In this editorial I would like to make a general and effective call for more diversity in information systems research genres. This is a general call that goes beyond a particular journal and attempts to provide some practical guidelines. One can be sympathetic to the vision of opening up to a new wider set of research and presentation genres such as the one developed in my former editorials (Rowe, 2010, 2011), but without some guidelines this call for increasing diversity in IS research genres might hang fire. In other words, to encourage potential authors to be more confident in these endeavors, we need to indicate which lampposts might be worth approaching when building their research. Journals play an important institutional role in signaling and promoting certain categories of research. Hence, the two objectives of this editorial are as follows:

  • To propose a new categorization for IS research genres: 1. Literature Reviews, 2. Theory Development and Research Essays, 3. Empirical Research, 4. Ethnographies and Narratives, 5. Issues and Opinion. Whether we should distinguish also an ‘alternative genre’ and what would be included in this category is open to debate.

  • To offer guidelines for three broad publication genres that are still too rare in our community. Those are literature reviews, theory development papers and research essays, and ethnographies and narratives. Before turning to the guidelines, I will briefly explain the motivation for promoting each genre, its scope or what it may cover.

The prevailing genres in IS research

Prior to proposing a new categorization of IS research genres, we should recognize the prevailing genres. Up to now EJIS has two categories ‘Research articles’ and ‘Opinion papers’, while others have several such as ‘Research notes’, ‘Research commentaries’, ‘Theory and review’, ‘Research essays’ and ‘Issues and opinions’ papers. Such classifications are broad and simple but also not entirely satisfactory in two ways. The EJIS taxonomy remains silent on theory development papers and literature reviews, while some taxonomies seem, by their emphasis on several ‘Research’ categories, to send the signal that theory development papers and literature reviews are not research. Taxonomies of research genres may need to be a little more detailed to stimulate diversity than currently is the case at EJIS and as clear as possible (Nickerson et al, 2012). I would argue that, in the European tradition, we are already well prepared to welcome contributions in three publication genres that I would like to promote because they are very valuable but too rare in the IS community: 1. Literature Reviews, 2. Theory Development and Research Essays, and 3. Ethnographies and Narratives. In addition this would also meet Bob O’Keefe's wish (2003) for a more focused and balanced set of empirical and theoretical papers.

These three genres can be distinguished first from prevailing empirical research. Empirical research is dominant in IS (Liu & Myers, 2011; van Osch & Avital, 2010). Literature Reviews, Theory Development papers and Research Essays are by definition distinct from empirical research. The third publication genre, Ethnography and Narratives, is not a prevailing empirical genre in IS research. In fact, in their survey of research articles in the Association for Information System (AIS) basket of six top journals over a 10-year period, from 1998 to 2007, Liu & Myers (2001) found that the four major research methods were by decreasing order: surveys (more than 50% in 2006 and 2007), case studies (about 35% in 2006 and 2007), experiments (between 10 and 20% depending on the years) and action research (reaching 12% in 2007). In IT impact research published in MISQ, ISR, I&O and EJIS prevailing empirical papers are by decreasing order of frequency: experiments, case studies, questionnaire surveys, secondary data and field study (Paré et al, 2008). These genres of research methods account for 90% of the published papers (id.). Between 1997 and 2007, irrespective of the topic, 22% of the articles published in EJIS were based on case studies, 18% on interviews and 17% on surveys (Dwivedi & Kuljis, 2008). Hence, these different genres (experiments, case studies, questionnaire surveys, interviews, secondary data and field study) constitute what we can refer to as the (prevailing) empirical research paper. Design-oriented research or action research could fit into the empirical genre or the new ‘theory development and research essays’ genre depending on the approach and contribution. These have become prevailing works even if historically less frequent in IS. This fit applies independently of the philosophy of science perspective adopted (Van Gigch & Le Moigne, 1989; Hevner et al, 2004; Winter, 2008), as long as there are contributions of high generality and relevance (Baskerville et al, 2011). In contrast, it is interesting to note that ethnography is not mentioned as such by Dwivedi and Kuljis, and appeared with only 2.6% of the papers in Paré et al’s study. One could agree that ethnographies and certain narratives have an empirical basis, but to promote them we need to make it a special category.

