OLKi aims to bring scientific interactions to the fediverse, and be a platform of choice to host datasets and related assets necessary to reproduce and analyze big data algorithms, but also interact (comment, annotate, poll and more) with them. The project is just starting but has hight ambitions.
The whole platform focuses on corpora (any set of files from single preprint pdfs to multi-GB datasets, usually but not limited to scientific output) and interactions around them.
Read more about it in:
- the landing page (short information)
- the wiki F.A.Q. (various questions answered)
- the white paper (technical information)
What OLKi is not
OLKi doesn't aim to replace data repositories (for instance like http://ortolang.fr/ or http://zenodo.org/). OLKi replicates some of their features, but is generally meant to be complementary to them, and intends to play nicely with them. OLKi can also host data, but doesn't plan/cannot by itself guarantee long-term data preservation. It has to do with the software than with the fact that every instance can have different data management plans. On top of that, we want to bring activity to corpora and focus on that - not on bringing long-term preservation of the data (that's your job!).
We plan to offer solutions for the following use cases:
- corpus review, annotation and overall interactivity with the fediverse: https://olki.loria.fr/blog/usecase/
- alerts, filetering and search regarding existing and newly added corpora on OLKi and harvested platforms (use case article to be defined)