My PhD project falls within the broad field of Science and Technology Studies (STS) and focuses on the topic of "users and technology production". In STS, users are considered a key actors group in the study of technological production, maintenance and evolution. Therfore they should not be neglected in studies that are interested in these phenomena.
This research is interested in the role of the users in the development of Free and Open Source Software (FOSS) projects. FOSS is a software development paradigm which embodies an ethical view of what software should be for society. Thanks to particular software licences and to the fact that the source code is publicly accessible FOSS allows for an open, collaborative and distributed way of developing and maintaining software.
Today, FOSS is a worldwide renowned phenomenon and widely investigated. Indeed, FOSS is a lively and multidisciplinary research topic. Software engineering, computer science, innovation studies, knowledge management, economics and also cultural and sociological studies are all interested in investigating FOSS and some of its aspects. However, the departing assumption of this research is that current research neglects (a part from very few exceptions) the relevance of the users. In a simplified view of the FOSS communities social structure it is possible to identify: project leaders, core developers, co-developers, active users and passive users. Most of the literature existing today is interested in everything that relates to the first three roles and all the related activities they are involved with, but the last two groups and their relation to the software development is largely neglected.
With the case study of Battle for Wesnoth and the use of (cyber-)ethnographic research method, this project aims at highlighting the role of users in FOSS projects. The following general questions inform this research: What kind of practices are the users involved in? What relations do exist between the users local practices and the software development ones? How do users experience their participation in the community? How does the software use mediate participation in the community?
Update: if you are interested, you can download here the full proposal. Unfortunately, at the moment is only available in Italian.
Update2 (20/03/2011): After the work on my reasoned bibliography (can be found here) the academic frame of this reasearch moved from STS towards the Participatory Design (PD) field. As soon as I have more time I'll update the content of this page
The constant growth of Free and Open Source Software (FOSS) attracted researchers from different fields over the last decades. A large body of knowledge accumulated about FOSS development: some scholars focused on providing explainations of motivations for individual participation in FOSS projects; others investigated its ethical foundations and the legal ones. Most researchers directed their attention on the self-organizing characteristics of FOSS communities, their governance mechanisms and their social structures. Other streams of research highlighted the innovative nature of FOSS development in relation to traditional software industry practices, the economic principles that make it a sustainable 'knowledge' production model, and the hacker culture that sustains many of the FOSS practices.
The departing point of this research is the little attention that the existing literature has paid to the users. Indeed, it is here claimed, this literature provide a reductionist account of FOSS:
Of course, it is true that this phenomenon is primarily about software and software development. Therefore, this aspect has a central role in the broad area of FOSS studies. However, while this focus has provided a lot of valuable knowledge it has also blinded us from seeing other aspects.
I use the so called onion-model for a FOSS community social structure, to simplify the point I'm trying to make here. Most of the literature existing today deals with the central layers (project maintainers, core and co-devs) or on the shifts from external (peripheral) layers to the central ones: from being a user to being a developer. What actually happens in the peripheral layers is rarely at the centre of the investigation: we know that active-users contribute to the project development by submitting bug reports, providing translations, writing/revising/translating documentation, helping and educating new users, but how does this happen?
If coordination and collaboration are key aspects of FOSS development, then it is important, for the 'ecosystem' of one project, to understand how within different activities and across them these aspects are played out.
In order to resume, the main goal of this research is: to understand the role of the users in the development of a FOSS project, in relation to those activities usually called peripheral.
The objectives set for this research favour an in-depth qualitative inquiry over a quantitative one. In particular, an ethnographic approach to the phenomenon of inquiry seems the most appropriate. Ethnography is a widely adopted method in sociological research even if it originated in the field of anthropology. It is used to study unkown cultures, specific organizational settings and community life in general. The method implies a lasting stay among the people who constitute the object of the research and therefore, within the culture to be studied. The main techniques for data collection are participant observation and interviewing. Inductive analysis (inspired by the Grounded Theory principles) will be used on the collected data.
In this research the object of inquriy does not have a physical location in the traditional sense. The Battle for Wesnoth's (BfW) community primarily exists on-line. Computers and Internet mediate all the interactions that community members engage in. For this reason, the ethnography cannot take place in the traditional sense: there is no place where the researcher can physically go and stay. However, successful studies that investigates human interactions that happens on-line do exist and are increasing. The revised form of ethnography used in these situations is known with many names: virtual ethnography, ethnography of on-line groups, cyber-ethnography. All of them are very similar and slightly differ in their epistomological stances. This research uses a cyber-ethnographic method.
The characterisation of the research field is the first point that distinguish a cyber-ethnography from a traditional one. The field is discovered during the ethnography and not decided a priori. However, some preliminary entry points are needed. For the case study of BfW the two following ones will be used as departing points:
The second point to be addressed is the access to the field, which immediately raises the issue of the ethnographer's presence. In BfW like in all FOSS communities, nearly all accesses are public. Archival records are publicly accessible and registration to the community media is open. On one hand, these aspects facilitate the 'access to the community life'. On the other hand, I also want to participate, and not only to observe (lurk). Therefore, I need a substantial identity which I can use to interact with community members and, at the same time, can legitimate a particular kind of participation (i.e. I might need to ask questions that could be considered off-topic, or i might need to go beyond the 'RTFM answer'). The purpose of this website is to represent that identity.
I will point members interested in knowing what my research is about to these pages. Moreover, from time to time, when I have enough material to make consistent reflections I will create new content, and leave it open to comments. In this page, I'll try to answer all doubts that the informants might have about the use of their data and about how their privacy will be guaranteed.
