2025-10-23 Frontend Working Group Meeting Notes: Catalog MFE, Admin Console MFE, New Visibility Option for Graded Sections

2025-10-23 Frontend Working Group Meeting Notes: Catalog MFE, Admin Console MFE, New Visibility Option for Graded Sections

 Date, time, location

 Discussion topic(s)

One, maybe two topics to be presented or discussed in depth in the upcoming meeting.

🎥Recording

 Participants

  • Adolfo Brandes

  • Brian Smith

  • Jacobo Dominguez

  • Jesse Stewart (WGU)

🤖 Summary

Here’s a detailed summary of the Frontend Working Group Meeting (October 23, 2025, 11:00 EDT):


🗣️ General Discussion: React vs. Backend Technologies

The meeting began with an informal chat about frontend vs. backend development.

  • Jesse mentioned learning Django, and Adolfo commented that backend frameworks often feel more stable than frontend tools.

  • They discussed how React evolves rapidly, often rendering past patterns obsolete (e.g., class vs. function components, Redux, hooks, sagas).

  • Brian humorously added that the rapid change in frontend work “makes you go gray faster.”

This section was lighthearted, establishing a shared sentiment about the challenges of keeping up with frontend changes.


🧩 Main Technical Topic: Catalog Microfrontend (MFE)

Brian brought up an agenda item regarding how to integrate the catalog microfrontend (MFE) with Tutor MFE (a microfrontend management tool for Open edX).
He relayed a message from Peter, asking whether the catalog MFE should:

  1. Be a Tutor plugin, disabled by default, or

  2. Be included directly in Tutor MFE, with a feature flag (ENABLE_CATALOG_MICROFRONTEND) turned off by default.

🧠 Discussion Points

  • Adolfo clarified the implications of both approaches:

    • Plugin approach: the catalog MFE would be a separate plugin that, when enabled, automatically sets the required feature flags to true.

    • Built-in approach: include it in Tutor MFE but keep the feature flags off by default, flipping them later when ready to go live.

  • Consensus leaned toward the plugin model, as it’s:

    • Easier to manage and enable selectively.

    • Less political (avoiding early inclusion in Tutor MFE before it's production-ready).

    • Already supported by an existing testing plugin that could be formalized and documented.

Decision: Move forward with the plugin approach for now, ensuring it can be easily toggled and documented.


⚙️ Secondary Topic: Admin Console Microfrontend (RBAC)

The group then shifted focus to the Admin Console MFE, which introduces Role-Based Access Control (RBAC).

  • Adolfo explained that it’s a new feature, not a replacement for existing pages.

  • It allows assigning user permissions for library management, forming the first iteration of broader RBAC improvements.

  • Brian shared his local dev environment and demoed functionality:

    • Adding users and assigning roles (e.g., Author, Admin).

    • Sorting and filtering roles.

    • Some PRs (e.g., delete functionality) hadn’t yet merged.

  • Jacobo confirmed that a sandbox environment was available for full testing since he couldn’t run it locally.

🧩 Key Notes

  • RBAC aims to extend beyond libraries in future iterations.

  • This is part of a long-term plan to replace the limited existing roles (e.g., superuser, course author).

  • Expected to be enabled by default once merged since it’s entirely new functionality.

  • Focus remains on getting it merged before the “cut” (release freeze) scheduled for Tuesday.


🧪 Demo: MIT’s “Never Show Correct Answers” Feature

Adolfo demonstrated a new MIT-developed feature addressing academic integrity in re-used courses.

📚 Problem

  • Instructors who reuse the same course yearly risk students sharing answers since correct responses are revealed after due dates.

💡 New Setting

Adds a fourth visibility option for graded assignments:

  1. Always show results

  2. Never show results

  3. Show after due date

  4. Never show correct answers but include in total gradenew option

This allows:

  • Grades to count toward the total,

  • But without ever revealing which specific answers were correct.

Adolfo tested the feature live, verifying grading behavior and identifying a minor issue where one lab displayed grades incorrectly (“real-time review failure”), which he planned to report.

 Action items

 Decisions