Issue Proposals from Mentors

[TheCedarPrince]

@asinghvi17 and I were chatting today from the mentor side on what could mentors potentially be asking potential mentees for help with. In particular, we were thinking that issues that mentors could write and be willing to mentor would more or less be their own self-contained issue proposal on GitHub. This has a few benefits:

1. We could create a bot that would check a given issue proposal to make sure it complies with what we expect of mentors to be writing for would be mentees
2. Have clear expectations for what mentees should do
3. All details a potential mentee would need is located centrally to that issue

A few prototypes have been started here for what this could potentially look like:

• Create Functionality to Download Datasets JuliaGeo/TigerLine.jl#3
• URL Helper Function JuliaGeo/TigerLine.jl#4

As you notice there are specific sections, checkboxes, and things of this nature. What else might be needed for an issue proposal that would fit a mentor/mentee matching program?

~ tcp 🌳

[asinghvi17]

IMO, the goal here should be to minimize the barrier to entry. What is obvious to a seasoned Julia developer (use HTTP.jl to query an API, Downloads.download to download files, what packages exist elsewhere in the ecosystem) is a complete unknown to many new contributors (though not all).

This is pretty much writing a comprehensive specification (as in software engineering), which can clear things up quite a bit in terms of what is desired from the issue.

As a sidenote - perhaps assigning packages/ecosystems some fixed number of slots on a Julia-community-wide board to advertise these issues could help in some ways:

1. Discoverability is a lot easier with a central board.
2. Package developers are incentivised to put a lot of effort into a few issues/specifications as opposed to very little effort in many issues. GeometryOps.jl is a good example of this - I routinely put up one or two line issues whenever I think of a cool new feature, but rarely flesh them out (primarily because it's assumed that all the contributors are sufficiently familiar with the idea, or at least will know what to ask). Having only 5 slots on this board per package would encourage me to create 5 detailed issues, and I could update the rest of them as the initial issues are completed.

[miguelraz]

@TheCedarPrince Perhaps it would be good to collate some of the JuliaCon2024 Diversity dinner notes/conclusions here.