Section 1 of lecture 12 has Duplicated what is briefly presented in lecture 11.

Idea: Modify syllabus to include a small bonus for students who submit all assignments?

Consider adding

• Examples of new and delete with classes.
• Consider an example of static storage duration.
• Qualitative comparison of Global variables with Static class attributes.

Clarify further differences between when we can modify passed arguments vs. Not.

need to change method name add_class to add_course. The name is confusing when we are trying to discuss Python classes

Rather than passing input arguments in prescribed number and order, we should encourage students to use command line flags to specify input. For example, instead of giving example for launching area computation program as

$ ./compute_area 10 20

we should provide example that takes input arguments like this:

$ ./compute_area --width 10 --length 20

This way allows students to specify defaults for all input variables, and to enter input argument in any order at the command line.

Clarify order of arguments for curried functions.

Should be on a Monday after break, as opposed to a Tuesday as listed on Schedule.csv

Make mention of SIMD - single instruction multiple data, and its relation to numpy operations. GPU's excel at SIMD, but CPU's are still capable of executing instructions in parallel. Other common numpy (vectorized) operations include np.log, np.abs, np.maximum

Use numpy to load files.

Put together simple set up guides for VIM and Emacs. Consider making config files downloadable.

Several of the C++ examples in lecture 10 have Python style # comments.

Order of arguments, details on how linker makes 1x pass.

Explain cerr and ensure consistency in returning non-zero values upon exit. Clarify relationship (or lack of) between cerr and exiting a program.

https://github.com/CME211/notes/blob/master/lecture-02/0-outline.md#printing

change to correct computation of area

Some of the material like iterators are already covered elsewhere. Consolidate material more appropriately. Or tie it in more cohesively.

To set expectations that useful intermediary data structures should be declared upfront.

Also, make init method initialize data attributes, and then load_name_data can populate them. But all attributes should. be defined in the init method.

Sometimes students have confusion around what it is that Make is doing for us (and what it's not, e.g. it does compile the program but it doesn't run it).

It turns out that AFS is not working well with git and GitHub.

See: https://web.stanford.edu/group/farmshare/cgi-bin/wiki/index.php/User_Guide#Directory_paths

We need to update notes and train students to use /farmshare/user_data/[username] where [username] is their sunetid.

Should be discouraged until we cover it properly and in detail. It may obfuscate the examples.

currently, there is a long list of readings in lecture 3. These should be distributed to other notes files.

We need to be a little bit more careful with the terminology used. I.e. we should remove dynamic array from introduction. We are really introducing the relationship between static arrays and pointers, the notion of a pointer (and the corresponding operations that can be applied to a pointer type), as well as the notion of Heap/free-store memory. We should really emphasize that as users of C++ whom are not writing software for other programmers, we shouldn't ever use naked news or deletes in our program, relying instead on the STL (wherein our dynamically sized (and separately, also heap-allocated) vectors come into play).

• Waitlist
• Why develop on Rice? (Academic, professional, and efficiency reasons).
• How to set up code completion via Emacs (and Vim), use 4-spaces instead of tabs, etc.
  See pinned Piazza posts.

Consider moving underflow mention to section on floating point.

It would be good to add some notes on integer representation before the numeric conversion section. The key point is that the Python (3?) int type supports arbitrarily sized integers.

https://github.com/CME211/notes/blob/master/lecture-02/3-numbers.md#numeric-conversions

Include Jupyter notebook output cells so notes are readable online.

Currently the markdown files and notebooks only have input code.

• update schedule for first 7 sessions
• add TA office hours somewhere
• add syllabus to notes repo
• add schedule to notes repo
• update hw1
• update hw2
• udpate hw3

other questions:

• sanity check?

Homework repo: https://github.com/CME211/homework

1. One of the print formatting statements is missing a digit (a 2 in between a 4 and a zero)
2. One of the print statement showcasing use of enumerate is missing the first line of output.
3. It wasn't Euler it was Gauss!

We get many questions on homework in the forum. It could be useful to have a document/primer explaining how to ask type(s) of questions. Inspiration can be drawn from a Stack Overflow QA template. E.g. if the question involves understanding how code(s) should work: (i) clarifying details around the question, and possibly a MWE (ii) what has been tried/considered? (iii) attempting to use the Search function (and correspondingly, comping up with a relevant Title for the question if a new one must be created), etc.

It would be good to document the relationship between Jupyter Notebook and IPython.

In the introduction to C++ lecture, we should mention const types and provide examples for their use. Homeworks and the course project code and material should use and/or refer to const types where appropriate.

I think the notes are not so good on this matter:

"The scope of a variable is the region of code within which a variable is visible."