Computational Modeling
Fall 2008

For today, you should:

1) work on Chapter 3, aim to finish today

2) prepare for a quiz (graph algorithms / performance)


Today:

1) survey summary

2) quiz 2

3) work on chapter 3

4) I'm available for questions


For next time you should:

1) read Chapter 4 and Barabasi/Albert paper

2) prepare an evaluation package

   a) choose 1-2 chapters you would like me to evaluate

   b) write your own rubric (you can use mine as a template
      and add or remove questions)

   c) fill out your rubric

   d) bring hardcopy of your rubric and chapter(s) to class


Survey summary
--------------

1) Total hours: about 60% at 12 hours or more, a few at 16+

Many of you report that you are working extra because
you want to, so that's excellent.

BUT: I am concerned that you overachievers don't understand
the meaning of the word "optional."

Suggestions:

   a) Since we are almost done with the swappy books, the amount
      of work will taper a little.

   b) Your book writing is probably getting more efficient.

   c) Optional really means optional.

   d) Try to stay close to 12 hours. 


2) Getting the material?

   Large majority say yes or almost.  Only a few who are
   uncomfortable.

3) Pace:  mostly just right, more saying too fast than too slow.

   Some requests for more notice on due dates (and honest
   advertising).

4/5) Comments about the books/class.

   Get rid of the bad books!
   More back of the room instruction/lecture.
   Pen and paper exercises?
   More coding, less book writing.
   More discussion of books.
   Choose books with more content, less history.
   Divide books by topic (or allow more shopping).
   Less reading.
   Tell us the point of the class!

6) More quizzes: majority yes, some dissent.

   Evaluative or thinking: overwhelming majority for thinking.

   Suggestion for a collaborative problem of the day.

7) More kinds of evaluation.

   Majority is ok with using the book as primary, but
   overwhelming call for me to provide feedback soon
   and at least a few times during the semester.

   Suggestion to get 2 peer evals for one chapter.

   Suggestion to calibrate expectation by preparation
   (tricky issue, gets right to the fundamental conflict of grading).

8) Flexible timeline vs. getting through everything?

   No hard deadlines.
   Select exercises.
   Provide code.
   "Free chapter", open-ended project.
   Depth is important -- don't just skim topics.
   Option to revise/re-evaluate chapter.
   Hand out the rest of (my) book.
   No need to write up all exercises in (your) book.
   
9) General comments:

   Physical model of real systems?
   Large class is suboptimal.
   Ninjas!
   Read more research papers.
   Large final project?