published in Kaiser et al. (2019a) and use it as a model for better structuring, describing and testing digital services based on vehicle usage data. We classify digital services of two projects by using the VDVC in our paper, an intermodal mobility service and a pothole and driving style detection service. Thus, we evaluate the VDVC and show its general applicability and usefulness in a practical context.

For years, the manufacturing industry has been investing substantial amounts of research and development work for the implementation of hybrid teams of human workers and robotic units. The composition of hybrid teams requires an optimal coordination of individual players with fundamentally different characteristics and skills. In this paper, we present a highly configurable simulation environment supporting end-users, e.g. manufacturing planners, to optimally prepare, evaluate and improve the collaboration of hybrid teams in the scope of production lines. For generating the optimal task assignment, a GPU-based high-performance optimizer is introduced into the simulation environment. The framework is embedded in a web-based distributed infrastructure that models and provides the involved components (digital human models, robots, visualization environment) as resources. We illustrate the approach with a use case originating from the aircraft industry.

For the last decades text-based templates have been the primary option to build dynamic web pages. Until today, no other alternative has rebutted this approach. Yet, using text-based templates has several drawbacks including: 1. blocking resolution, 2. programming languages heterogeneity, 3. limited set of templating features and 4. opinionated idioms. In this paper we show how a domain specific language (DSL) for HTML (such as HtmlFlow, Kotlinx.html or React JSX) can suppress the text-based templates limitations and outperform state-of-the-art template engines (such as JSP, Handlebars, Thymeleaf, and others) in well known benchmarks. To that end, we use the Spring Framework and the sample application PetClinic to show how a DSL for HTML provides unopinionated web templates with boundless resolving features only ruled by the host programming language (such as Java, Kotlin or JavaScript).

Modern versions of the HTTP protocol, such as HTTP/2 over TCP and the upcoming HTTP/3 over QUIC, use just a single underlying connection to transfer multiple resources during a web page load. The resources are divided into chunks, optionally multiplexed on the connection, and reassembled at the receiver’s side. This poses challenges, as there are many different ways simultaneously requested resources can share the available bandwidth, and not all approaches perform equally well with regards to achieving low loading times. Making matters worse, HTTP/2’s prioritization system for directing this multiplexing behaviour is difficult to use and does not easily transfer to the new HTTP/3.

In this work, we discuss these challenges in detail and empirically evaluate the multiplexing behaviours of 10 different QUIC implementations, as well as 11 different prioritization schemes for HTTP/3. We find that there are large differences between strategies that can have a heavy impact on page load performance, of up to 5x load time speedup in specific conditions. However, these improvements are highly context-sensitive, depending on web page composition and network conditions, turning the best performers for one setup into the worst for others. As such, we also critically evaluate the ability of the newly proposed HTTP/3 prioritization mechanism to flexibly deal with changing conditions.

More and more, researchers in humanities and companies need large classified document data-sets. These users are not familiar with information retrieval or data science notions. For data scientists, there is also often a need for those classified document data-sets as ground truth. There are multiple tools that allow users to carry out this classification task on large data-sets, involving always a quite expert level in computer and data science. More over, these tools are not usually oriented to the domain of micro-blogs or do not always take into account meta data and attached images as additional dimensions to improve the classification. In this work, we present a platform to enable end users to classify large document collections of several hundred thousands documents in an assisted way, within a humanly acceptable number of clicks, with no coding and without having data science and information retrieval expert knowledge. The system includes a graphical user interface with several classification assistants doing text- and image-based event detection, geographical filtering, image clustering, search services with rich visual metaphors to visualize their results and finally Active Learning (AL) with different sampling strategies. We also present a comparative study on the impact of using different and interchangeable AL components on the number of clicks needed to reach a stable level of accuracy.

With the rise of mobile devices, users operate applications in a large variety of contexts. In each of these contexts, a user may have different requirements and preferences regarding an application’s user interface. The context describes the current physical and social environment of the user, his activity and locomotion, as well as the current location and time. Hence, different user interfaces may be more suitable in specific contexts than others. At the moment, a user interface is an integral part of an application, consequently also limiting its usefulness in certain contexts as the information is not presented in the best possible way. Therefore, we propose a way to decouple the user interface from a specific form of representation. Through this decoupling, it is possible to dynamically adapt user interfaces to the user’s specific needs in a context, to increase the value of the application for the user. In this paper, we introduce a general framework for the context-dependent adaptation of user interfaces and evaluate it in the specific context of travel information systems. Travel information systems are particularly suited to evaluate such a framework, as they are usually operated in many different contexts – before a trip, during a trip, and after a trip. The adaptation framework transforms between system-oriented messages and user-oriented messages. The user’s context, the output device capabilities, and the user’s preference, all influence the choice of the actual representation for user-oriented messages. We implemented a prototype of the proposed system and conducted an experimental evaluation focusing on scenarios from the domain of travel information systems.

Social media platforms have become important channels of promotional communication from marketers to users with the goal of enhancing purchase intention. One important element of the communication mix is paid advertising on social media. In this context, the interplay between general attitude towards advertising on social media and specific attitude towards an individual advertisement as drivers of purchase intention has not yet been addressed in extant social media research. This study investigates this issue by distinguishing between these two levels of attitude and examining their antecedents as well as their impacts on purchase intention. The empirical test of the nomological network with survey data shows that platform enjoyment is an antecedent of general attitude towards social media advertising whereas perceived meaningfulness and brand awareness drive specific attitude towards the individual advertisement. Perceived intrusiveness has no impact in a social media setting. Specific attitude shows a stronger impact on purchase intention than general attitude although both are significant. The findings point at the need for a clear conceptual differentiation between both levels of attitudes and the role of their antecedents for an appropriate design of paid social media advertisements.