Second, these genres can also be distinguished from the common opinion papers. The prevailing opinion papers are naturally positioned as solutions, more or less controversial, to an institutional or disciplinary challenge. How to bridge relevance and rigor? How wide should the scope of the field be? And, correspondingly, which communities should we interact with to participate in the building of our own collective identity? Such are typical questions accommodating several opinions as long as, of course, these opinions come with solid arguments. When the arguments are strong enough, opinion papers are good and necessary to nurture the sense-making and the legitimacy of our activities, and as such, they are most welcome. Some good recent examples appeared in the debate on design science, which was introduced by a collective editorial (Junglas et al, 2011) and furthered by two opinion papers (Baskerville et al, 2011; Österle et al, 2011). EJIS has always published such opinion papers but also mixed or confused this genre as defined here with more conceptual or theoretical papers that I will refer to later in this editorial. It is also worth mentioning that a special kind of publication genre that we have singled out, not as a paper (as it is considered a research article), but as part of a series, is the ‘contrarian’ paper. The series led by Joe Nandhakumar and inspired by Richard Baskerville offers scientific debates and potential controversy, laid on very scientific grounds. One of the best examples we have is the paper by Hunter (2010) who replicates Barley's study with a different method, finds different results and gets commented by Marios Koufaris (2010). In this sense contrarian papers can participate in the building of a controversy, but they are not opinion papers in the usual sense. They are empirical papers, although generally shorter, more similar to a research note.

Three genres too rarely seen in IS research

The three genres that I would like to encourage are as follows:

Literature reviews

The IS community does publish literature reviews, but with caution. This genre is probably the most well-known of the three above, and there is no need to describe it at length. It has long been identified as one of the main types of papers expected to grow scientific knowledge in more appropriate directions (Webster & Watson, 2002). Literature reviews can make substantial contributions when they reveal such things as research gaps, operating theories, frameworks and unrecognized assumptions. The identification of new research avenues is a great service to the community and leads to high citation impact. However, these contributions are more often trivial than significant. EJIS sometimes publishes papers that offer significant contributions to knowledge based on rigorous and deep literature reviews. But literature reviews coming typically from Ph.D. dissertations might not correspond to the EJIS editorial aspirations. In sum, we are all in favor of very good literature reviews. But these have to be up to the quality standards that are expected of an EJIS journal article.

My editorial experience with literature reviews at Systèmes d’Information et Management and EJIS leads me to discourage single author submissions. The likelihood to meet the publication standards expectations greatly increases if at least two colleagues with experience on the problem (in the domain) are collaborating. A lot has been said and proposed on the methods required to collect, organize, document and summarize the set of empirical and non-empirical papers that are relevant to the problem such as meta-analysis, meta-synthesis and the like (e.g., Truex et al, 2011; Wu & Du, 2012). This paper collection effort has to be significant and substantial, that is, done with some intelligence and applied to a sufficient set of papers. But the most crucial point is that, in order to analyze the phenomenon and get interesting results, researchers need to have a good conceptual framework, not necessarily a theory, but a set of coherent macro-concepts. This conceptual framework will help them to analyze theoretically the dimensions of each concept and thus code the data. Sometimes a theory has long been proposed and many researchers agree that this theory is relevant to analyze the phenomenon. Let's take the case of transaction theory being used to account for outsourcing. An interesting result would precisely be to show that this theory is not sufficient to account for the phenomenon (Alaghehband et al, 2011). But when there is no such theory or when the relevant theories are rarely used, as in the case of organizational transformation for instance (Besson & Rowe, 2012), researchers have to come up with a new conceptual framework. In that case interesting research gaps are more likely to be identified. But the ultimate goal of a literature review is to transform the identification of these research gaps into research avenues. This is not a self-evident result and this last part too often never materializes or falls short. Not that it is neglected but to say something that is not a direct outcome of the results of the data analysis requires careful interpretation and a sense of the relative value of the research avenues themselves because these avenues might not be closely related to the gaps. In following the example of Schryen (forthcoming), researchers divide their literature review in three parts, ‘Synthesis of knowledge (what do we know?)’, ‘Identification of lack of knowledge (what do we still need to know?)’, and ‘Proposition of paths for closing the knowledge gap (how can we get there?)’; authors should not forget that the first part requires a good sense of the problem, a capability to reframe it (Weber, 2003a) – at least on the conceptual part (cf. 2.) – and some synthesis capabilities; the second part requires the more classic systematic and analytical skills researchers generally have, and the last part requires more advanced speculative abilities and intuition.

