Editorial: Publishing conference proceedings for low-budget conferences: a case study of ISIC 2022
Laura Rothfritz, and Elke Greifeneder
ISIC 2022 took place in late September 2022 in Berlin and was hosted by the Humboldt-Universität zu Berlin. ISIC: The Information Behaviour Conference is the academic home of the information behaviour community and held bi-annually in different locations. The proceedings of the conference are published in the open access journal Information Research. The aim of this editorial is to offer insight into the processes of publishing conference proceedings for a low-budget conference and to help future organizers in their planning. This editorial is written by the conference chair of ISIC 2022, Elke Greifeneder, and Laura Rothfritz, the proceedings chair of ISIC 2022, both members of the hosting site, the Berlin School of Library and Information Science.
A low-budget conference is a conference that requests reasonable conference fees, but would not be able to run successfully without the strong engagement of the local conference organizers. Low-budget means essentially that the conference is organized without an external event company and that all organizational tasks have to be fulfilled by members of the community. This includes both the local organization such as catering, room planning or name badges as well as the reviewing and publishing part of the conference proceedings.
The publication of the proceedings by Information Research staff is a commitment to the community since staff do not ask for extra payment to publish the proceedings, compared for example to proceedings published in ACM journals, such as the iConference proceedings. Before deciding on publication venues of proceedings it is thus important to calculate the costs for either local staff or for an editing and publishing service.
Low budget conference organizers rarely realize at the beginning that the most difficult part of the conference are the proceedings. Few organizers have editorial experience and lack the knowledge of the many steps that proceedings involve. In fact, few bids for proposals for conferences offer details about the work that the proceedings involve. This editorial wants to shed some light into the steps needed to move from article submissions to published articles.
Use of conference management systems
Low-cost conferences do well to invest in a good conference management system, since it helps tracking submissions much easier than self-made spreadsheets and enormous emails sent to hundreds of people. Conference management systems allow the sending of bulk e-mails to (groups of) authors or (groups of) attendees and, most importantly, to track submissions on various variables, for example to mark submissions that do not fit the requirements, where final uploads were missing or to identify authors of accepted paper that are not yet registered.
ISIC 2022 worked with the conference management system conftool which offered intensive services at the one-time cost of a conference registration to ISIC 2022. A comparable, but at present more expensive, alternative is EasyChair using the pro version.
Low-cost conferences also ought to consider planning with teams with responsibility instead of assigning crucial tasks to one person and to document open, ongoing and finished tasks so that in case of a fall-out, (for example, due to COVID), the team can take over. For ISIC 2022, we worked with Trello as a task management system in addition to the use of conftool.
Templates and reviewers
Submissions for ISIC 2022 included full and short papers and authors were required to use a template provided and developed by the organizers. The template was available for Word and LaTeX, but only three authors used the LaTeX template. ISIC 2022 received 60 full and 23 short papers. Directly after submission, a student assistant spent roughly 20 hours on checking each submission for fulfillment of the requirements including lengths of submission, use of template and anonymization. Ten percent of submissions exceeded the word limit and were given two days to shorten their paper and another other nine percent were non-anonymized, which we rectified ourselves where possible.
We used the automated reviewer assignment function in conftool, which allows for assignments based on topics of expertise that reviewers had to choose beforehand. For ISIC 2022, we made an explicit choice not to go for paper bidding, but it is certainly an alternative. Conftool also automatically checks for conflicts of interest, which is strongly recommended instead of manually checking it during the assignment process. We assigned two reviewers per paper, because the community is small and we did not want to overburden reviewers. 105 researchers from 22 different countries completed reviews for ISIC 2022. Unfortunately, editors’ reality of reviewers’ punctuality also held true for ISIC 2022 and of the 166 required individual reviews, 27% were submitted late. Conference organizers do well to either go with more reviewers per paper, so that a missing review does not hurt as much, or to plan for the extra two weeks that are required to receive late reviews.
Acceptances were sent out in March 2022, that is seven months before the conference took place. Authors had then four weeks to include reviewers’ feedback and to upload a final version.
Checking revisions, proofreading, formatting
Publishing the proceedings also offered a very good learning opportunity for students at the Berlin School of Library and Information Science, Humboldt-Universitaet zu Berlin. We offered a research project seminar where a team of nine students worked together with the proceedings chairs. Each student was assigned four or five papers and was responsible for communication with the authors. Students had the opportunity to experience the peer review process very closely: for example, their initial task was to check authors’ revisions and comments to reviewers’ feedback, thus deciding if the content of the submissions was ready for publication. This process took about three weeks and was closely monitored and supported by the instructors and the paper track chairs.
In a second step, the proceedings team of students and instructors formatted and edited each submission according to Information Research author instructions and checked all references for their styles and validity. This took another three to four weeks, because in some cases it required additional communication with the authors. All authors received a copy of our final file. The outcome of this team effort were 37 long papers and 13 short papers as doc-files, formatted and edited according to Information Research standards. The time-frame to achieve this was a whole semester (about 18 weeks). Each student paper editor worked on average 10-15 hours per submission, including communications with the authors. By July 2022, all short and long papers were copy edited and formatted according to the guidelines. They were not yet converted to HTML, which is a requirement for publication in Information Research.
