Editing Design Stories (OLD)

The EDURange Design


The guiding principles:
    1. EDURange is flexible for creating new exercises, so that it supports polymorphism.
    2. Solving the exercises should require analytical skills and abilities.
    3. EDURange allows the instructor to see what individual students are doing, so that they can provide real-time feedback.
    4. Easy to use

'''For the instructor'''

* Seeing what individual students are doing
* The instructor should be able to clone an exercise and add students or groups and start the exercise easily
* The instructor should see the status of a scenario and re-start if necessary
* The instructor should be able to see the sequence of commands (including their time) that a student typed, what the context was (host and directory) and what the resulting shell output was. This needs to be presented in such a way that the instructor can quickly determine if the student is making progress.
* The instructor should be able to see the answers that students gave to the questions.
* The instructor should be able to be able to offer a hint, after seeing that a student has a question or wants a hint and interact with the student via chat.
* The instructor should see a ranked list of students who need help. ML should rank the  students.

'''For the student'''  

* The student should be able to register for an exercise that the instructor created.
* The student should be able to run/use the exercise (once started) from any computer with a browser.
* The Guide should be clear, in particular, the questions should be auto-graded unless they are essay questions.
* The students should see which questions they have answered correctly
* The tasks should follow a logical sequence

'''For the researcher/developer'''

* The scripts that analyze the data should be easy to run.
* The final data should be collected in a single location at least for each single instance of a scenario.
* It should be possible to pool data from multiple runs of the same scenario. Any data that is exported should be anonymized It should be possible to verify functionality, e.g. that a scenario will start without error, that a question will be graded correctly, that a system will usable by n users simultaneously, etc.

'''ML'''

* The ML system should be able to use multiple models for predicting student challenges and should be milestone-specific
* It should be possible to run a trained model and make predictions during a scenario  Author The author should be able to make changes to the Guide.
* The author should be able to specify host and network information
* The author should be able to specify learning objectives for each task in a scenario
* The author should be able to specify milestones, i.e. indicators that a learning objective has been met.