Category Archives: Proceedings

Continuous Assessment in Software Engineering Project Course Using Publicly Available Data From GitHub

Title: Continuous Assessment in Software Engineering Project Course Using Publicly Available Data From GitHub

Authors: Henrik Gustavsson (University of Skövde), Marcus Brohede (University of Skövde)

Abstract: This paper describes an approach for assessment in a large software engineering project course. We propose an approach for continuously collecting information from a source code repository and collaboration tool, and using this information for assessing student contributions and also for assessing the course as a whole from the teacher’s standpoint. We present how we display metrics for how the students perform in relation to some of the requirements of the course. We argue that continuous summative assessment feedback to the students on how they are performing in the project is a suitable strategy for ensuring active participation from the students for the duration of the project course.

Download: This contribution is part of the OpenSym 2019 proceedings and is available as a PDF file.

FLOSS FAQ Chatbot Project Reuse – How to Allow Nonexperts to Develop a Chatbot

Title: FLOSS FAQ Chatbot Project Reuse – How to Allow Nonexperts to Develop a Chatbot

Authors: Arthur R. T. de Lacerda (University of Brasilia), Carla Silva Rocha Aguiar (University of Brasilia)

Abstract: FAQ chatbots possess the capability to provide answers to frequently asked questions of a particular service, platform, or system. Currently, FAQ chatbot is the most popular domain of use of dialog assistants. However, developing a chatbot project requires a full-stack team formed by numerous specialists, such as dialog designer, data scientist, software engineer, DevOps, business strategist and experts from the domain, which can be both time and resources consuming. Language processing can be particularly challenging in languages other than English due to the scarcity of training datasets.

Most of the requirements of FAQ chatbots are similar, domain-specific, and projects could profit from Open Source Software (OSS) reuse. In this paper, we examine how OSS FAQ chatbot projects can benefit from reuse at the project level (black-box reuse). We present an experience report of a FLOSS FAQ chatbot project developed in Portuguese to an e-government service in Brazil. It comprises of the chatbot distribution service, as well as for analytics tool integrated and deployed on-premises. We identified assets that could be reused as a black-box and the assets that should be customized for a particular application. We categorized these assets in architecture, corpus, dialog flows, machine learning models, and documentation. This paper discusses how automation, pre-configuration, and templates can aid newcomers to develop chatbots in Portuguese without the need for specialized skills required from tools in chatbot architecture. Our main contribution is to highlight the issues non-English FAQ chatbots projects will likely face and the assets that can be reused. It allows non-chatbot experts to develop a quality-assured OSS FAQ chatbot in a shorter project cycle.

Download: This contribution is part of the OpenSym 2019 proceedings and is available as a PDF file.

Predicting Open Source Programming Language Repository File Survivability From Forking Data

Title: Predicting Open Source Programming Language Repository File Survivability From Forking Data

Authors: Bee Bee Chua (University of Technology Sydney), Ying Zhang (University of Technology Sydney)

Abstract: Very few studies have looked at repositories’ programming language survivability in response to forking conditions. A high number of repository programming languages does not alone ensure good forking performance. To address this issue and assist project owners in adopting the right programming language, it is necessary to predict programming language survivability from forking in repositories. This paper therefore addresses two related questions: are there statistically meaningful patterns within repository data and, if so, can these patterns be used to predict programming language survival? To answer these questions we analysed 47,000 forking instances in 1000 GitHub projects. We used Euclidean distance applied in the K-Nearest Neighbour algorithm to predict the distance between repository file longevity and forking conditions. We found three pattern types (‘once-only’, intermittent or steady) and propose reasons for short-lived programming languages.

Download: This contribution is part of the OpenSym 2019 proceedings and is available as a PDF file.

Getting Started With Open Source Governance and Compliance in Companies

Title: Getting Started With Open Source Governance and Compliance in Companies

Authors: Nikolay Harutyunyan (Friedrich-Alexander University Erlangen-Nürnberg), Dirk Riehle (Friedrich-Alexander University Erlangen-Nürnberg)

Abstract: Commercial use of open source software is on the rise as more companies realize the benefits of using FLOSS components in their products. At the same time, the ungoverned use of such components can result in legal, financial, intellectual property, and other risks. To mitigate these risks, companies must govern their use of open source through appropriate processes. This paper presents an initial theory of industry best practices on getting started with open source governance and compliance. Through a qualitative survey, we conducted and analyzed 15 expert interviews in companies with advanced capabilities in open source governance. We also studied practitioner reports on existing practices for introducing FLOSS governance processes. We cast our resulting initial theory in the actionable format of best practice patterns that, when combined, form a practical handbook of getting started with FLOSS governance in companies.

