LDT Seminar – Master's Project Milestone Assignment


Approach: Theories

There are several different “approaches” to address in your proposal.  One is your theoretical perspective on the challenge you address.

What approach to learning informs your design? Explain your theoretical framework – a mile-high view of the big ideas in your proposal about how people learn. The point is not to show that you read everything in 333A or 328, but rather to help your reader understand where you’re coming from. Explain how this approach has potential to help learners reach the learning goals.

It can be challenging to know what to include here. It might help to look at past projects. For instance:

  • Contrasting cases is a core mechanic in Feeling Talk
  • Creative confidence is the key outcome in Kibuni
  • Periscope uses augmented reality to promote theory of mind 
  • TandemArt facilitates conversations between kids and caregivers, because that’s how they learn
  • NatureQuest supports close observations of nature through narrative and family interactions.

For more examples, I strongly recommend reading some past project reports (at sdr.stanford.edu) to get a sense of how this might look for different types of projects. 

Approach: design of the solution

Another way to address “approach” is to think about how you will arrive at your final design. You probably identify more with one or the other of two scenarios:

  1. You know (pretty much) what you will be making. You might find it useful to list the key features of the proposed project, with brief explanation and rationale for each feature. You will want to explain why this is the right technology for this problem.
  2. You need to learn more (or much more!) about the learner before you can say anything about what your solution will look like.You will probably focus more on the design principles that you will use to focus your efforts and inform your design decisions. Somewhere later down the line you will need to decide what technologies to use; what can you say now about which ones will be most likely and/or how you will decide

If you expect any challenges, discuss how you might address them.



The approach to learning that informs my design is a combination of the Protege Effect, project-based learning, and TPACK. The expert, teacher, or content creator is here called the “user” insofar as it is the person who is interacting with LXD during the course creation process.

The Protege Effect will be elicited through a virtual student who will prompt the user to teach him by asking leading questions, making suggestions, and warning the user about excessive use of one style of teaching as well as the lack of content, reflection opportunities, or detailing of previous knowledge. The virtual-student closes the gap between the content ideation and the actual student’s experience. Through immediate feedback, the virtual student will elicit the user to think deeply about content choices and aid in the process of deciding the learning progression that must be in place.

The project-based learning approach simply entails that the user is engaged in a project while using LXD itself. The project is the course creation process itself, within which scaffolds are presented to the user. In addition, the approach utilizes backwards design principles embedded in the interactions the virtual student has with the user. The idea being that the heuristics and strategies invoked by the virtual student are guided by these approaches without necessarily making them explicit.

Finally, LXD aims at increasing the user’s Technological Pedagogical Content Knowledge (TPACK) by offering simple media editing tools, pedagogical scaffolds, as well as content produced by other users that can be incorporated into the course creation process. LXD in itself is a technological tool that will increase the user’s TPACK by presenting the necessary information, background knowledge, and content that supports the user’s ideation and publication strategies.


Existing solutions (“competition”):

LXD is a construct that for the purposes of this project, will build upon an existing LMS or CMS – let’s call it LCMS for simplicity. This LCMS provides a base from which to start off with. Creating an entirely new LCMS from scratch is unfeasible and not necessary in order to test LXD’s effectiveness. I am currently analyzing which platform will be the best suited for this project. Here’s a list of the ones I have shortlisted:

  • Moodle – open-source CMS
    • Positives
      • Total freedom to create
      • Mature platform with thousands of plugins
      • Large community to interact with
    • Negatives
      • Cumbersome to customize
      • Old looking base interface
      • Old HTML base – no use of HTML5 affordances
  • Coursera
    • Positives
      • Could work with existing content publishers on Stanford
    • Negatives
      • The focus is curating online courses done with any online tool (I think)
  • Udemy
    • Positives
      • Content publishing tools is one of the most user friendly I’ve seen
    • Negatives
      • Would have to negotiate with Udemy access to their platform’s source code
  • Udacity
    • Positive
      • Focused on tech courses – familiar to me
    • Negative
      • Have never seen their course publication tool
      • Would have to negotiate with Udacity access to their platform’s source code
  • EdX
    • Positive
      • Candace Thille might have contacts to get access to the company
    • Negative
      • Have never seen their course publication tool
      • Would have to negotiate with EdX access to their platform’s source code

At the moment, Udemy is looking like my favorite candidate.

I also intend to talk to VPTL at Stanford to understand what are the usual difficulties professor have in the process of creating their online tools.


LDX will be a web-based tool which will overlay the existing LCMS with text, image, and video triggered by analyzing the steps and content being published in the course. This is where artificial intelligence comes to play. Let’s say that the user has published a 30 minute video – LDX might suggest that the video should be shorter. If the user publishes 50 pages of text with no images, LDX might suggest that images illustrate concepts more powerfully that text alone. LDX might prompt the user to insert a knowledge-check or reflection activity once the user has published 5 pieces of content. The idea is to provoke the user to think about how the learner will be processing the content towards learning.  

The key features of LDX are:

  • Virtual Student
    • 3D character that talks to the user
    • Guides the user through the process of creating the content
    • Asks questions about the content and format of the course as it is created
  • Media Editing Suite
    • Video editor
    • Image editor
    • Text editor
  • Course Publication Tool
    • Create course structure and progression
    • Add media
    • Create assessments (quizzes, multiple choice, reflections, and etc)
  • Curated Content
    • Access to similar courses to get examples
    • Ability to link to external material for student’s reference
  • Coaching
    • Peer-to-peer help to go through courses

This is definitely ambitious for the time and resources I have for this project. The biggest challenge, other than the sheer volume of features, is the Virtual Student. I would have to partner up with someone who has experience and access to such technology in order to create a prototype.