Recently, research infrastructures (RIs) are recognized to take on a crucial role in the national innovation system in driving open innovation via shared use that include offering measurement service to external users. Generally, these services are performed by staff in line with the social division of labor as scientists and technicians in research institutes. However, services rarely lead to research collaborations because research collaborations generally do not stem from services per se and require researchers to engage more with partners in initiating collaboration. This article focuses on its boundary of services and research collaborations by conducting an ethnographic survey of four years since 2016 to analyze epistemic cultures in eight Japanese state-of-the-art RIs. It first examines what is occurring at the boundaries and offers two frameworks, i.e. the subject-driven approach and the user-driven approach, for finding potential solutions to the issues that impedes service from extending into research collaboration.

中文翻译：

---

在日本最先进的学术研究基础设施中的服务和研究合作的边界

最近，研究基础设施 (RI) 被认为在国家创新体系中通过共享使用（包括向外部用户提供测量服务）推动开放式创新方面发挥着至关重要的作用。这些服务一般由科研院所的科技人员按照社会分工进行。然而，服务很少导致研究合作，因为研究合作通常不源于服务本身，并且要求研究人员在发起合作时与合作伙伴进行更多的接触。本文通过对自 2016 年以来四年的民族志调查，分析八个日本最先进的 RI 的认知文化，重点关注其服务和研究合作的边界。