Papers Papers/2020



Proceedings of the 6th International Workshop on Software Knowledge

November 12-14, 2015, in Lisbon, Portugal

ISBN: 978-989-758-162-5

Conference Link: http://www.ic3k.org/SKY.aspx

Foreword: 1 SKY: A WORKSHOP AND MORE In terms of years of a human life, SKY is still in its infancy, since this is only the sixth edition: 1st SKY’2010 – in Herzlia, Israel 2nd SKY’2011 – in Paris, France 3rd SKY’2012 – in Barcelona, Spain 4th SKY’2013 – in Vilamoura, Portugal 5th SKY’2014 – in Rome, Italy 6th SKY’2015 – in Lisbon, Portugal. But as a Workshop, SKY is becoming more mature in its capabilities, while still young and dynamic. The SKY Workshop is held on Thursday, 12/November/2015. In this year SKY extends over two days, to contain the various initiatives, besides the regular Workshop: -SKY Challenge 2015 – with a well-developed set of 3 tracks, a definite step further from the previous year; -SKY Intensive School – with invited in-depth lectures on a few topics of interest to the Software Knowledge community, on Friday, 13/November/2015 -Three invited speakers! Two of them deliver plenary keynotes to the IC3K Conference, one keynote in each day. 2 THE SKY WORKSHOP Quantitatively, we continue with the same numbers of refereed and selected papers as in previous years: 4 full-papers, 4 short-papers and two invited talks in the Workshop day. The contributed papers are collected in two sessions: the first one in the morning and the second one in the afternoon. In terms of contents, two subjects are noticeable: 1- Natural Language, as a ubiquitous basis for knowledge in all domains; 2- Medical applications, as a challenging and useful domain in which language and knowledge are important. 3 FIRST SESSSION: POETRY & NATURAL LANGUAGE It is quite surprising to have two papers related to Poetry in a Workshop dedicated to Software Knowledge. But second thoughts about the papers reveal that Software is to a large extent based upon Natural Language. 3.1 Software, is it Poetry of Prose? The paper by Exman and Plebe starts with a dialogue between the authors in the style of Galileo Galilei. They set the stage for the problem of whether software is essentially syntax or mainly semantics. Upon this background software is compared with Poetry – sometimes with rigid syntax, but free and largely associative semantics. Two poems are given representations in terms of UML class diagrams and statechart. Then software is compared with Prose – which is more colloquial. In the opposite direction a class diagram of a software design pattern is shown to be essentially natural language. The final conclusion is that software has similarities to both Poetry and Prose, as it is more Natural Language than formal programming languages. 3.2 Invited Talk: The meaning of Language and using it Creatively (PoeTryMe) The talk by Hugo Gonçalo Oliveira, from Coimbra University in Portugal, is divided into two parts. In the first part, it presents ECO, an approach for learning wordnets automatically from textual resources of natural language. Wordnets are useful to resolve ambiguities, characteristic of natural languages, in contrast to programming languages. The second part of the talk overviews PoeTryMe, an intelligent system for poetry generation, with semantic and form constraints. In such a case, ambiguity is considered an advantage, rather than a difficulty to be solved. 3.3 Natural Language Processing Applied to Public Health Fraga and co-workers analyze how authors of papers write in a given domain. For instance this means, which groups of words are frequently used, most common grammatical categories, patterns found in the text. In order to actually test software tools developed for the natural language analysis purpose, a domain was chosen – e.g. genetic engineering – and used to demonstrate the techniques and the results obtained. 4 SECOND SESSION: FROM REQUIREMENTS, THROUGH EXCHANGES TO DOCUMENTS & ARCHIVES This session deals with infra-structure tools and techniques needed for the several stages of Software life-cycle. From the ubiquitous requirements from which one starts development, through exchanges among existing programs and data, to maintenance facilitated by documentation and long-term archiving. 4.1 A Model Supporting Requirements Elicitation for Content Rich Systems Goldin proposes in her paper a GQR (Goal-Question-Result) model to deal with requirements elicitation in content rich systems. The problem to be solved is the excess of content details; one cannot see the forest for the trees. The GQR model elevates the discussion to contextual structures by scoping it through a definite goal. The model is illustrated by applying to a medical domain and to a CMMI software environment. 4.2 Tracing Requirements through Model Variables Boschen and Rudat describe a particular notion of traceability, between model variables and text segments. This kind of traceability naturally connects formal to informal requirements. This is done semi-automatically, based upon an algorithm combining metrics, involving semantic distance. The approach is described in detail and illustrated by means of a prototype implementation. 