00-better-science.Rmd

# 'a Science Collective' for Better Science

We believe that science can make faster progress, use resources more
efficiently, and can be a more rewarding experience for scientists if they
could follow their curiosity in a constructive, collaborative and efficient
research environment which is kept as free as possible from legacy traditional
research structures.

We believe that people with scientific interest should be stimulated and
supported to pursue their curiosity, also as a part-time effort beside other
occupations or as a full-time scientist working independently from the existing
institutional model. This would leverage the enormous potential talent that is
available, as is already done in open source software communities.

We want to achieve this by creating a new type of autonomous, decentralised
research organisation that is committed to the highest standards of the
scientific method, helps scientists thrive as part of a global community,
stimulates new connections, fosters open research collaborations and provides
key infrastructure (decentral governance, finance, computational and IT
resources). This organisation will also strive to include non-traditional
scientists: people who are eager to learn, participate in and contribute to
science but have not followed, or have left a career in academia or commercial
research.

## Values for Better Science

- Scientific output should be diverse and not limited to publications. Documented datasets, code, educational material, public outreach, and other dissemination outlets (e.g. blogs, protocols, peer-review reports) are just as valuable and necessary. We believe all these outputs should be equally acknowledged.
- Scientific output should be openly licensed and accessible. Common licenses (CC, MIT) and repositories (OSF, GitHub, Zenodo, figshare) should be used widely.
- Scientific results should be fully and independently reproducible. If data can’t be shared for privacy or legal reasons, a simulated dataset that mimics the original data should be provided with the open source analysis code. Reproducibility should be easy to carry out and should be well documented.