Autogenerate DOIs (with Zenodo?) based on releases/tags

[RaoOfPhysics]

At the moment, PDFs get pushed to PeerJ. But you could use the GitHub-Zenodo integration to snapshop the whole repo and give it a DOI.

[agitter]

We started discussing auto-DOI generation in a different project (greenelab/meta-review#13). @dhimmel recommended either:

1. the DOI should refer to just the manuscript and redirect to a view of the manuscript
2. the DOI should refer to the entire git contribution history and redirect to some archive

Our current approach is to manually release versions of a manuscript and use PeerJ, bioRxiv, etc. to mint a DOI (option 1). But this is a good place to continue discussing other options if you have thoughts.

[dhimmel]

At the moment, PDFs get pushed to PeerJ.

Only partially true. The preprint posting is entirely manual. Manubot manuscripts have been manually published to PeerJ Preprint, bioRxiv, and Figshare. I don't know of any cases where someone has automated publishing to a service that mints DOIs.

[vsmalladi]

I have looked into this. I was thinking I could make a tag for each draft. My worry is that it would confuse ppl for which DOI to cite once the work is published.

[RaoOfPhysics]

FWIW, @vsmalladi, you can always use versioned DOIs: http://blog.zenodo.org/2017/05/30/doi-versioning-launched/

[vsmalladi]

@RaoOfPhysics That is true. Maybe that is the way to do it.

[castedo]

Folks interested in this feature I think will be interested in some plans I'm making. I've stared writing about it and asked some questions here: https://community.crossref.org/t/doi-registrant-using-digital-fingerprints-of-jats-files/2671 I plan to write more details about it. Feel free to ask me questions about it. There's a lot more to explain.