Concept Note: A Modular Open edX Experience

This is a vision-forward concept doc to outline the value, scope and potential market fit for modularity. It is a draft and intended to stimulate discussion, and to inform a use case-driven narrative around current discovery work around course modularity, atomic learning units, and a lightweight LMS/learning core.

Beyond MOOCs: A Modularized Open edX Experience

Concept Doc

Abstract and Goal:

Open edX was built to deliver full, end-to-end courses. The course-centric model is deeply embedded in every level of the platform and platform architecture, such that the underlying structure results in difficult barriers to creating alternative, more nimble models of content delivery. Furthermore, the current platform core has grown into a gargantuan monolith that creates myriad challenges around extensibility and performance for developers and course teams who need agile customization options even within the end-to-end course experience. This Initiative proposes to create a new, radically simplified, modularized and scalable system that enables diversified experiences in building courses, learning experiences and course pathways. In addition to creating end-to-end courses from the top down, it would offer an alternative option for users to create, design, build, sequence and deliver learning experiences from the ground up.

Value and Impact:

Value to educators, faculty, SMEs, course designers, course authors:

  • Improves authoring by enabling a pedagogy-first pathway through the course design experience. Educators are freed to focus on designing learning objectives first, and building and sequencing learning pathways to achieve them, from the ground up, rather than molding learning pathways into a prescriptive, top-down course→module—> unit structure.

  • Empowers educators/course teams to define the boundaries of learning experiences and align them to more particular and diverse learner needs and outcomes.

  • Creates opportunities for educators to integrate adaptive learning strategies and tools to improve learner outcomes.

  • Creates platform flexibility required to explore alternative models of content sharing and content reuse.

Value to developers:

  • Enables faster development cycles by removing dependencies inherent in the edx-platform monolith

  • Provides an easier, more reliable path for extensions developers as extension authors should be able to build and test against a smaller repo than the edx-platform

  • Drive performance enhancements

Guiding Use Cases:

While the majority of current use cases for Open edX revolve around full course delivery, there is a defined and growing interest in alternative, more nimble models of content delivery. For example, use cases to design and deliver more flexible learning experiences include:

  • Delivering a single module as a complete learning experience, or a unit within a module as a complete learning experience.

  • Designing a learning objective around a single video or webinar, with associated comprehension questions, and delivering as a complete learning experience. 

  • Delivering customized sets of questions, assessments or question banks as complete learning experiences. 

  • Designing a learning pathway that sequences together a single video, a skills assessment, and a chapter from a text as a complete learning experience.

The underlying user assumption beneath these use cases is that the learning experiences could be built without many of the dependencies that come with installing and running a full Open edX Instance.

Relatedly, there are defined and growing use cases that require more flexibility when building an end-to-end course. For example

  • Delivering a full course with the flexibility to “mix and match” specific modules or units for different audiences without having to build separate courses for each “mix and match” iteration. Relatedly, the capacity to integrate adaptive learning engines or tools into the course to guide “mix and match” pathways.

  • Point-and-click options for integrating third-party LTI components.

Market/Ecosystem “Fit”:

The above use cases illustrate a fundamentally different approach to designing learning experiences. In the current course-centric model, course authors start from the top of the course and work down, from the overall course objective to designing and building modules, units and subunits that fit the prescribed structure. The above use cases flip that process on its head, and enables authors to start with a singular learning objective, and then build and sequence the right combination of contextual materials into a learning pathway to fulfill the learning objective. The final result may be a full course, or it may be any number of sub-course-sized learning experiences.

Currently, there is no platform or tool in the edtech space that fully and easily meets this need:

  • Most mainstream LMS systems like Canvas, Blackboard, Sakai, are stuck in rigid course-centric models, and others that have some potential for customization to meet this need, such as Moodle, require extreme configuration. 

  • Mainstream OER content libraries like Merlot, Openstax, OER Commons have extensive, discoverable, curriculum-aligned libraries of teaching content, but lack the platform and infrastructure to sequence them into customized, meaningful and measure-able learning pathways.

  • Mainstream adaptive learning platforms provide data-driven infrastructure to enable integration of adaptive learning in the classroom, but without the platform and content to enable better learning outcomes at scale in massive asynchronous environments.

The platform could be designed to fit in the center of these spheres and fill the niche of content delivery and sequencing that they don’t, while simultaneously fitting into each sphere with the right extensibility strategy.

Scope and Approach:

Refactoring Open edX for modularity impacts all aspects of the platform, most significantly architecturally and from a course authoring and user experience/UX perspective.

Platform and Architecture - The Learning Core

There is an Initiative to design and create a learning core platform, by extracting core learning concepts and data models into a new openedx-learning repository. The new core platform would be the necessary infrastructure that could support an option for users to create, design, build, sequence and deliver learning experiences from a “ground up” approach, rather than a “top down” approach. 

Extensive discovery work to define an implementation strategy for refactoring the platform architecture to create a learning core that can support the above use cases is here: https://github.com/openedx/platform-roadmap/issues/67

Next steps:

  1. Design and execute a proof-of-concept project to illustrate the approach and outcome of the above implementation strategy. We discussed defining a raw data model to demonstrate how the learning core would function, particularly around the use case of arranging content into sequences. This work would serve a dual benefit of enhancing the current content library work.

  2. Define contingencies, such as: the Initiative would be architected and implemented in such a way as to preserve/not disrupt the current course authoring experience.

  3. Discovery work around open questions, such as:

    1. How would users who currently run Instances navigate and/or take advantage of the benefits of the new learning core with Instances running on the current core?

    2. What are the boundaries of the modularity/core? What elements of the current platform core and included, which are not, and what is the underlying rationale. 

  4. An investigation into trade-offs and alternative implementation approaches

Course Authoring - A Lightweight and Simplified Studio

In order to explore the goals of a user-friendly authoring experience to drive adoption, we should run discovery work in parallel to investigate what the user experience could and should look like on top of the Learning Core.

Next steps:

  1. Design and execute a proof of concept project such as sketching a visual mock-up of the user journey through a “bottom-up” learning design experience. It would answer the question: What does the authoring environment look and feel like to build, sequence and deliver a unit as a standalone learning experience via the new learning core platform? This work could serve the dual benefit of informing work to improve the current course authoring experience in Studio today, which contains a number of barriers and challenges.

  2. Build feedback loops with the community and/or with users on the design documentation, early and often.