I think what's refreshing about Omar Souleyman is the party — it's fun. It's really alive and very urgent. And he's not above using synths, electronics, drum machines and YouTube. He's really eager to make something that's vibrant today. I always heard interesting stories that he has one man called Mahmoud Harbi who is a longtime collaborator — he writes poems for Souleyman. When they are really warmed up and going for it at a good-times party, Harbi stands next to him on stage and chain-smokes. Then he will whisper poetry in his ear that he's writing at the moment. Omar will sing it immediately in the microphone and run around the room, exciting people there. I thought it was quite exciting for a poet and an emcee to work together.

Figure 3 illustrates at what age ceased ‘indie’ journals stopped publishing. Most journals survived the first 2–5 years period, whereas the mortality rate rose in the critical 6–9 years period. After that, the number of journals ceasing dropped sharply, indicating that the surviving journals had found stability.

The development over time of active ‘indie’ OA journals before and after 2002 is shown in Figs. 1A and 1B. A journal was counted as ‘active’ in a particular year if it was still publishing articles in that year. Before 2002 the number of active journals grew very rapidly from a total of 76 journals in 1995 to 207 journals in 2002. The year 2002 was the cut-off year to be included in the studied cohort, meaning that no new journals were added to the data set after this point in time. After 2002, the number of journals in the cohort decreased steadily to the 127 that stayed active in 2014.

The average number of articles published was 31 per year with 74% publishing 0–30 articles, and 9% 60 or more. The study also contains interesting data about the workload done, revenues etc.

“The key question for OA publishing is whether it can be scaled up from a single journal publishing model with relatively few articles published per year to a comprehensive major journal with of the order of 50–100 articles annually.” They further note: “The continuation of the journal relies very heavily on the personal involvement of the editor and is as such a risk to the model. Employing staff to handle, for example, management, layout and copyediting tasks, is a cost-increasing factor that also is a threat to the model.” Both questions are still highly relevant today.

Earlier studies A number of previous studies, both snapshots and some with longitudinal elements, have shed light on different aspects of such type of journals, which for short we will call “indie” journals.

Often the enthusiasm of the founders and their personal network can carry a volunteer-based journal for a few years. But at that same time this type of journal, which lack the support of employed staff and a professional publishing organization, are threatened by many dangers. The editor may change affiliation or retire, or the support of the university hosting the journal might be withdrawn. Authors may stop sending in good manuscripts and it may become more and more difficult to find motivated reviewers. Not being included in the Web of Science, and the impact factor that follows, may in the long run limit the number of submissions severely. On the positive side of the balance the emergence of open source software for publishing (i.e., Open Journals System) and cheap or free hosting services like Latin American Scielo have facilitated the technical parts of publishing.

Most of the OA journals founded in the 1990s were of this variety, later many established subscription journals (particularly society ones) have made their digital versions freely available immediately or with a delay. This has been particularly noticeable in countries where cheap or free national or regional electronic portals have become available, like Scielo, Redalyc, and J-stage. Since around 2003 the OA market has become increasingly dominated by professionally published journals, which finance themselves by charging authors so-called article processing charges, APCs. At first such journals were being launched by open access publishers like BioMedCentral and PLOS, but in the last couple of years the major commercial and society publishers have increasingly started new OA journals and have also converted some subscription journals to APC-financed models.

Open Access (OA) is nowadays increasingly being used as a business model for the publishing of scholarly peer reviewed journals, both by specialized OA publishing companies and major, predominantly subscription-based publishers. However, in the early days of the web OA journals were mainly founded by independent academics, who were dissatisfied with the predominant print and subscription paradigm and wanted to test the opportunities offered by the new medium. There is still an on-going debate about how OA journals should be operated, and the volunteer model used by many such ‘indie’ journals has been proposed as a viable alternative to the model adopted by big professional publishers where publishing activities are funded by authors paying expensive article processing charges (APCs). Our longitudinal quantitative study of 250 ‘indie’ OA journals founded prior to 2002, showed that 51% of these journals were still in operation in 2014 and that the median number of articles published per year had risen from 11 to 18 among the survivors. Of these surviving journals, only 8% had started collecting APCs. A more detailed qualitative case study of five such journals provided insights into how such journals have tried to ensure the continuity and longevity of operations.

A longitudinal study of independent scholar-published open access journals