Reasons to have an institutional registry

README.md

Why would an institution want to deploy a registry? On a high level, adding a piece of software to an institution registy would be tailored to the needs or goals of the institution. E.g., it could be that there is additional support given or curation (hence the incentive for the researcher) or even automation. It could allow for any of the following:

1. the software will be proudly showcased as part of the institutions set
2. It can be a place to link software to funding / grants, e.g., someone can visit the registry to understand where tax dollars are coming from
3. it can be a place to summarize / understand what an institution is working on (or where there are gaps)
4. for institutions that want to give back to open source (e.g., companies do this, Indeed is a good example) they have a summary of what is most important and can direct investment!
5. an API provides data for those interested in combining across registries or further study.
6. could include a review aspect, e.g., could be reviewed by someone at institution
7. could be hooked up with a bot or similar to share on social media
8. some number of hours given / year to help projects in the database - "submit your software here in this form and we will help"
9. place for new student to find lab/what they want to work on - "I'm a new student in astronomy what are people working on?"
10. "find similar software" - and introduce opportunity for collaboration - is there a way to work together? Learn from one another?
11. "submit your software and get help to... package it? Integrate into workflow managers? containerize?"
12. submit your software and - every year we find similar projects, have an annual meeting and put people working on similar things in the same room. Create "research software alliances." that cross registries.
13. look at people using common dependencies, get feedback about what could be better to the project
14. it could be a registry provides a linked code browser (e.g., akin to https://cs.opensource.google)
15. makes it easy to cite, e.g., go to the search.select some number, and click "cite" to generate citations
16. maybe it builds software into environment easy to try on HPC system?.
17. could there be a direct way for someone to "try it out" on an institutional resource?
18. could this registry be a temple for institutions to share / easily install other institution software?

While a lot of these ideas could be cool, none of them (until the last three) got me excited about (as a developer) submitting my software to such a registry. I got excited about points 16-18, and I'll follow up in my email! Note that many of these bullets are trying to deviate away from standard publication and ask "What do we want to do instead?" While papers might be appropriate for some research software, they are not for all.