T3.5 Peer reviewing the research flow A **research activity** can be intended as the sequence of actions that aims to prove an initial thesis in order to bring novelty to a research field. Every research activity is carried out in an **experimental context**, intended as a combination of tools, methodology and data, which builds upon and produces research products. Research products can be of several types: documents, such as textual narrations of the activity or sub-parts of it, more conceptual products, such as theorems, theories, ideas, processes, or more tangible ones, such as data, tools, devices, software, applications. Research activities can be evaluated only if the underlying **research flow** is adequately represented, i.e. modelled. This is because a common (agreed-upon) representation of the research flow allows other scientists (or services) to verify the quality (i.e. efficacy, novelty, repeatability, etc.) of the research activities. For example, the narrative description of a research activity in an article only support a vague/minimal level of repeatability. As a consequence, the peer review process is negatively affected and becomes less trustable/effective. Different levels of modelling of the scientific process underlying the research activity enable corresponding degrees of “transparent evaluation” [EC terminology]. For example, to achieve fully transparent evaluation research communities may agree on common digital representation of their research flows in order to enable automated evaluation of research activities via common e-infrastructure services. **Goal of Task 3.5 (Peer-reviewing the Research Flow)**: The task proposes a model for representing a research flow and new methodologies and metrics for (openly) evaluating (peer reviewing) it and its products. The model and the methodology should be applicable to both “positive” research activities, i.e. the thesis is proved to be right, and to “negative” research activities, i.e. the scientist fails to prove the initial thesis (which may not be necessarily wrong). In any case, publishing the products of the research flow will be useful to other scientists not to repeat the same mistakes. **Methodology**: One of the main characteristics of the research flow model should be to allow the repeatability of the whole process, and therefore the task will initially investigate various (existing) solutions supporting repeatability of research activities across distinct scientific disciplines (starting with those of the OpenUP use cases: arts & humanities, social sciences, energy and life sciences), in order to identify their common concepts and functions. The task will then propose a model, metrics, and tools supporting Open Peer Review of research activities (in the form of the D3.2 wiki). The survey resulting from the work of T3.2 will be used to verify possible redundancies and identify useful commonalities between open peer review of literature/research data and peer review of research activities. **Outcome (D3.2 Wiki):** preliminary outline *Section 1. Description of use-cases*: describing repeatable research flows across different research infrastructures. Identification of common concepts and functionalities. *Section 2. Survey of existing research flows*: models supporting repeatability and peer review of the research flow: Survey of models proposed in the literature for describing the research flow and enable repeatability Survey of models for peer review of the research flow *Section 3. A model for evaluating the research flow* - How to model the research flow for repeatability: this model could be an existing proposal or an extension of it; - How to evaluate the research flow: framework for the evaluation of the research flow, inclusive of metrics, methodologies, and tools. *Section 4. Validation of the model*: Based on the model, describe/evaluate the repeatable Research Flows of the OpenUp use-cases.