idea on new system

readme.md

Here I would like to propose the desgin of a web-based e-learning system.

Workflow

Test Givers

For test givers, the workflow will be:

1. Create a document of questions in RStudio with R Markdown
   • use all R and R markdown legacy
   • use a new defined function to replace variables(see Example below)
2. Export it as a collection of
   • markdown files (.md,.rmd)
   • plot images (.png,jpg,etc.)
3. Preview as a student
   • in a web browser
   • check questions and parameters
4. Upload exported collection to the server
5. Done!

Students

For students, they will:

1. Login the system
2. See a list of individually selected questions with randomly selected parameters
3. Answer and click save.

System admin

As the administrator, she/he will:

1. Install Node.js to render self-customized R markdown documents into web html pages.
2. Install a CMS system
   • for students to log in, save, edit, log out
3. Implement a uploading service
   • usually a ftp service

Example

This is a R markdown example that a test giver will assemble questions.

### Problem 1

Two numbers N and {{M}} have LCM = {{LCM}} and GCF = {{GCF}}. Find N.

'''{r echo=FALSE}
df <- data.frame(M = c(16, 8, 4),
LCM = c(48, 24, 12),
GCF = c(8, 4, 2)
)
df
'''

M	LCM	GCF
16	48	8
8	24	4
4	12	2

And one student will see:

### Problem 1
Two numbers N and 16 have LCM = 48 and GCF = 8. Find N.  

The idea is to using a data frame to output all possible value range/combinations.

Another example is as follows,

### Problem 2
If the area of a circle is {{N}}pi square feet, find its circumference. 

'''{r echo=FALSE}
df <- data.frame(N=(3:6) ^ 2)
df
'''

N
9
16
25
36

The student will see,

### Problem 2
If the area of a circle is 25pi square feet, find its circumference. 

Advantange

1. Compared to Latex, R Markdown is much easier to implement, customize and more efficient in compilation.

2. With R Markdown, we could fully use the R package, functions, markdown syntax, equations, plots, tables to generate high-quality documents.

3. With self-customized functions in Node.js, it is convenient to extend the syntax for furthur individual needs.