Write up some outreachy reflections?

[choldgraf]

A year ago we participated in our first (and only) round of Outreachy. I think that it went really well, as we were able to welcome two new members of our community (👋 @GeorgianaElena and @leportella).

If we'd like to repeat the process and to make ourselves more successful at doing so in the future, it'd be helpful if we wrote down our experiences, lessons learned, etc when participating in mentorship programs. It may also help provide others in the Jupyter community with patterns that they could try to follow.

So, I wonder if as a start we could codify what were the most important things that we learned from Outreachy. What went well, what could be improved, etc. Maybe @yuvipanda or @minrk have thoughts that they could either copy/paste or write down from somewhere? Maybe @leportella or @GeorgianaElena have recommendations or tips they'd give to future interns or future mentors? Then we could add them to the team documentation in this repository.

Some links I can think of:

• Intro post about jupyter's involvement in outreachy
• Georgiana's Outreachy blog post
• Leticia's Outreachy blog post

[minrk]

I think it went super well, and plan to mentor this coming winter when I have time again. My main thoughts are:

1. this is a super valuable way to spend our time, if our first experience is any indication. I think we should try to build a habit of participating in Outreachy at least every year, and possibly every session (twice a year), if we can.
2. it does take quite a bit of time, and we need to plan for that
3. the application process requires a lot of work (more than the internship itself, I think), both before and during:
   • we need to prepare repos with strong CONTRIBUTING docs for novice contributors (this has improved a lot since our first go, great job everyone! ❤️ )
   • we need to have well defined internship projects that are
     • reasonably scoped for 12 weeks of an intern
     • stretch goals are a great way to allow for variable productivity
     • valuable to get done so people care, but not a blocker for our full-time folks to be frustrated by waiting for.
   • we need lots of "micro tasks" for reviewing applicants. This is the hardest part - lots of discrete PRs that are easy enough to do for newcomers, but non-critical enough that we don't need to hurry them along, since several applicants will start work and move on without finishing. Adding prometheus metrics was a good source of these, but I'm not sure we have enough similar tasks. This is where we need the most help
   • reviewing / helping applicants on micro tasks is a lot of work as well during the application period.

Implementing whole new features like traefik proxy and native auth were good fits because they met real needs, while also not conflicting with ongoing work. A nice green field for interns to work in, though I imagine this can also possibly feel isolating if the work doesn't get integrated promptly.

If we have early tasks to start thinking about as a team, I think it would be:

• coming up with future internship project ideas
• coming up with sources of lots of micro tasks for applicants
• planning for mentor time and funding

[GeorgianaElena]

In addition to the links Chris added, I want to point out that @leportella has five amazing and inspiring blog posts about the Outreachy internship on her blog: report 1, report 2, report 3, report 4, report 5 :heart: You can find my impressions about the internship (which are also my first attempts of writing blog posts) here.

I must say that I loved everything about the internship and it was the best one I had. I especially liked and that:

• there was a separate repository for the application phase and this made it very clear right from the start what you had to do
• the micro-tasks were so nicely and clearly written
• every task ended with an encouragement to ask questions or support and this helped create a safe place where I felt less afraid to ask for help
• @minrk and @yuvipanda were really patient and lightning fast in answering questions :zap:
• other members of the community were sometimes reviewing the PRs and congratulated us for a merge. This made me feel that the contribution mattered and that even if I wouldn't have been accepted, I would have at least been helpful.
• there was a great road map of the project defined through a set of issues, opened by @minrk (:heart:) which provided great instructions and helped a lot with the organization, sense of progress and timeline.

I wouldn't change anything from the Outreachy experience. And I realize now that the things I loved the most were the ones that @minrk listed as the ones that take the most time and effort. So, thank you again for everything that everybody did to make the internship that awesome. You are amazing :heart: :sparkles:

[meeseeksmachine]

This issue has been mentioned on Jupyter Community Forum. There might be relevant details there:

https://discourse.jupyter.org/t/jupyterhubs-experience-with-outreachy-lessons-learned/4803/1

[choldgraf]

Hey all - I've tried to distill the points above into a structured post that I've just added to the community forum here:

https://discourse.jupyter.org/t/jupyterhubs-experience-with-outreachy-lessons-learned/4803

It's marked as a "wiki", so anybody should feel free to edit it, or comment and suggest any changes to be made. I've also set the author to "Jovyan" since these aren't my ideas, so much as the community's. Thanks everybody for your help in collecting this experience!

If someone thinks there's a better place to put this content, let me know and I'm happy to do so. It seemed like the community forum is a decent place to store it in a way that's easily discoverable / linkable / etc. I think we could also turn it into a blog post if we think it could be useful to others outside the community. Let me know what you all think!