The virtue of brevity

Abstract

Before the rise of online-only publishing, journals, including the ISJ, operated with page budgets, which determined the annual limit of pages that could be published. A decade ago, at the ISJ, this amounted to some 576 journal pages. However, with the advent of online-only publishing, the concept of a page budget has disappeared. The total number of pages increased gradually, at first to 666, to 800 and then it jumped to over 1000. Wiley, the publisher of the ISJ, has informed us that there is no limit at all to how many pages (or articles) we can publish a year, so long as quality standards are maintained. In 2023, the ISJ published 43 research articles, 8 editorials and 3 book reviews, a total of 1458 journal pages or 253% of the pre-online-only model. We expect that in future these numbers will continue to rise.

Not only has the number of accepted articles increased but also has their length. It used to be the case that we requested authors to keep within 8000 words. This was supposed to be an all-inclusive word count from title to references and appendices. With the passing of time, the restrictions of paper length have largely disappeared. Table 1 below presents the guidance given to authors for the AIS Senior Scholars' List of Premier Journals as well as several others. While there is some wiggle room, the general guidance tends to sit within the 8000- to 12 000-word range, or 38–55 pages. This is similar to adjacent fields: the Academy of Management Journal allows 40 double-spaced pages including references, tables, figures and appendices,1 and Organisation Studies permits up to 13 000 words encompassing references, tables, figures and appendices.2 Computer Science journals tend to publish briefer papers. For instance, IEEE Transactions on Pattern Analysis and Machine Intelligence allows 35 double spaced pages3 and IEEE Communications Surveys & Tutorials charges authors $220 for per page over 30 pages (up to 8 pages).4

Today it is not unusual for a first-round submission to be in the 12 000-word range and for a finally accepted article to be 15 000. Even 20 000-word epics are by no means unheard of. Why all this verbosity? One reason is that authors and reviewers get trapped in what we call a text-accumulation trap. Authors submit papers with as much information as possible, hoping to instil confidence in the review team as to its completeness, as well as to prepare for both expected and unexpected reviewer demands. Reviewers then read the paper, yet ask for even more detail, and authors feel obliged to respond in kind. Over multiple revision cycles, satisfying all the reviewers requires ever more words. Reviewers should be mindful of the text-accumulation trap. When requesting authors to incorporate large amounts of new material, it is necessary to weigh the value of this extra material against the risk of unnecessarily bloating the manuscript.

In addition, it is easier to write more than less, easier to include more literature, more cross-references, more examples, more arguments and more details. In contrast, it is harder to be parsimonious with words, even if our theories are often lauded for their parsimony. Blaise Pascal noted in the 17th century ‘I would have written a shorter letter, but I did not have the time’.5 It takes much careful thought to reduce the length of a manuscript, yet to preserve the essential message.

Editors used to say things like ‘your article is accepted but … I need you to cut at least 3,000 words’. However, with the page-budget restrictions of print journals largely gone, editors are less inclined to make such demands. We have not seen such a request in many years and our less experienced readers may be shocked to read this historical anecdote! With the disappearance of this means of reducing manuscript length, the previously balanced system is now out of balance. As a result, submitted manuscripts get ever longer, ever more turgid and, regrettably, ever less enjoyable to read.

In this editorial, we argue for brevity rather than verbosity. First, it is instructive to draw analogies with the short story, as opposed to the epic novel. Boyd (2005, p. 238) writes that ‘there are twenty-page short stories that are far more charged and gravid with meaning than 400-page novels’. Is the same true for work in our own field? Would our work be more impactful, more read, more appreciated if it was shorter, crisper and more precise and if we took the time and trouble to alembicate our insights before submitting them for review? For authors, there are several reasons to submit a shorter, more concise manuscript.

One is that the main ideas will stand out more clearly, not be hidden among text that becomes noise precluding the reader from getting to the main message. No manuscript will attract support from reviewers if its contributions are invisible.

Another reason is that reviewers almost invariably suggest modifications and additions that need to be undertaken before the manuscript can be accepted. If the manuscript is already long, adding text means that something else may have to be deleted which can create new issues and the feeling of a lack of progress.

As we enter the era of generative AI, there is the potential for a rise in the number of submissions that are well written and grammatically correct but that drown in their own verbosity, constituting clear exemplification of unnecessary padding. The flip side of this, as many of us have experienced, involves students resorting to AI tools to summarise a paper. Although this adds some value, it also highlights the downside of papers that are overly verbose: the initial response from many readers is a reluctance to read it.

