Apologies if this is a stan-users question.
I hope to have a Stan case study read to contribute this week and was wondering what the procedure is for doing that?
Do I create a fork of the example_models repo and create a pull request, or just share my repo? Not quite sure what the best way to go about it is.
This is hopefully the first of a few case studies I would hope to write for the insurance space, the later ones building on the work of the earlier models.
Thanks
This forum’s fine.
If you can create a pull request that’d be great.
What format is it? (Jupyter/knitr/R/Python, etc.?)
We’ll review it to make sure it runs and make sure
the stats look OK.
And definitely do one first before submitting pull
requests for other ones.
I’m interested in making a case study as well as companion to a paper I’m working on. I have two additional issues to OP, 1.) I am not a GitHub user, so I don’t know how that would affect the submission process. And 2.) I am a US federal employee. So while I know I have permission to disseminate the information, I am not sure how to deal with copyright issues, as despite being the author I am not the personal holder of the copyright, the US government is. I am still talking to people about this in-house, but it is new to them and me. I thought you all may have some insight into the process as the publisher of the case-studies that I could use while trying to figure out this process.
Sorry nobody answered this right away.
Are you currently allowed to publish papers and distribute open-source code? A Stan case study is just the combination of the two. All that we require is that you have the permission of the copyright holder to release the material under an open-source license.
Is there a version control system you prefer to GitHub? We’d like some way to distribute the code that goes along with the case studies. If not, we can just put everything in Stan’s example-models GitHub repo (which we’ll have the right to do if you get permission for the open-source licensing).
No worries. I needed the time to do more digging myself. I can’t say I know this all for certain yet, but what I think I have learned is that if I were to write a case study as a US federal employee I would be creating what is called a work of the United States Government. Here’s the Wikipedia page.
The code and text would be considered public domain inside the United States, but the US Government would still hold the copyright in jurisdictions outside of the United States. Is that acceptable? I am still trying to figure out what all the appropriate legalese I would be required to place on such a document.
I am still stuck in the past and do not use a version control system. If I can send you the copies of the relevant files somehow for you to host, that would be good.
I don’t know the implications of U.S. public domain. We need a license to distribute these things globally as that’s what we’ll be doing by putting them on our servers. If you want to put them on your server somewhere, we could link to them.
Version control lets you play easily with others and saves you a lot of mental anguish around backups and reversions and naming. It’s well worth the investment, especially if you have colleagues that can help if you get stuck.