In sum, literature reviews are useful when they provide some synthesis and a vision of part of the future in a knowledge universe where most objects are more and more atomistic. They are not so useful when they only rely on counts based on already well-known categories and with little discussion of the results. They are most welcome but researchers should first identify relevant phenomena that are amenable to such effort and build teams with the corresponding qualities.

Theory development and research essays

I will not be original in calling for more pure theory development (i.e., without data) papers. This publication genre is clearly distinct from the theory-testing papers, in particular the case studies that EJIS publishes frequently. Too often reviewers reject these pure theory papers on the grounds that EJIS does not publish this genre. But evidence shows this is not entirely true (e.g., Johnston & Gregor, 2000). I am happy that EJIS has earned a good reputation for empirical papers, but I think that, as Europeans, we could do better and that we should also accept pure theory papers as long as they are of the highest quality. Many that we have received have not met these expectations. However, my feeling is that, with all the institutional pressures to publish, not sending a positive and welcoming signal for theory papers that fit our policy would lead to even more incremental papers, with a quantitative standard methodology and a theory-testing inclination. In the continental tradition, we have known for centuries, if not since the Greeks, that we do not only learn by experience (Locke, 1995) but that categories of knowledge are also constructed by reason (Leibniz, 1765; Kant, 2003).

Our discipline is constantly being populated by apparently new technological artifacts that call for new observations (Wang, 2010). Human perception, visual or audio, is often defeated. All scientists know that. When trying to explain or understand social behavior we often have to access the actors’ intentions and cannot only rely on better instruments to measure behaviors. Furthermore we should also cultivate our critical thinking instead of fighting a losing battle when trying to update research results against the overwhelming flow of data about new technologies that we receive each day. The act of being reflexively critical is essential because (1) it helps see knowledge connections with other domains of application and thus questions the generalizability of our knowledge contribution, (2) critique is part of the mission of journals such as EJIS and (3) it is an underutilized resource that is not time-consuming but, if well-used, can greatly enhance the potential of our contributions. It is our reason that allows us to sort the ivy from the chaff. Moreover it is our critical thinking that allows us to take into account normative concerns and to anticipate and project the possibilities for designing the future. If IS is also about designing, we cannot only rely on observation. We also need theory to guide our reflection and endeavors. The editorial board has noted that both MISQ and JAIS have very aptly taken the theoretical turn in the 2000s and I am convinced other journals should as well. Turning now to guidelines for theory papers, I recommend reading Ron Weber's comments (2003b) and his recently developed version (2012), where he recalls that a good theory needs constructs and well-developed relationships. This is certainly a very clear guide for those who, like him, consider that theories should explain and/or predict. However, I believe that EJIS should be open to all epistemologies. As a critical realist, I was happy to see that Ron recently extended his first comments by considering that theories are not only for explaining and predicting (Weber, 2012). The philosophy of social sciences also considers understanding as an important goal for theorizing (Gregor, 2006). Understanding is also deeply rooted in the interpretivist and human sciences tradition (Habermas, 1988). Moreover, among the few theories we have in IS, theories such as Colonial Systems theory (Porra, 1999) are not always presented in terms of constructs and relationships and yet these can be powerful and inspiring. Moreover, if Gregor (2006) argued that there are normative design science theories, why shouldn’t IS journals accept theory-only papers in design science?

Second, Weber continues to insist on parsimonious explanations. Parsimony is a goal of explanatory research, which is also very useful for conveying knowledge when teaching. However, for many of the fuzzy phenomena we study, it is very difficult to distinguish the context from the central problem we attempt to study. This is precisely what case studies are good for. Rich description of the context is a condition for generalizing further and I would argue that too often we apply models or refine theories to use them for situations that have not been described before and for which we know very little, such as IS governance or IS-enabled organizational transformation. As I have argued in my former editorial, we must also get closely acquainted to the phenomena and develop narratives reflecting events and actions where actors are not absent of the scene (Ramiller & Pentland, 2009). This will take us to my last point (cf. 3).