Finally, we observe that it is increasingly difficult to find people who are willing to review 12,000- to 20,000-word monster submissions! Potential reviewers reply to invitations, saying ‘yes, I like the topic and I know the method, so I am willing in principle to review it, but … how long is the manuscript?’. When the truth is revealed and they see that they are in for a marathon read-and-digest effort, their willingness to continue wilts noticeably. A few do stay the course, but to get two good reviews, we may have to invite 15–20 people, which unavoidably delays the whole review process. It might be better to start off with a shorter submission of 6000 words and to revise it upwards towards 8000, rather than starting with 12 and heading for 20.

Beyond the rather tedious exposition of numbers, there are arguments for the value of brevity itself. A haiku, or a call for laconicism, takes the analogy too far, but we should be contemplating how we can focus our thoughts on what really matters and distilling our research into consumable portions. Maybe we can take a cue from journals like Nature or The Lancet that deal with complex concepts and life-saving studies, while keeping word counts under 5000.

A genre where this is particularly apparent is the practitioner paper. We seriously doubt if practitioners, as readers, have the time, energy and patience to wade through excessively long articles. At the ISJ, we ask that practitioner papers not exceed 8000 words, inclusive of all contents.6 We also have detailed guidelines and address the topic of practitioner papers in a recent editorial (Davison et al., 2023). Practitioner papers should mirror genres commonly used in formal business communication: concise and to the point business reports and memos. Abstracts should read as executive summaries, focusing on the real-world context of problems and key recommendations. It is no surprise that the Harvard Business Review asks authors to submit an 800-word narrative. Practitioners are more likely to value shorter, more pithy work that dispenses with unnecessary literature, method and analysis, though this material could be retained in an appendix. Practitioners value what you are doing and why: the motivation, problematisation and question. They will want some background, but not in excess, and only to help them ensure whether the situation is applicable to their specific context. They probably do not want excessive references. They will want to know briefly what you did and how, but will value more the prescriptive (and proscriptive) recommendations that you have the confidence to make. This implies a writing style focused on action for the practitioner reader who faces these real-world problems or challenges. However, do not mistake the call for brevity for abandoning a logical chain of reasoning from observation to recommendations. Indeed, practitioners will want to assess the believability of recommendations especially if these have substantial, even existential implications for their organisations. It may be that this shorter version of the research output should parallel a longer version. Use concise tables and informative figures to condense verbose text in the body of a paper. Ancillary materials could be consigned to an online appendix. Communications of the ACM, some years ago, specified a tight word length (4000), a short list of references (max 10) and tables/figures (max 4 each). That may be too restrictive for our tastes, but even if we increase the numbers by 50% the result will still be consumable.

In this second issue of volume 34, we present eight articles.

In the first article, van Offenbeek et al. (2024) note that when there are misfits between information systems and people's work practices, workarounds often emerge in response. In complex work environments, working around information systems is regularly required to get tasks done. Many workarounds are functional and may become legitimate practices. However, in the current study, the authors show how some workarounds can, in turn, lead to new or increased misfits for subsequent users. The authors report on a multi-year, in-depth, qualitative exploration of the use of an electronic health record system in two tertiary hospitals. The authors report how escalating patterns of misfits and workarounds emerge through the interplay among work practice routineness, misfit characteristics and non-compliant workaround behaviours. By differentiating the concept of non-compliance, the authors arrive at a conceptual distinction between, on the one hand, three misfit-aggravating workaround patterns and, on the other hand, those that leave the misfit unaffected or even reduce it.

In the second article, Gong et al. (2024) suggest that social factors play a critical role in motivating player participation and commitment to online multiplayer games. Many popular mobile massively multiplayer online games (MMOGs) adopt social network embeddedness (SNE) functions to optimise players' social play experience. SNE changes the traditional pattern of MMOG social play by porting acquaintance relationships (e.g., Facebook friends) from social networking sites to the virtual game world. Drawing on the affordance framework and social capital literature, the authors propose a theoretical model that integrates the factors of SNE technology affordance (identity transparency and information transparency), players' social experience (social interaction, social support, shared vision and social pressure) and affordance effects (play performance and play frequency). The model was validated through a longitudinal field study, in which both subjective and objective data were collected. The study enhances the theoretical understanding of social relationships in players' game participation results from the SNE aspect.