The length of my stay in the field is planned to be about nine months, followed by about three months of analysis. During this time-frame I plan to locate and familiarize with the key community practices involving users participation. To discover the links between these practices and the ones related to software development (in the strict sense). I will do so by reading communications amongst members (either on IRC, mailing-lists or forum) and in the cases where this will be possible, also to participate in these discussions.
The case chosen for this research is the turn-based strategy game Battle for Wesnoth (BfW). The choice to use this FOSS project as main case rest on several reasons related to the objectives of the research. The interest on the role of users (and usage) in FOSS development called for a case study on an active project which allows for heterogenous ways of contributing and which also integrates on user made content. It's true that most FOSS projects (in a way or another) present these charateristics, but BfW stood out from my selection process as the best candidate.
I selected about 40 projects from the two biggest FOSS collaborative platforms (Sourceforge and Freshmeat) filtering the selection on the project's activity level. In roughly one month, I discarded most of these projects and remained with less than ten projects. Amongst these (4 videogames, 2 audio/video editors and one small managerial software), BfW highlighted the most intereesting characteristics for the current research. The preliminary selection has been done by looking at the projects' Contribute pages, main websites, forums and roadmaps. This are the key characteristiics that favoured BfW over the other projects:
The Battle for Wesnoth, or simply Wesnoth, is a turn-based strategy game with a fantasy setting, designed by David White and first released in June 2003. In Wesnoth, the player attempts to build a powerful army by controlling villages and defeating enemies for experience. White based Wesnoth loosely on the Sega Genesis games Master of Monsters and Warsong.He wanted to create a freely-available, open source strategy game with very simple rules, but one that had strong artificial intelligence and that was challenging and fun. (ref. Wikipedia)
The result of this research will be a PhD thesis, which probably will not be published for large distribution (even though I'd like that), and will be read nearly exclusively by a small network of academics (namely my supervisors, some colleagues and few other researchers). Most likely, however, this research will provide material for few publications (in academic international journals). These publications will include partial analysis of the data and some reflections on the methods I used to obtain and elaborate them.
There is a third way in which the data might be spread: by being published on this website. I do not consider this case as publication of research resutls, rather as a part of the undergoing research. Blog posts and other content published here will be in the form of reflections or partial analysis of specific situations (the data) relevant for the research. This form of content will be open for discussion, comments and debate.
Here, I try explain how I intend to respect and protect the privacy of the people who inform my research in the case of the first two types of distribution (PhD thesis, journal articles).
Here, I use the word "data" to refer to all the material that I will gather during the following 10 months and which I will use to build my analysis.
Mainly, from the BfW community: online forum, IRC, mailing lists, wiki, private discussions/interviews.
All the data that will be collected are those ones produced for 'public consumption' (mailing list archives, forum discussions, public IRC logs) or those ones voluntarily given by the informants upon request (interviews).
Mainly, these data will be textual data. For example, forum or mailing-lists threads, chunks of IRC logs, official documents (rules, guides, tutorials).
E-mail addresses and/or nicknames will be collected only to be able to identify participants in the discussions and for targeting potential interviewees. In no way the e-mail addresses will be disclosed.
No IP addresses will be collected in this research.
Manually. All the discussions (which seem potentially relevant) will be read and manually collected.
Where possible and worthwile, the Firefox plugin ScrapBook Plus will be used to store locally a copy of the page/discussion. For other media (IRC, mailing lists) I will use the saved logs and archives directly within the used application.
In this research, no automated data mining tool or crawler will be used.
I will be the sole researcher handling these data in raw form, no third party is involved.
In case that data are needed for supporting the presentations (in spoken or written form) of the research results, all the identifiers (nicknames, e-mail addresses or other ID forms) will be changed, to preserve the identity (and privacy) of people involved.
Most of the data, however, will be used for the anlysis only, and will not be published.
For any inquiry about the use of data you can add your question as a comment to this entry, or send me an e-mail. I'll try to answer as soon as possible.
The document, attached at the bootom of the page, is an attempts to frame the academic debates that are more connceted to my research work.
The core of this bibliography relates to Participatory Design and to the current debates that can be found within this field of studies. Namely, distributed participatory design and continuous design (aka: design-in-use, meta-design). The other major section of the reasoned bibliography deals with the interest on user participation in mediation processes between design/use, within FOSS studies. The final (and smaller) section enters into debate of game studies and it focuses on the area of game (story) writing.
The two major parts are rather comprehensive, while the last one might need some updates in the future.
The chapter attached below is intended as the first full blown draft of my dissertation's theoretical chapter. It is an ordered and focused reflection that will (theoretically) frame the key arguments and findings of my doctoral research. It is derived from the work previously done on the Reasoned bibliography and from the first period of field work. The chapter will certainly be modified before the final completion of the dissertation, but it served as a key milestone, a sort of 'deliverable', required for the completion of the second academic year in my doctoral programme. So far, it received praises from the scientific programme board and positively evaluated.
The chapter, provisionally titled Users Participation in Free Software continuous and distributed design addresses the three main concepts/phenomena that frame the title itself: users participation; free software (paradigm/development); and continuous and distributed design. Broadly speaking, the arguments developed in the chapter cross the three areas of Participatory Design, Free Software studies, and Information/Software Systems development. Each main section departs from the traditional definition of the phenomenon in the pertaining area and it reaches a refined and up-to-date conceptualization of the same phenomenon.