For all the reasons above, pure theory papers, that is papers that allow for normative developments (Mingers & Walsham, 2010), or that build on past empirical research to enlarge the traditional research goals (Gregor, 2006), or are essays on the ontology and philosophy of information systems (Beynon-Davies, 2010; Kettinger & Li, 2010), or essays on the philology of information systems (Cordoba et al, 2012), or critical research allowing to rethink our theoretical categories and research genres (cf. the Kleinian EJIS Special Issue) should be most welcome in our journals (Markus & Saunders, 2006). All these genres have in common that they allow us to rethink our philosophical assumptions. Literature reviews and theory papers, as just described, have in common that, to make a contribution, they need to go beyond the orthodox knowledge and show why a different way of thinking promises to be fruitful.

Papers that produce a new typology or an analytical framework can also be interesting and can be considered as theory papers (Gregor, 2006). It should not always be necessary to test these typologies in the same paper as the one offering the theoretical development. If the theoretical description of the typology is strong enough, and based on an in-depth treatment of the compatibility of characteristics between different constructs that are part of the theory, why would we also, in the same paper, require some theory testing? Why should we keep also a restrictive understanding of types as being necessarily tested against some performance criteria, as Doty & Glick (1994) argue? Most certainly, it is interesting to complement theoretical knowledge with empirical data describing the different forms that actually exist in the field. But if the theoretical arguments are correct, this empirical knowledge will only enrich its description and reinforce the theory by delineating more precisely boundary conditions for the occurrence of the phenomena (De Corbière & Rowe, 2010). In the very classic sense a good theory paper rests on arguments that build on the literature and stand by their consistency and compelling logic.

Now one must wonder if it is possible to provide a simple guideline for all these types of theory development papers so that they can be more easily (auto)-evaluated? To be honest, I am not so sure. What seems to matter is if they add or change existing theories with arguments offering high face validity. But apart from reason and expert judgments they are more difficult to evaluate than the prevailing genres of empirical research.

Research essays, which usually relate to research methods (see the two Special Issues on Qualitative Research and Quantitative Research, which appeared this year in the first two issues, and the Grounded Theory Special issue in the first issue of 2013) or research practice (e.g., Damsgaard & Lyytinen, 2011; even if this one appeared as an ‘opinion paper’, which it is not) or specifically refer to (i.e., extend or criticize) a paper that was recently published in the journal, are arguably of a different nature. But my preference is to name them in the category otherwise journals may not receive such papers and to decide later if they can be singled out as a category. For the moment I suggest grouping it with the Theory Development papers since certain types of research essays are also found in the Theory and Review department of MISQ Markus & Saunders (2006) and since both are notoriously difficult to evaluate. Their evaluation is difficult because we are not very accustomed to the genre and therefore it will take patience, goodwill and practice to make the editorial team converge in its recommendations. But isn’t the situation the same in fact with empirical papers? For the best ICIS and MISQ papers Matthew Jones (2004) convincingly argued that, with the exception of his observation that ‘inter-textual reinforcement appearing to be at work in positivists’ reuse of others’ instruments and constructs’ (and we note that inter-textual reinforcement may apply across all types of genres), there was no explicit rule that came out of the analysis of the methodologies of the best papers published. He therefore suggested that, rather than being the best followers of the scientific method, the papers were successful because they were telling a story that was convincing. This maybe the sign that a good methodology requires an adaptation of the method to the problem and that a good paper has to adjust the presentation of the method in a way that takes into account the strengths of the research. That it defeats a standardization and prescriptive approach to a method also shows the need for creative thinking in other genres than theory testing papers.

Ethnographies and narratives

We need powerful and smart techniques to describe situations rarely observed and for which a better understanding may have important consequences such as how CIOs coordinate with the Top Management Team members, how the IS function transforms and why it is not often perceived as contributing to value creation, how information systems deliver services over complex supply chains, or how people work with IT in control-command rooms in nuclear plants. Beyond the observed reality, there is also a need to understand better how intentions develop and how people take stances or make compromises if one accepts that management is at least partly about paradoxical injunctions.