4.3 Conceptual Object Exchanges by Non-Programmers Exman and Shimoni show that existing programs can be extended by non-programmers once an adequate and transparent infra-structure is provided. They believe this approach to software development will have increasing importance along time. Their basic notion is that “conceptual objects” – i.e. instances of classes with a clear conceptual meaning for the non-programmer – can be transferred from a program to another one without requiring programming knowledge. Their approach is given a proof-of-concept in the domain of platform games. 4.4 Flexible Evolvable Data Brokers Bailin and Bakker deal with a flexible and evolvable data broker, applied in a certain medical center. It was designed to accept a variety of data sources and destinations with little or no additional coding – a theme having similarities to usefulness for non-programmers, as discussed in the previous paper. The independence from database schema – using schema-less RDF – allows the broker to be used in other institutions and applications. 4.5 Wiki based Living Documentation with Domain Knowledge The paper by Yagel deals with software documentation that should be useful during system development and after product delivery for system maintenance. Its goal is to facilitate up-do-date documentation with less effort of the software engineer. The technique adds domain knowledge, to the idea of living documentation, by using a common Wiki. Among other advantages Wiki is both easy to manipulate and a proven tool for collaboration. The solution is obtained by an enhanced Wiki which marks domain concepts, to enable documentation tools to detect these concepts and generate more relevant documents. 4.6 Software Archiving for the Future Binun and co-workers tackle a thought provoking topic, viz. long-term software archiving. The crucial problem is the large amounts of digital data of various types, which according to our experience, are rapidly made obsolete by technology obsolescence. The proposed approach is based upon preserving programs as bit-blocks. The idea is to have a simple machine that is able to execute a single command to interpret these bit blocks. 4.7 Keynote: Formal Ontology, Patterns and Anti-Patterns The SKY Workshop day is concluded by a Keynote given by Prof. Giancarlo Guizzardi. His subject is Formal Ontology, Patterns and Anti-Patterns for Next-Generation Conceptual Modeling. His claim is that Conceptual Modeling has reached a state of too much complexity. It needs three complexity management tools: Ontological Design Patterns, Ontology Pattern Languages and Ontological Anti-Patterns. 5 SKY CHALLENGE 2015 The SKY'2015 Challenge has three different tracks to explore new solutions for well-known software engineering problems: first track – consists on analyzing quality in software product lines reusing the datasets available in the SPLOT repository to validate the accuracy of the proposed feature models; second track – consists on providing solutions for recovering traceability links between requirements in a software development process. The solution should formally represent requirements and automatically build a traceability matrix between requirements in different scopes: stakeholder, system, functional and non-functional requirements; third track – consists of analyzing existing software repositories to extract insights of the development processes. To do so, solutions should reuse the datasets and information available from different social coding platforms such as GitHub or Bitbucket to provide and discover the implicit knowledge generated in both the static social network and the stream of events. 6 SKY INTENSIVE SCHOOL The SKY Intensive School, whose first edition is in this year, deals with topics of importance for the Software Knowledge community, i.e. those involving Knowledge regarding all aspects of Software. The SKY Intensive School activities, each lasting about 45 minutes, are conducted only by invited researchers and practitioners, having significant experience in their respective fields. In this edition there are four lectures and one Keynote by Ralf Bogusch from Airbus Defence and Space, entitled “Ontology-based Systems Engineering - The Smart Way of Realizing Complex Systems”. Special emphasis has been given to the extensive industrial experience of all the lecturers. The intended audience of the SKY Intensive School encompasses active professionals (researchers and practitioners) and students in their advanced years of study or learning toward their higher post-graduation degrees. At the end of the SKY Intensive School, participants are expected to have a solid introduction to the specific fields dealt with during the SKY Intensive School enabling posterior deepening and actual application in the participant work area. 7 CONCLUSION The SKY’2015 Workshop can be characterized by various novel initiatives amplifying its range, while keeping its dynamic spirit. Acknowledgments Our thanks to the papers’ authors and to the invited and keynote speakers, whose contributions enable together a stimulating SKY’2015 Workshop. We are grateful to the dedication of the Program Committee members that did hard work resulting in the best and more representative papers among those submitted. We wish to thank the very helpful professional staff of IC3K/INSTICC. Finally, we are glad to acknowledge the continuous support of SKY Workshops by our institutions, The Jerusalem College of Engineering – JCE - Azrieli, in Jerusalem, Israel and UC3M – Carlos III of Madrid University, in Madrid, Spain. (More)


Vol. 1 - 978-989-758-162-5