Publication of the proceedings in Information Research was envisioned for autumn or winter 2022, that is after the conference took place. The proceedings of the previous conference, ISIC 2020 were published in two issues in December 2020 and in March 2021, nearly eight months after the conference. Since all papers were ready for publication in July, we, as local team, decided to go for a different approach than previous ISIC conferences (and different than most conferences): all papers were published as preprints on the Humboldt university’s institutional digital repository edoc server prior to the conference. What do preprints mean in terms of tasks?
First, we created a distinct ISIC collection on the Humboldt repository with a landing page leading to all short and full papers, but also to selected posters and panel publications. In a second step, all papers were converted to PDF files.
Secondly, all authors were contacted and asked to sign a copyright form so that we were able to publish articles on their behalf.
Last, since the edoc server assigns digital object identifiers (DOIs) as soon as an item is created and the metadata record is populated, it was also possible to include DOIs and licensing information in the published PDFs’ metadata. We used exiftool to semi-automate this process. As a result, all 50 submissions were published as open access preprints, including persistent identifiers, long-term preservation and CC-BY licensing.
Information Research, as a journal, currently works with HTML as a file format for publication. This meant for the process of publishing the proceedings that all 50 submissions had to be manually converted to HTML. This process took about three weeks and was very labour-intensive. It also requires different skills than what is required in terms of proof-reading and editing. Low-budget conferences might consider hiring someone for at least one of those two activities: (1) editing and (2) (HTML) conversion.
We experimented with automatic tools to reduce the manual, time-consuming copy and paste work from Word to HTML, but unfortunately, automating this process did not work out. For example, Pandoc, a program for converting file formats, was not feasible in this case, since the Information Research HTML template was too individually designed and there was no time to write and test a custom script for this conference. In the future, a custom Pandoc script for DOC to HTML conversion could be of great use to all authors submitting to Information Research, though.
Post-conference publication vs conference supplementary material
For ISIC 2022, we manually included a link to each preprint into the schedule of the conference program. This approach of preprints and inclusion of links in conftool turned out to entail two advantages:
First, from an author’s perspective it was possible to promote conference attendance and individual talks by referencing and pointing to the finished submission. In addition, during and after presentations, authors were able to refer to their papers and discuss points made in more detail based on the fact that it was possible for the audience to read the final submission.
Second, from the conference participants’ perspective, the proceedings preprints served as a supplement material to the presentations given, enabling deeper discussions and more targeted questions to the presenters. They also aided a better-informed decision-making which sessions to attend.
When this editorial was written in October 2022, the preprint articles had been accessed 1600 times. Already after the first day of the conference, the preprints had been accessed 900 times. We view this as confirmation of our vision that proceedings as supplementary material have as much merit than as post-conference publications.
Overall, it should not be underestimated how much time and effort it takes to publish conference proceedings, especially for a low-budget conference. For upcoming conferences, proceedings chairs and conference organizers would be best advised to plan for an adequate amount of time for reviews, checking revised version, checking style and formatting of papers, conversion to desired formats and good metadata assignment. We recommend at least four to five months.
Another important lesson is that it does take a team to build a village. The ISIC 2022 conference proceedings would not have been published on time without the effort from our great team of students. The students who served as paper editors for the proceedings were: Helene Hellmich, Samantha Tirtohusodo, Helen Bubiger, Sophie Kobialka, Celina Karlsson, Nina König, Benedikt Knaup, Sebastian Berg, Kristina Langrock, Laura Somschor and Anastasiia Iarkaeva.
We strongly advise against a single person responsible for the whole process, especially if proceedings are planned to be ready in time for the conference. Because of the procedural character of the publishing stages and the time-sensitive planning that is required, it is important for several people to be involved and knowledgeable about each step, in case the proceedings chair has to take time off e.g. because of sickness.
We believe that our approach to publish all submissions as preprints prior to the conference was a very good decision, despite the considerable extra work it required. Conference proceedings traditionally work like ‘special issues’ to a journal, with the same amount of time from submission to publication. It is, however, beneficial to conferences in general to think about proceedings in a different way: as supplementary material, published ‘just in time’.
About the author
Laura Rothfritz is a lecturer and PhD candidate at the department of Information Management at the Berlin School of Library and Information Science at Humboldt University Berlin. She acted as proceedings chair for ISIC 2022. Her main research interests are in the area of research data management, infrastructure studies and open science with a special focus on socio-technological perspectives on the relationship between data infrastructures and societies, and a string relation to STS (Science and Technology Studies). She can be contacted at firstname.lastname@example.org
Elke Greifeneder is Professor of information behaviour at the Berlin School of Library and Information Science at Humboldt-Universität zu Berlin, Unter den Linden 6, 10099 Berlin, Germany. Previously she was an assistant professor at the University of Copenhagen. Her research interests are standardization of methods, digital methods, natural environments, and human information interactions in a digitalized world. She can be contact at email@example.com