We must bear in mind that we can never model or represent reality without selecting and bracketing out various aspects not only because we have limited time and resources, and because some of the situations I just referred to are not easy to access, but because we are always describing from a certain point of view (Gadamer, 1975; Klein & Myers, 1999).

To observe situations such as the ones evoked above requires negotiating windows of presence on site, playing a dual consulting-researcher role or doing ‘auto-ethnography’. Since experience is the main source of learning, immersion into the lifeworld of those who live what we want to study is the best way to go. Ethnography is a privileged research method if we can devote enough time and effort and are able to observe and feel what is happening or not. In fact, through immersion the researcher not only gains an in-depth understanding of the actors’ viewpoints but of their broad context in which they work. Myers (1999) identifies three types of ethnographies: (1) the holistic school that requires an identification with the social group being observed. ‘Going native’ is a precondition to better absorption and understanding of the culture to convey the meanings of a lifeworld; (2) the semiotic school that analyzes symbolic forms – interestingly this can also been done without being there when one can access such material (Sarker & Sahay, 2004; Beynon-Davies, 2010); (3) the critical ethnography that questions the status quo and domination conditions by a rich dialog between the ethnographer and the people in the research setting. The literature also refers to virtual ethnography (Sarker & Sahay, 2004), which may sound as an oxymoron but relies on face-to-face observation of the participants and on symbolic observation and interaction with them in the electronic space.

But it is not enough to simply live with the natives to draw the lessons. To develop this ability ethnographers memorize (record traces in some way) what happened and suggest or allow to infer from the sequence of events why things happened. Causes and reasons can then be identified and theories developed through narratives. What distinguishes narratives from canonical variance theories is the inclusion of focal actor or actors and an identifiable narrative voice (Pentland, 1999). Even in the realist tale mode, narratives ‘carry cultural values because they encode implicitly or explicitly, standards against which actions of the characters can be judged’ (id., pp. 712–713). Since narratives do not remove actions, actors and events, their vividness and accuracy can be better recognized by managers (Ramiller & Pentland, 2009). Ethnography allows the writer to build narratives accounting for an experience, to argue about relationships, and to describe and qualify objects. It can be nicely complemented by other methods such as grounded theory (Sarker & Sahay, 2004; Mattarelli et al, 2011). As an ethnographic technique, the narrative does not aim at describing what has been witnessed, but what has been lived or done. The ethnographer at play can depart from the rigor of the representation sought by the scientist and evoke a discursive experience (Harvey, 1997). Story-telling is thus a way to organize a discourse and the resulting narrative. Classically, story has two meanings. One can distinguish the story, i.e. the set of events that we want to report, and the narrative, i.e. what is being told.

Hyvärinen (2008) considers three narrative turns: in literary studies in the 1960s, in historiography and in the social sciences in the 1980s, the latter being characterized by a positive appraisal and humanist approach to psychology and culture. Riessman (1990) identifies three genres, (1) the proper story (Labovian model with the following parts: abstract, orientation, complicating action, evaluation, result, coda, (2) habitual narratives and (3) hypothetical narratives. Among the great diversity of narratives (Barthes, 1977) several types can be of special interest to the IS community. I will only briefly mention some of them, according to their function:

  • The ‘professional life narrative’, which can help make sense of a particular social position, reveal a personal trajectory and carreer evolution in IS. Such narratives would vividly complement cognitive maps describing situations such as the challenges facing women in the IS function (Reid et al, 2010), or the difficulties faced by different actors in IS development (Baskerville et al, 2006).

  • In the grammarian and Labovian tradition, based on the analysis of linguistic sequential variations short narratives can be used for reorganizing work and processes (Pentland, 2003). This type of narrative shows that (mini) case studies can also be a good research design to develop narratives. In such situation they would nurture the ‘ethnographies and narratives’ category rather than the (prevailing) ‘empirical research’ one.

