Undergraduate Research with Prof. Norm Matloff: Ground Rules

Undergraduate research can greatly enhance a student's insight into his/her field. However, it is crucial to understand that this is very different from an ordinary course. Please keep the following in mind:

In most cases, you are producing code that people will actually use. So, it must be RIGHT. This is not like an ordinary course, in which a student gets most of the credit if the program is mostly correct.

Please adhere to the following rules (mostly pertaining to R):

• Before pushing code to GitHub, make sure it works in installed form. Code that works during your development process may depend on variables, files etc. that you create during that process -- but which will be missing in the installed version. Also, file names can change; what was in inst/examples will now be in examples.
• Please limit code line width to 80 characters, no TABs.
• Please do NOT use roxygen. Also, if you wish to use R6 classes, please clear it with me first.
• Concerning comment lines in your code: Again, you are NOT writing comments to avoid being docked points by your TA in grading; they are for the benefit of future users who may wish to modify/enhance the code, and for the benefit of YOURSELF: Commenting will help organize your thoughts as you write, and will enable you to read your own code several weeks later. More on this below.

Your code should have numerous clear, helpful comments. BEFORE you start writing a code file, place some comments at the top, explaining what the code is supposed to do, and in broad terms, how that will be accomplished. BEFORE writing a function, write comments stating the roles of the arguments, and the structure of the return value. DURING the process of writing a function -- NOT later, when you're done -- you should write comments explaining what the code is doing.

Finally, please sure to make continual forward progress every week. Yes, illness, work in your other classes and so on will mean that your progress will be less during some weeks, but do not give your research project last priority. If you do, you will have very little work done by the end of the quarter, and no one will be happy. :-(