Outreachy report: December 2025

• All 21 interns signed their internship agreements. 20 intern agreements were forwarded to Software Freedom Conservancy on December 04, 2025. The remaining intern agreeement was forwarded to Software Freedom Conservancy the following day.
• The deadline for mentors and interns to submit Feedback #1 was December 15, 2025. 14 interns received feedback from mentors we considered satisfactory to be sent to Software Freedom Conservancy before December 20, 2025. 6 interns received late, but more adequate feedback—they were forwarded to Software Freedom Conservancy on January 02, 2026. 1 intern’s first payment authorization remains blocked—their mentor feedback, although positive, doesn’t include details about the work performed in the first weeks of their internship. We expect to hear from their mentors this week.

Happy new year! 🎉

The theme of the month was debt in technical and policy capacities. In early December, Sage informed me they were leaving Outreachy’s organizing team. Four organizers remain: Karen Sandler, Omotola Omotayo, Tilda Udufo and I. We’re now in a transitional phase: I’m filling the gaps their farewell has left. Part of this process is the uncovering of long-standing website bugs and assessing whether the code produced corresponds to the expected business logic. It started with an investigation about why we couldn’t access feedback submissions tied to the first feedback cycle of this December cohort.

I learned that Sage Sharp wrote a script to create a new RoundPage object. RoundPage receives several important parameters, such as important cohort deadlines and the cohort’s slug. Each Active internship object will inherit RoundPage dates as their default values. Changes to the RoundPage object after the creation of Active internship objects, however, will remain isolated to that RoundPage object. Changes to a single Active internship object are also executed in isolation. The business logic behind that programmed behavior is:

• The cohort schedule establishes standard cohort dates. All internships are established with those default dates.
• We allow mentors to request extensions and, in special circumstances, late starts. The manipulation of dates tied to a single internship should be done in isolation—an extension for one intern doesn’t imply in an extension for all interns.

I learned that all cohort dates are calculated and inserted into the script by hand. They may vary according to Software Freedom Conservancy demands (e.g. tax season, fundraising) and/or community needs (e.g. full alignment with Google Summer of Code). While that has given us flexibility to create a cohort schedule, the lack of tests leaves some room for human error and tight deadlines. The difficulty in accessing feedback submissions originated from a mistake in the script: Sage Sharp accidentally entered several dates with the year of 2025 rather than the year of 2026. As several cohort dates were now in the past, our dashboard had nothing to display in the present.

Additionally, our schedule established that first stipend payments should be made by January 14, 2026. Feedback #1 submissions were due on December 15, 2025. That’s 30 days on paper, respecting the principles of NET-30 payments. However, we often have to request more detailed feedback submissions from mentors so it complies with internal payment authorization policies, and a lot of mentors take time off during the holidays. That seeemingly adequate 30-day interval between our Feedback #1 submission deadline and our first stipend payment deadline doesn’t sound that adequate anymore.

We have experimented with changing feedback submission questions to coach mentors to write more detailed reports, and we have succeeded in training some mentors to submit all relevant information. But as new mentors come in, we always have to spend time following up with mentoring communities about vague feedback submissions.

I want to establish a new process for defining cohort dates. I want to write tests that take into account the time to process feedback submissions, including following up with mentors; sensitive periods such as holidays; and the validity of dates being inserted into RoundPage object parameters, avoiding mixing years as we’ve seen in recent December cohorts.

Another new process being established is the offboarding of Outreachy organizers. For example, Outreachy organizers are in either of two groups on our Zulip server: Owner or Administrator. Organizers with Administrator privileges cannot remove organizers with Owner privileges. Additionally, organizers with Owner permissions have access to funcionalities blocked for Administrators (e.g. changing the server’s message retention policy). Some BigBlueButton meeting rooms used for weekly activities were created under individual accounts, as were some private repositories.

As for Open Mentorship Handbook news, after a much needed break from creative activities, I am now feeling more energized to write. I’m reading On Writing Well by William Zinsser to become a more effective handbook writer and editor, and some of its principles are helping me get into the habit of writing daily. We have a defined outline for the handbook, but in January, I will write with no specific sequence in mind. I will let myself piece the handbook naturally. Once we have a draft, the exercise of rewriting and editing will start, and we’ll make all connections we’ll need.