Evaluate switching to OSF for archiving results.

epiforecasts / covid-rt-estimates

National and subnational estimates of the time-varying reproduction number for Covid-19

https://epiforecasts.io/covid/

MIT License

34 stars 17 forks source link

Evaluate switching to OSF for archiving results. #182

Open seabbs opened 3 years ago

seabbs commented 3 years ago

As the data verse solution appears to be unreliable and there is no evidence the issues will be able to be fixed it might be sensible to move to another archiving solution. The open science framework is one option that appears to be well supported and is used by many researchers.

OSF here (project example): https://osf.io/3ed8x/

R package for uploading and downloading: https://github.com/ropensci/osfr

sbfnk commented 3 years ago

This is an excellent idea. At the same time we should review the amount of data that needs to be stored - especially, I‘d suggest not storing graphs (or if we do want graphs, at least use vector graphics - at the moment we‘re moving a lot of white pixels around the world) and avoiding redundant information.

https://help.osf.io/hc/en-us/articles/360054528874-OSF-Storage-Caps