Citable record? about pydl HOT 8 CLOSED

Is there some reason not to use ASCL?

from pydl.

My understanding is that ASCL is a registry of software, but does not produce true DOI's -- I think it's more of a database of software. But I could be wrong about that.

from pydl.

Ah, OK. I took a quick glance, & it was not obvious that ASCL provides a DOI. I'll look into zenodo. My concern though is that ASCL has been around for years, but I've never heard of zenodo. Is that really the choice for long-term curation of citations?

from pydl.

@adrn, I now have a zenodo account, but I have no idea whatsoever what to do now. Please provide instructions on how to proceed.

from pydl.

Alternatively, does the term "citeable record" imply a DOI by definition, or is it possible to have a "citeable record" that does not have a DOI?

from pydl.

I've bit the bullet & am prepared to create a Zenodo record for this, but it appears that publication only happens on a new release. I need to publish an existing release. Please help!

from pydl.

I think you can download a zip of the last release of Pydl (from GitHub, e.g. https://github.com/weaverba137/pydl/archive/0.6.0.zip) and upload this to Zenodo manually. You have to enter some info manually but it should be possible to create an entry from an existing release. Let me know if you get stuck - I've done this a few times.

from pydl.

OK, that worked: https://doi.org/10.5281/zenodo.1095150. I'll submit a PR on the Astropy 2 paper to add the citation.

PS, for the record, this guide was helpful: https://guides.github.com/activities/citable-code/

I'd still like to discuss whether the definition of "citation" requires a DOI by necessity, but that's separate from this issue.

from pydl.