  • Learning accounts (Boje, 1991; Soulier, 2000), which can be based on realism through a chronology (van der Blonk, 2003) or interpretivist research (Schultze, 2000; Niehaves, 2006; Sandberg & Mathiassen, 2012) are useful for memorizing and learning from experience. The latter, the confessional account uses autoethnography, which refers to the situation when the native has been in a particular position and wants to reflect upon it; the confessional genre would be appropriate to the Professionally Qualified Doctoral Student (Klein & Rowe, 2008). Even if the author and the subject are not the same person, in a narrative a large space can be given to reflexive conversations so that the reader can in turn reflect upon personal experiences, to the self and relevant ethical issues (Coffey, 1999, Olsen, 2012).

In offering guidelines for such ‘narratives’, three main remarks will be enough for my purpose. First, as there are many genres in ethnography and narratives, each of them should be evaluated with respect to the main genre(s) and to the epistemology it adopts. Beyond the well-known criteria developed by Klein & Myers (1999), which fit with interpretive ethnographies, Myers (1999) evokes more generally four criteria: (a) contribution (novelty and capacity to convince the journal editorial board of this), (b) rich insights (one way to address this being to consider whether it contradicts conventional wisdom), (c) significant amount of data collected (involvement of the researcher on the field to get data; contextualization, multiple stakeholders perspectives), (d) sufficient description of the method. Depending on the narrative genre (realist tale; impressionist tale; confessional tale) and corresponding type of ethnography (‘scientific’ for the realist, ‘at play’ for the impressionist (Harvey, 1997) evaluation criteria will be different. For Schultze (2001) the confessional account should exhibit five characteristics: the willingness and ability to become immersed, the availability of a supportive thought community, the tenacity to carry on despite ambiguity, the discipline to write self-reflexive field notes, and the ability to identify ways in which this self-reflexive material can be used to make substantive contribution to IS research. Second, as for any type of research the evaluation should look both at the narrative and what is not told. Hyvärinen (2008) puts emphasis on how language practices, positioning and expectation analyses ‘direct attention to the fact that narratives not only account for past experiences but position speakers within networks of social and cultural expectations’ (p. 457). Through many linguistic expressions, evidence of negative expectations tell what did not happen and serve the sense-making function of past and relevant futures (id.) and thus can significantly contribute towards the building of process theories (Langley, 1999). Third, in genres that depart from the realist one, aesthetic criteria and the capability to provide insight with the writing style matter. When I first wrote in a former editorial that we should be more flexible on writing styles (Rowe, 2011), I meant that we should accept more diversity there, not that anything unconstructed would be accepted. In ‘writing: a method of Inquiry’ Laurie Richardson & Elizabeth Adams St Pierre (2008, p. 478) explain ‘we do not triangulate: we crystallize’. They refer to CAP ethnography (creative analytical processes ethnographies), for which they use four evaluative criteria (substantive contribution, aesthetic merit, reflexivity, impact). They indicate ‘Science is one lens and creative arts is another. We see more deeply using two lenses’ (id., p. 480). It would be nice if presentational genres in IS research could also include aesthetic merit and reflexivity in their criteria while keeping substantive contribution at the foreground. Ethnographies and narratives should be encouraged in that direction.

Conclusion

In this editorial, I have proposed a new categorization for IS research genres: 1. Literature Reviews, 2. Theory Development and Research Essays, 3. Empirical Research, 4. Ethnographies and Narratives, 5. Issues and Opinion. This new categorization has consciously left out many of the most innovative presentational genres, such as those using video, play or theatre (Avital & Vandenbosch, 2000; van der Blonk, 2003; Schultze & Avital, 2011), for future developments by the experts. I have also left out other genres that we think do not fit well with the theoretical positioning of many top journals. In particular these journals are not the best place for data-driven research papers with long surveys or description of work systems for industry or governmental agencies. One reason is that such papers often lack the theoretical interpretation and discussion that these journals mostly favored. Even if they are really useful for academic teaching or for decision-makers as well as for further positioning research, these papers are best suited in their own publishing channels such Pivot, a new journal launched by Palgrave MacMillan.

Although useful, these guidelines should not become a straitjacket, especially when creativity is emphasized and valued. However, they should help convey a spirit which, once appropriated by members of the community, will help them see how their own ideas and visions can transform into innovative papers.

