“Over the last decade or so, there has been a steady transition in scholarly publishing away from a traditional subscription based revenue model for publishers towards open access models where published articles are freely available to readers1 . During the early part of the transition, author-pays models, where a researcher finds money to pay article processing charges (APCs), were shown to be sustainable under certain conditions by publishers like BioMed Central and PLOS and grew in popularity among commercial publishers2 . In more recent years, concerns about rising APCs and lack of access to publication funds in many disciplines, coupled with funder mandates3 aimed at accelerating transitions to openness, have led to a number of new business models, from so-called ‘diamond’ open access4 where publication costs are covered by a third-party fund, to transformative agreements, such as ‘read and publish’ aimed at enabling journals to move from subscription to open access models with institutional support5 . Alongside all of these sits ‘green OA’, in which authors self-archive a version of their article in a suitable disciplinary or institutional repository while the published version appears in a subscription-based journal. This increasingly complex landscape poses a problem for universities as they find themselves administering a diverse range of open access agreements. At the same time, very little research has been done into how universities deal with open access. Anecdotally, approaches to OA funding are varied and sometimes ad hoc. In general, it appears that university libraries often distribute information to researchers and scholars about sources of open access funding, but no clear picture exists of how funds are allocated or monitored. With this in mind, in late 2021, we launched a community survey, supported by MoreBrains Cooperative6 , about the current state of the open access landscape7 . With 64 responses from 22 countries, although this is a relatively small sample, several themes emerge strongly, some of which we had already intuited, and some that were more surprising….

Half (32) of all respondents reported low levels of trust in the management of OA publishing and associated charges compared to 39% (25) who reported that they neither trusted nor distrusted the status quo and just 11% (7) who reported moderate or high levels of trust. There was strong support for open APC data (43), open standards for data exchange (41) and clear institutional ownership of data (42), with about 65% of all respondents claiming that each of those measures would increase trust. Seven of the eight free text responses also mentioned transparency and improved reporting as being desirable. Although no single measure emerged as a clear first priority, these ideas share a common theme of greater coordination and coherence across the many stakeholders involved in OA. In a similar vein, community governance structures for OA data were favoured by over half (34) of all respondents….”