Abstract
In the early days of the world wide web, browsers were developed to navigate through (static) HTML web page documents. This has changed dramatically, and nowadays web pages are dynamic, expressed by programs written in regular programming languages. As a result, browsers are almost operating systems, having to interpret/compile such programs and execute them within the browser itself. Currently, while JavaScript is the main de facto language to express web pages, it does have various short comings and performance inefficiencies. WebAssembly, a new portable and size/load efficient alternative developed by major IT powerhouses, is seen as the future substitute. As WebAssembly aims to be more performance efficient than JavaScript, we aim to look at this current status and present a preliminary study on the performance of these two, based on their runtime and energy efficiency. Preliminary results show that WebAssembly, while still in its infancy, is starting to already challenge JavaScript, with much more room to grow. Additionally, our benchmarking framework is also made available to allow further research and replication.

Abstract
Since the beginning of the Semantic Web and the coining of the term Linked Data in 2006, more than one thousand datasets with over sixteen thousand links have been published to the Linked Open Data Cloud. This rising interest is fuelled by the benefits that semantically annotated and machine-readable information can have in many systems. Alongside this growth we also observe a rise in humans creating and consuming Linked Data, and the opportunity to study and develop guidelines for tackling the new user interaction problems that arise with it. To gather information on the current solutions for modelling user interaction for these applications, we conducted a study surveying the interaction techniques provided in the state of the art of Linked Data tools and applications developed for users with no experience with Semantic Web technologies. The 18 tools reviewed are described and compared according to the interaction features provided, techniques used for visualising one instance and a set of instances, search solutions implemented, and the evaluation methods used to evaluate the proposed interaction solutions. From this review, we can conclude that researchers have started to deviate from more traditional visualisation techniques, like graph visualisations, when developing for lay users. This shows a current effort in developing Semantic Web tools to be used by lay users and motivates the documentation and formalisation of the solutions encountered in the studied tools. Copyright © 2021 for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).

Abstract

Abstract
We propose a mechanism to raise the abstraction level of source-code analysis and robustly support multiple languages. Built on top of the LARA framework, it allows sharing language specifications between LARA source-to-source compilers, and enables the mapping of a virtual AST over the nodes of ASTs provided by different, unrelated parsers. We use this approach to create a language specification for Object-Oriented (OO) languages and add support for three different LARA compilers. We evaluate it by implementing a library of 18 software metrics using this language specification and apply the metrics to source code in four programming languages (C, C++, Java, and JavaScript). We compare the results with other tools to evaluate the approach.

Abstract
In this paper we present a Deep Learning-based methodology to automatically classify 12 of the most representative grapevarieties existing in the Douro Demarked region, Portugal. The dataset used consisted of images of leaves at different stages of development, collected on their natural environment. The development of such methodologies becomes particularly important, in a scenario in which ampeleographers are disappearing, creating a gap in the task of inspection of grape varieties. Our approach was based on the transfer learning of the Xcepetion model, using Focal Loss, adaptive learning rate decay and SGD. The model obtained a F1 score of 0.93. To clearly understand the predictions of the model, and realize which regions of the image contributed the most to the classification, the LIME library was used. This way it was possible to identify the parts of the images that were considered for and against each prediction.

Abstract
This tool paper presents the High-Assurance ROS (HAROS) framework. HAROS is a framework for the analysis and quality improvement of robotics software developed using the popular Robot Operating System (ROS). It builds on a static analysis foundation to automatically extract models from the source code. Such models are later used to enable other sorts of analyses, such as Model Checking, Runtime Verification, and Property-based Testing. It has been applied to multiple real-world examples, helping developers find and correct various issues.