In the future, the IS community may also consider ‘Alternative Genres’ as a new and larger class of publication genres, which would encompass the narratives and presentational innovations such as theatre, video and other types of presentational genres. The alternative genre of publication is not only a new type of research output (e.g., a movie), and as such potentially a sixth type of genre or creation comparable to the status of those that have enriched modern literature (Duff, 2000), but it is also a supplement and an enhancement of prevailing genres. Therefore the question that remains open is whether the ‘alternative genres’ category should be created as such. The main benefit is to have an identified vehicle for getting our most creative ideas out. However, when it comes to establishing a formal category named ‘alternative genres’ this might – in the long run – be counter-productive. Papers published under that label might be discredited as outsider work, niche interests, etc. The high idea that we have for the alternative genres should position it as fully qualified and valid research.

To conclude I would like to invite the submission of research or opinion papers in EJIS that would help us flesh out in more detail the guidelines on alternative publication genres such as those mentioned above or any other specific publication genre (Stahl, forthcoming). I am also pleased to invite you to attend panels on theater and alternative genres at ICIS in Orlando where some of our Associate Editors will further stimulate these ideas and to submit to a future special issue in EJIS on alternative publication genres.

In this issue of EJIS

This issue includes six prominent IS and IT articles. We begin by an article that explores the IS discipline evolvement by tracking the publications of both EJIS and MISQ over a 15-year period. Next is an article about the applicability of knowledge sharing in a Chinese situated Knowledge Management System (KMS) project. The third article is a study about nearshore location characteristics. The fourth article discusses the benefits and dangers of IT enjoyment while using an IS tool: social network websites (SNW). The fifth article studies the intersections between organizational routines during the introduction of a health IT system. Finally, the last article adopts the sociomateriality framework to present important findings that concern boundary objects during an Information System Development (ISD) implementation project.

In the first article ‘Information systems as a discipline in the making: comparing EJIS and MISQ between 1995 and 2008’, three scholars: Josè-Rodrigo Còrdoba, and Alan Pilkington from the University of London and Edward W.N. Bernroider from Vienna University of Economics and Business investigate the IS as a discipline embracing Abbott's (2001) proposition that social science disciplines are part of a wider cultural and social system. Abbott suggests three iterative stages for a discipline: (a) Differentiation where the discipline claims ‘jurisdiction’, (b) Competition where the different distinctions made about problem diagnoses, treatments or inferences between the two generate competition within and between disciplines, and (c) Absorption where a set of distinctions and connections between them absorbs or ingests others who were previously their competitors. For their study, the researchers analyze the articles published in both MISQ and EJIS between 1998 and 2001 to determine in which exact stage they were. They further zoom into 5-year time brackets to refine their analyses and avoid too large segregations. For this purpose, they use bibliographic citation and co-citation to map knowledge distinctions and their connections. This produces clusters of themes and links between them for each journal, respectively. The diagrams help the researchers interpret the corresponding discipline stages, and identify which themes constitute mainstream IS discipline and which, on the contrary, surfaced, or vanished. The study also indicates a high dominance of published research with a positivist orientation.

In the second article ‘To share or not to share: a critical research perspective on knowledge management systems’, the researchers: Mei-Lien Young from the Nanhua University in Taiwan, Feng-Yang Kuo from the National Sun Yat-Sen University in Taiwan and Michael D. Myers from the University of Auckland, New Zealand investigate the relationship between cultural values and norms and the production of knowledge in a KMS. For this purpose, their case study is situated around the SCTNet KMS used by Taiwanese teachers. Forty-nine participants from different genders and rankings are interviewed over a 6-month period. They use the principles for critical research advanced by Myers & Klein (2011) and in particular Foucault's concept of the gaze. The gaze reflects a feeling developed in reaction to the implementation of a technique to control those who are gazed upon. The gazers need not be there physically; however, they leave the gazed upon with the feeling that they are constantly being scrutinized for their behavior. The researchers equally use the concept of Face, which is very relevant for the Chinese culture. Face is one's situated identity reflecting the subject's sensitivity to how s/he appears in the eyes of others. The study findings show that the Taiwanese teachers influenced by Confucian principles are reluctant from participating in public KMS because of this feeling of the ‘invisible eye’ or of the need to ‘keep quiet to save face’. As these feelings develop, they collaborate and share content more freely in private KMS spaces with their peers and people they feel close to. A broader study finding is that cultural variables such as those witnessed among the Chinese teachers population may strongly affect KMS usage, sharing, creation and distribution of knowledge.

