OpenRetractions

[blahah]

We can make a nice sciencefair enrichment datasource from the https://github.com/fathomlabs/open-retractions data.

This will allow each paper to be linked with updates (corrections, expressions of concern, retractions and withdrawals for example).

Each day when the update scripts run, in addition to pushing any new data to github, they can also be pushed to an sciencefair feed.

As results are streamed in sciencefair, they will be looked up by ID on the fly in the OR dataset.

The information can then be displayed on context - in a way yet to be decided.

[todrobbins]

This is a fabulous idea that enriches the original datasources!

[blahah]

Thanks @todrobbins - in general this will be a whole category of datasources. Such a datasource can be made from any kind of data that is not a paper itself, but enriches the paper somehow.

If you think of any examples of such things, please do open new issues.

They need:

• to be able to map to a paper identifier like a DOI, PMID, PMCID, arxivID etc.
• to be able to be pulled from some existing source regularly and streamed to a p2p resource, i.e using an API or data dump
• ideally to be free of problematic licensing issues

Examples:

• pubpeer comments (see PubPeer #18)
• other post-publication peer review platforms
• contribution badges or other author information
• links to code or data repos

My hope is that other people will have more, and more interesting, ideas than me! Also that platforms will be interested in taking part in creating and managing their own datasources.

I should write this up as proper documentation.