Download: This contribution is part of the OpenSym 2019 proceedings and is available as a PDF file.

Qwant : Innovation, Sovereignty & Privacy: Building the Web we want for Europe

Title: Qwant : Innovation, Sovereignty & Privacy: Building the Web we want for Europe

Author: Tritan Nitot

Facilitator: Olivier Berger, Telecom SudParis

Download: This contribution is part of the OpenSym 2018 proceedings and is available as a PDF file.

Software Heritage – Browsing the Free Software Commons

Title: Software Heritage – Browsing the Free Software Commons

Author: Stefano Zacchiroli

Facilitator: Olivier Berger, Telecom SudParis

Download: This contribution is part of the OpenSym 2018 proceedings and is available as a PDF file.

How to estimate the value of open intangible assets?

Title: How to estimate the value of open intangible assets?

Author: Robert Viseur, University of Mons

Abstract: Open innovation practices are widespread in the industry. The software sector, marked by the rise of open source, is a striking example. This paper presents the preliminary results of an exploratory research on estimating the value of open intangible assets. Our approach favors simplicity and relies on a partial automation of the evaluation. An evaluation structure, distinguishing the different forms of value involved in open intangible assets, particularly applicable to software, is provided as an illustration of the proposed approach.

Download: This contribution is part of the OpenSym 2018 proceedings and is available as a PDF file.

How much are digital platforms based on open collaboration? An analysis of technological and knowledge practices and their implications for the platform governance of a sample of 100 cases of collaborative platforms in Barcelona

Title: How much are digital platforms based on open collaboration? An analysis of technological and knowledge practices and their implications for the platform governance of a sample of 100 cases of collaborative platforms in Barcelona

Authors: Mayo Fuster and Ricard Espelt

Abstract: From the early cases of FLOSS and Wikipedia, the digital collaborative model of production and consumption has rapidly expanded to other spheres. This article explores to what extent this expansion has maintained the open character of the initial model, specifically the extent to which platform projects follow an open collaborative approach in their technological and knowledge policies and practices, and if this is also reflected in an open approach to governance. The empirical analysis is based on a sample of 100 cases in Barcelona. On the basis of this analysis, we conclude that open modalities of collaborative digital platforms are not prevalent. Around a third of the sample present open modalities of the dimensions analyzed. Different areas (technological, knowledge, or governance) showed different levels of diffusion of open practices. The cases which tended to be open in one dimension also tended to be open in the other dimensions. That is, the analysis points to a correlation between technological, data, and knowledge policies and open and democratic collaborative economy models. These results suggest the importance of open technology and knowledge in adopting an open and democratic collaborative model.

Download: This contribution is part of the OpenSym 2018 proceedings and is available as a PDF file.

Wikipedian: a social identity between work and contribution

Title: Wikipedian: a social identity between work and contribution

Author: Léo Joubert, Aix-Marseille University

Abstract: Contributors to the Wikipedia “free encyclopedia” identify themselves and are identified as “Wikipedians”. A Wikipedian does not leave his job when he becomes a Wikipedian. Nor does he become a Wikipedian in his workplace. The worker’s identity and the Wikipedian identity coexist in the social identity of an individual. On which patterns does this coexistence between worker’s identity and Wikipedian identity operate? Beyond the differences specific to the social identity of each contributor, we will try to show that singulars transactions all take place according to a finite number of patterns that it is possible to count. At this stage of our analysis, we are able to distinguish five identity patterns: employment, learning center, alternative development, continuity in upset, parallel arena. Our model aims to better understanding of why a contributor stay in Wikipedia and identifies himself as a contributor.

Download: This contribution is part of the OpenSym 2018 proceedings and is available as a PDF file.

A clustering approach to infer Wikipedia contributors’ profile

Title: A clustering approach to infer Wikipedia contributors’ profile

Authors: Shubham Krishna, Romain Billot and Nicolas Jullien

Abstract: Recent studies have improved our knowledge about the different types or pro€files of online contributors, from casual to very involved ones, through focused people. But they use very complex methodologies, making their replication by the practitioners limited. We show on both Romanian and Danish wikis that using only the edit and their distribution over time to feed clustering techniques, allows to build these pro€les with good accuracy and stability. Œis suggests that light monitoring of newcomers may be sucient to adapt the interaction with them and to increase the retention rate.

Download: This contribution is part of the OpenSym 2018 proceedings and is available as a PDF file.