In the third article ‘Everywhere and nowhere: nearshore software development in the context of globalisation’ co-authored by Pamela Abbott from Brunel University and Matthew Jones from Cambridge University, the location characteristics and attractiveness of the software development project are discussed. A typology of espoused, unanticipated and remediable location characteristics is furthermore developed. The study's particularity is that the location is discussed from a service provider's perspective. For this purpose, it illustrates two cases where location decisions for the software development projects were similar, but the results were significantly different. The study suggests that the location decisions in the two cases are based on place and space-based logics that draw from both the International Business and Information Systems literature on location characteristics. But it also attracts special attention to the less known contextual variability of these location characteristics such as the surrounding culture and socio-political environment.

In the fourth article, ‘The benefits and dangers of enjoyment with social networking websites’, Ofir Turel from California State University and Alexander Serenko from the Lakehead University explore an IS usage model applicable on SNW. The enjoyment of IS usage is argued to lead to high engagement, which is perceived as being positive, but could conversely lead to habit forming that might push for IT addiction. The latter effect has been mostly ignored in the literature as an enjoyment causal effect. The study also shows some notable theoretical and practical implications. For theoretical implications, we highlight the divergent outcomes of IT perceived enjoyment and that IT use shall no longer be always considered as a desirable phenomenon. As for practical implications, users and their surrounding community should become conscious of these potential IT usage harms and try to regulate usage through training, dissuasion, etc. SNW developers might also need to revise their responsibility both officially (through disclaimers for example) and unofficially (through website design and associated features).

The fifth article ‘Mediating the intersections of organizational routines during the introduction of a health IT system’, co-authored by Laurie Novak, Cynthia Gadd, Shilo Anders and Nancy Lorenzi from Vanderbilt University School of Medicine, and JoAnn Brooks from Syracuse University examines the problem of managing interdependencies between different organizational routines in information system projects. With a qualitative case study approach, conducted at an educational hospital, and by using Feldman and Pentland's model of organizational routine, the researchers demonstrate how an Informatics Support Team, composed of nurses, is able to manage the unintended consequences arising from the introduction of a barcode technology for medication administration routine. This medication administration routine intersects with three previously stable-related routines: pharmacy verification and scheduling, next shift medication administration as well as with management reporting and oversight routines. This paper highlights that the routines are mostly nested with each other (inter-related). These intersections become a major focus of analysis to get a better understanding of the impact of unintended consequences due to the introduction of the IT artifact on a particular organizational routine, and the subsequent staff's workarounds. This paper also provides a good illustration of the criticality of the metastructuring activities in the management of various ‘problems’ in post-adoptive phase. The role of mediators is rendered therefore of prime importance at the routines intersection.

The sixth article, ‘Sociomateriality and boundary objects in information systems development’, co-authored by Bill Doolin and Laurie McLeod from the Auckland University of Technology, studies an ISD longitudinal case to illustrate project-related artifacts as boundary objects with a sociomaterial perspective. Particularly, the study analyzes a prototype developed at earlier stages of the ISD, and its relevance as a boundary object for the project's expected and unexpected results, situations and performances. Its theoretical framework allowed the researchers to anchor several findings of their study. First, the prototype developed for the ISD (as a boundary object) was found not to be an independent technical object, but formed a boundary object in the performance of the sociomaterial practices of different stakeholders. In turn, its performance as a boundary object helped to configure those situated practices. Second, the prototype emerged temporally in practice as a sociomaterial assemblage, being reconfigured over the course of the project in response to particular situations. Third, certain possibilities for knowledge transfer, its interpretation and assigned interests between project participants using the prototype were the results of the sociomaterial agency produced from the constitutive entanglement of human actor and material artifact. Fourth, those diverse sociomaterial assemblages incorporating the prototype were performed differently across different occasions, sites and participants, producing varying effects. Finally, each project participant group enacted the prototype as a boundary object differently in specific sociomaterial practices. Thus, rather than a singular boundary object, a multiplicity of boundary objects were performed.