Ramon R. Palacio

What Information Would You Like to Know about Your Co-worker? A Case Study

The lack of face-to-face interaction in Distributed Software Development (DSD) projects is a great obstacle for informal communication, which is the way in which we normally discover more about the co-workers along with building trust. Since we are attempting to reinforce the team spirit and improve trust between co-workers, we have analyzed what information software engineers consider to be important to know about their virtual colleagues. In this paper we describe the preliminary results obtained from this study and outline the features of a tool, called Trusting Social Network, with which to support DSD and help stakeholders to know each other better.

Collaborative Working Spheres as support for starting collaboration in distributed software development

One important tendency in software development has been the globalization of its industry. Software developers are frequently required to work in groups which are distributed throughout multiple geographic locations. People and processes (e.g. software design) are consequently distributed. Providing the support that will allow distributed software developers to collaborate at a convenient moment, is therefore extremely complicated both for the person making contact and the person being contacted. We therefore propose a characterization and a model of Collaborative Working Spheres (CWS) through which to understand the management of activities in distributed software development (DSD). Our work first aims to characterize DSD activities mainly through a survey of literature. We then describe a projected scenario to illustrate how DSD activities can be supported by technology and finally we propose a characterization of CWS through which to represent how DSD workers can collaborate at a moment which is suitable for all involved participants.

Trusty: a tool to improve communication and collaboration in DSD

Distributed Software Development (DSD) projects frequently confront the problem of a lack of face-to-face interaction, which is a great obstacle in informal communication. Since informal communication is the means by which people normally discover facts about their co-workers, thus leading to their mutual trust, we have designed a tool called Trusty with which to support DSD. In this paper we describe the main characteristics of Trusty, which provides mechanisms to support communication, coordination, knowledge management and other capabilities such as the statistical analysis of those networks which are valuable in virtual environments.

A very low complexity near ML detector based on QRD-M algorithm for STBC-VBLAST architecture

STBC-VBLAST MIMO space-time codes combine spatial multiplexing with diversity gain to achieve both high spectral efficiency and link reliability. In this work, we present a new near maximum-likelihood detector based on QRD-M algorithm for DSTTD architecture. We analyze its performance in terms of Bit Error Rate and complexity. We use an uncorrelated MIMO channel model to prove the code performance and decoder complexity. We show that a near maximum likelihood detector based on QRD-M algorithm shows a better performance in terms of complexity without degradation in terms of Bit Error Rate with respect to [11] for DSTTD space-time code. Therefore, we conclude that this scheme is an attractive candidate for hardware implementation.

Understanding Architectural Knowledge Sharing in AGSD Teams: An Empirical Study

Nowadays, the use of agile methodologies (AM) in Global Software Development (GSD) -- known as AGSD -- is increasingly common. However, AM and GSD are not completely compatible. On the one hand, in AM people interactions (face-to-face) are preferred over document-based communications to share knowledge. On the other hand, in GSD knowledge sharing is conducted through documents to minimize the effect of the inherent four distances (physical, temporal, language and cultural). This means that tacit knowledge is preferred in AM and explicit knowledge is preferred in GSD. These differences between AM and GSD affect many aspects of software development, for instance: Architectural Knowledge Management. According to the literature, in AGSD it is preferred to convey Architectural Knowledge (AK) by frequent interactions across sites through unstructured and textual electronic media (UTEM) (chats, emails, forums, etc.), that is, AK is articulated in these media. UTEM leave a textual record of the transmitted information, thus leaving an unstructured log of the shared AK of the project. In this paper we present an empirical study to understand AK articulation in UTEM in AGSD teams. Our results consist of an ontology that represents the involved aspects in AK articulation in UTEM in AGSD teams. Additionally, we identified eleven categories of interactions across sites through UTEM, where requirements and coding themes are prominent. Finally, we found that AK in UTEM is perceived as important, regardless the interaction frequency. These results lead us to think that a tool to structure and exploit AK in UTEM is needed in AGSD, in order to bridge the gap between AM and GSD.

