Improve citation management

[shaunagm]

There's a set of related questions here:

1. When someone "cites HARK", how should we encourage them to do so?
2. When someone "cites HARK", how do we discover that they've done so?
3. How should we refer to and display HARK-derivative items, especially REMARKs? Should REMARKs have DOIs?

1: Chris suggested in an email, and I agree, that we should set up something like AstroPy's acknowledging.html page as our solution to question 1. Chris thinks we should use the SciPy proceedings paper as the reference, which is fine by me. (Currently, the closest thing we have to this is an this autogenerated citation in Zenodo, which I dislike relying on since it will change with every major release of HARK.)

2: Right now we do Google Scholar searches for "econ-ark" and the two DOIs on Zenodo. If we settle on the SciPy paper, for instance, we can search for that instead of the Zenodo DOIs, although we probably still want to keep an eye out on "econ-ark". Google Scholar is someone limited in the references it finds so it would be nice to be able to access Web of Science as well.

3: This overlaps with an existing discussion we've been having about how REMARKs work - how we should store/tag/display them, what the process for adding a new REMARK should be. This issue has a proposed process for adding REMARKs, and there's an email thread about how best to display REMARK metadata (in Zotero, on the website itself via a content management system, etc.)

Chris mentioned having a few REMARKs ready to be added soon. Hopefully we can come up with an initial process soon and try it out with the new REMARKs.

[shaunagm]

Note: see discussion of CFF in this issue.

[shaunagm]

Moving over Chris's comment from an issue in the HARK repo:

The last presentation at the PASC19 session on management of open source projects was by a person from the The Research Software Directory in the Netherlands. They have a blog

The software is available at GitHub

He illustrated a technology for integrating Zenodo, Zotero, and GitHub.

In particular, whenever there is a new release, references to the software are updated in Zenodo and Zotero.

• GitHub-Zenodo integration to mint DOIs whenever there is a release.
• Research Software Directory has an Admin interface with concept doi
• Nightly scripts run to update references.

Again advocated using the cff file format: Google search looks for this and makes it easier to find your resources.

This is integrated with a "cite this software" web page.

[llorracc]

This is related to Econ-ARK/HARK issue 339

[llorracc]

I think we still have not worked out the proper cff file format for citing Econ-ARK. @sbenthall, please take a look at the links in this OverARK issue and those referenced therein. That should be in plaintext (and a "copy" button") on the "Acknowledging" page.

[llorracc]

@DrDrij, in the "acknowledgments" section of my BufferStockTheory paper (which is about to go to the journal), I link to the "https://econ-ark.org/acknowledging/" page; but I would also like to be able to have a direct link "\href{[what is the proper url]}{(Click for bibtex reference)}" that takes either takes them straight to the corresponding button. (Also, please expose to view the exact text that is being copied.)

[sbenthall]

I have a question.

It looks like the recommended use of CFF is to put the file in "the root of your code repository".

In this issue, it looks like it's imagined that a separate CFF would go into subdirectories of a repository, such as individual REMARKs.
#37

It's unclear to me whether this is standard use of CFFs and if software tools built around the CFF format would recognize this practice. Has this been thought through by others here yet?

[DrDrij]

@DrDrij, in the "acknowledgments" section of my BufferStockTheory paper (which is about to go to the journal), I link to the "https://econ-ark.org/acknowledging/" page; but I would also like to be able to have a direct link "\href{[what is the proper url]}{(Click for bibtex reference)}" that takes either takes them straight to the corresponding button. (Also, please expose to view the exact text that is being copied.)

@llorracc Re-reading your comment. Is the Bibtex unique for the paper? If so, where would be ideal to host the file? (The copy bibtex to clipboard button is a little complicated. It uses a small html hack and javascript. So I wouldn't recommend embedding it elsewhere).

It's unclear to me whether this is standard use of CFFs and if software tools built around the CFF format would recognize this practice. Has this been thought through by others here yet?

@sbenthall I don't believe its been tested out yet.