search

nnf-cbn / 2019-unconference

Organisation of the 2019 unconference
3 stars 4 forks source link

BOF: Where to store stuff #5

Open cfblaeb opened 5 years ago

cfblaeb commented 5 years ago

Where should we store stuff?

Code: Github, gitlab, bitbucket, sourceforge, other Data: figshare, dropbox, amazon, mendeley data, other managed or self-hosted?

pvtodorov commented 5 years ago

Synapse is a platform built for tracking and version controlling data. It's possible to record any steps done to the data as provenance. The other useful feature is being able to provide users/collaborators with access to only specific files. I've seen this used by projects like the Mount Sinai Brain Bank as a way to grant only approved individuals access to the data. It is also capable of running on user defined storage such as an Amazon S3 bucket or a self hosted server.

I also recently used Dryad which is great for providing a durable data and code package that's citable (via doi) and can be included into publications.

phantomas1234 commented 5 years ago

@cfblaeb you mean stuff that will be created during the event or in general? Regarding the content created during the event we were thinking about just using the repo here to link out to content wherever it might be created (google drive, other github repos, etc.). If this is more about where to store data in general, then I think this would be a great topic for a session 😃 In addition to what @pvtodorov already mentioned, let me add https://dvc.org/ here.

cfblaeb commented 5 years ago

@phantomas1234 I was thinking "in general" and as a topic for a session.

martamatos commented 5 years ago

I've recently come across fairdom, which focuses on having the data in a standard format and properly annotated. Access is also easy to define.

Midnighter commented 5 years ago

Other options for the discussion in the session:

Code & Data: https://codeocean.com/ Data: https://zenodo.org/ https://quiltdata.com/

martamatos commented 5 years ago

Since we talked quite a bit about fairsharing and fairdom, I just checked both on the re3 database, and one difference between these platforms seems to be the data access, which is open on fairsharing, while on fairdom it can be open or restricted (e.g. only the reviewers for a given paper can have access)