In the third article, Müller and Sæbø (2024) describe how the Scandinavian Journal of Information Systems was hijacked by cybercriminals who created a fraudulent website to scam authors. The criminals sent fake acceptance letters and charged authors for publication, while the website appeared legitimate and was linked to Scopus, Elsevier's well-known citation database. The Editor-in-Chief discovered the scam after an author reported receiving both an acceptance letter and a desk rejection for her manuscript. The investigation revealed that several authors had been scammed, and the fraudulent website appeared in online searches. Despite reporting the crime to various authorities, the fraudulent site remains operational, highlighting the difficulties in stopping such cybercrimes. The authors discuss implications for the IS community and strategies for combating this nascent threat.

In the fourth article, Laumer et al. (2024) examine the conditions under which applicants in personnel selection processes perceive algorithms to be fair and develop a theoretical foundation for algorithmic fairness perceptions. Drawing on an online application scenario with eight experimental groups (n = 801), the authors analyse the determinants for algorithmic fairness perceptions and the impact of interventions that are based on transparency and anthropomorphism. Embedded in a stimulus-organism-response framework and drawing from organisational justice theory, the authors reveal four justice dimensions (procedural, distributive, interpersonal and informational) that determine algorithmic fairness perceptions. The results further show that transparency and anthropomorphism interventions primarily mainly affect dimensions of interpersonal and informational justice and highlight the importance of algorithmic fairness perceptions as critical determinants for individual choices.

In the fifth article, Durani et al. (2024) recount how historically, gatekeepers, like elites and institutions, controlled visual information to serve their agendas. However, the rise of social media has given rise to ‘visual audience gatekeeping’, where users shape information flow. This evolution offers empowerment but is not without risks, as visual content's persuasive nature can be exploited for persuasion and manipulation. The authors scrutinise visual audience gatekeeping in the context of the Russo-Ukrainian War, focusing on a Russia-related Reddit subforum. As audience gatekeepers, users disseminate visual content that aligns with their embraced social reality. The outcome is a ‘visual echo chamber’ characterised by limited information diversity, thereby corroborating but seldom interrogating the worldviews of audience gatekeepers. Amid heightened tension, this cycle intensifies, increasing the potential for radical visual narratives to gain ground. The study culminates in a novel theoretical model of visual audience gatekeeping, offering broad implications for the fields of visual communication, gatekeeping theory and social media platforms.

In the sixth article, Hardin et al. (2024) use a theory-driven, deductive approach to examine the concept of virtual team efficacy. The authors draw on social cognitive theory and trust theory to develop a research model in which virtual team efficacy relates indirectly to virtual team outcomes through trust and perceptions of problems with time differences, geographical separation and cultural differences. Findings from a multi-wave survey administered during a comprehensive virtual team project confirm the research model hypotheses. The authors conclude by emphasising the importance of expanding the virtual team efficacy nomological network because, unlike collective efficacy concepts developed for traditional team settings, virtual team efficacy focuses on the technological components characteristic of virtual team projects.

In the seventh article, Lumbard et al. (2024) examine how open-source software continues to play a critical part of our shared digital infrastructure. The authors explore how open-source software is understood to meet acceptable standards of health and sustainability as it is relied on as a key component of organisational work. Open-source community research has explored issues of health and sustainability, often focusing on the characteristics of single open-source projects. Using social comparison as a research frame, the authors explore how open-source project health and sustainability are understood, not in isolation, but across communities within software ecosystems. The authors provide insights for how people explore and compare interactions present across open-source communities, demonstrating open-source community health, not a set of discrete metrics, but is an ongoing social construction.

In the eighth article, Aguirre-Urreta et al. (2024) engage with the debate on formative modelling (‘formative measurement’) in the information systems discipline and in particular the claim that measurement model misspecification, where a reflective model is used instead of a more appropriate formative model, is widespread. The authors argue that this misspecification cannot be true because models mis-specified in this way would fail the measurement validation procedures used with reflective models and thus would not be publishable. The authors present two extensive simulation studies to support their contention. The simulation results show that in most cases where data originate from a formative model, estimating a reflective model would not produce results that satisfy the commonly used measurement validation guidelines. Based on these results, they conclude that widespread publication of models where the direction of measurement is mis-specified is unlikely in IS and other disciplines that use similar measurement validation guidelines. Building on recent discussions on modelling endogenous formatively specified latent variables, they also demonstrate that the effects of misspecification are minor in models that do pass the model quality check. Their results address important issues in the literature on the consequences of measurement model misspecification and provide a starting point for new advances in this area.