Open research

MishMash works in the open by default. Here, we describe what the centre has undertaken to do with its publications, data and code.

Why “open research” and not “open science”

The centre uses the term open research rather than open science on purpose. Artistic research produces exhibitions, performances, scores, instruments and documentation, not only articles and datasets. A term that covers only science would leave out half of what MishMash makes, and with it the people who make it.

Publications

MishMash results are generally published in open access channels, often with early versions deposited in preprint archives so that the work is readable before a journal’s process is finished. Given the breadth of the centre, outputs range from monographs and book chapters to journal articles and conference papers.

We highly encourage co-authorship across disciplines and institutions. The level of co-publishing can be seen as a proxy for how well we are able to engage in interdisciplinary and interinstitutional dialogue.

Data

Research data should follow the FAIR principles, which means that they should be findable, accessible, interoperable, and reusable. Datasets are deposited in trusted repositories with a persistent identifier, such as DataverseNO or Zenodo, and registered in NVA, which is where the results page is generated from. Artistic material belongs in repositories dedicated to creative practise, such as Freesound and the Research Catalogue. Material from libraries and museums connects to the European cultural heritage infrastructure.

Data are released in standard, open file formats, with the documentation and software needed to read them, and with a data management plan. The rule is that data should be shared “as open as possible and as closed as necessary”, following the dataset’s privacy and copyright limitations. MishMash also actively engages in dialogue with relevant partners about how this can be done in practice.

Where the centre can choose, data carry a permissive licence, normally CC BY. The site’s own data are published at mishmash.no/data/.

Code

Software and hardware developed in MishMash are released as open source, on public platforms, and under permissive licences. Licensing is a live discussion in the network, since the interests of a freelancer, an institution and a company are not the same.

The website itself is an example of our open approach: the website code is MIT-licensed and available on GitHub, while the content (text, images, etc.) is CC BY, as set out in the terms of use.

Why it matters

Open research is a requirement from our main funder (Research Council of Norway). However, in MishMash, we see that open research is not only meeting a funder’s requirement. Permissive licences make it possible to put tools in the hands of people and institutions that cannot buy them, in Norway and far beyond. This is part of what the centre means by working responsibly with AI.