Promoting Elderly-Children Interaction in Digital Games: A Preliminary Set of Design Guidelines

In this paper, we propose a set of guidelines to facilitate the design of digital games to support elderly-children interaction. We conducted a literature review to identify preliminary elements in the elderly-children interaction process and assessed elder-child interaction during their using two digital games to complement the initial findings. Based on these results, we proposed a set of guidelines to aid in the design of digital games. To validate the guidelines, we conducted an evaluation on the design of digital games with a group of 12 postgraduate students from two local Computer Science programs. Results are promising, indicating a high perception of usefulness, ease of use and intention of use of the proposed guidelines by the group of developers.

Providing Support for Starting Collaboration in Distributed Software Development: A Multi-agent Approach

Developing a system to provide support for distributed software developers (DSD) to get into collaboration is very complicated. On the first hand, it is necessary to know the characteristics of their daily work activities. On the other hand, technical aspects must be considered, such as obtaining information on the context and on the data flow of their activities. This requires information from the individual and group work environment. Therefore, we propose a model to help in the development of this type of system. Firstly, our work aims at modeling the information flow of DSD workers based on a literature survey and on our own experience. Secondly, we describe a projected implementation scenario based on the multi-agent proposed system.

Supporting component presence notifications in software development

One of the current approaches to software engineering is Component-based Software Engineering (CBSE). As its name implies, CBSE has to do with the development of open and distributed systems by assembling a set of components. However in practice it is difficult to follow the characteristics demanded by this working philosophy because when developers want to update or modify a component, they do not have the required and sufficient information to determine the degree of impact that a change will have. In order to understand the management of components in CBSE ten experts in software development were interviewed. With the information obtained, the requirements for a system that we call Component Presence Notifier were defined. This system supports the extraction of information from different repositories of the organization and presents it as elements of a graphical user interface (GUI). This information is provided to the developer while working with any of the registered components in order to increase his/her awareness of the impact of the proposed modification. As future work, this prototype will be evaluated for usability by a group of distributed software developers.

CWS: An Awareness Tool to Support Starting Collaboration in Global Software Development~!2009-07-01~!2010-05-03~!2010-05-17~!

Software development organizations are facing a paradigm shift towards Distributed Software Development. This shift introduces situations from which organizations may benefit (e.g. highly skilled human resources, development groups closer to client location, etc.); but also introduces challenges to which organizations have to adapt (e.g. coordina- tion difficulties, inadequate knowledge management and communication, and lack of inter-virtual-team trust relation- ships). In this work, we particularly study the lack of timely adequate opportunities for informal interaction, which has been identified as an underpinning foundation to overcome coordination, communication and trust limitations. To achieve this, we introduce and define the concept of Collaborative Working Spheres (CWS), through which developers can obtain information related to the personal activities of their distributed colleagues. CWS allow identifying opportunities for col- laboration in suitable moments both for the one making contact and the one being contacted. We notice that other exam- ples of technologies, including the telephone and instant messaging are used by developers for starting collaboration; however, they do not provide enough information from the personal activity of the person being contacted. We argue that with CWS, software developers will be able to become aware about the status and progress their partners have achieved in some activity, and use this information to inform their starting collaboration. We illustrate this concept with the design of a CWS-based messenger tool that supports Collaborative Working Spheres for Distributed Software Developers. The re- sults of an initial evaluation provide encouraging evidence on the perceived usefulness and ease of use of the proposed CWS-based messenger tool.

Low complexity maximum-likelihood detector for DSTTD architecture based on the QRD-M algorithm

This paper presents a new decoder algorithm for the double space–time transmit diversity (DSTTD) system. The decoder is based on the QRD-M algorithm, which performs a breadth-first search of possible solutions tree. The search is simplified by skipping unlikely candidates, and it is stopped when no promising candidates are left. Furthermore, the search is divided into three concurrent iterations, making possible a fast, parallel implementation either in hardware or software. After presenting an analysis of the capacity and diversity of DSTTD, we present performance results showing that the proposed decoder is capable of achieving near maximum likelihood performance. We also show that the proposed algorithm exhibits lower computational complexity than other existing maximum likelihood detectors.