Alan Isaac
Alan Isaac
==================
**Reminder:**
The Segregation Exercise is due Wednesday.
Allocate at least 10 hours to complete this assignment.
Post your questions now (to the *Segregation* thread).
Do not wait until the last minute!
**Reminder:**
Spend **lots** of time on your project.
By now, grad students should have committed
some code to the Subversion repository.
(See the thread *Version Control Using Subversion*.)
By the end of this week, all students should have
- refined the model outline
- implemented a syntactically correct
“skeleton” program based on the outline
- implemented some key procedures
**Reminder:**
Timely and substantive participation in the discussions is
a graded activity. Please participate in **discussion**.
Posting a summary of a reading *begins* a discussion.
Assignments
------------
- Experiment with Schelling's segregation model as implemented at
NetLogo > File > Models Library > Sample Models > Social Science > Segregation
Post questions or observations in the *Model's Library* thread.
- Learn how to get data out of a simulation using Behavior Space:
https://subversion.american.edu/aisaac/notes/netlogo-intro.xhtml#behaviorspace
http://ccl.northwestern.edu/netlogo/docs/behaviorspace.html
Use table output, and try to read the output file into your favorite spreadsheet.
Please discuss your experiences in our discussion group.
(You need this skill for the Segregation Excercise.)
- Make progress on the Segregation exercise.
- This week you need to make substantial progress on your Course Project.
I'd like to see discussion of your project on the *Project Topics* thread;
do not simply try to develop it in isolation or you will attempt too much.
I will reduce other demands of the course,
and I'll try to pick readings that you should find helpful for your project.
Readings
--------
**Required reading:**
- Start [railsback.grimm-2011-pup]_ ch. 9.
(Emphasis: the use of Behavior Space.)
- http://ccl.northwestern.edu/netlogo/docs/behaviorspace.html
- [rauch-2002-atlantic]_
http://www.theatlantic.com/magazine/archive/2002/04/seeing-around-corners/2471/
Graduate students should additionally read:
- [hegselman.flache-1998-jasss]_
http://jasss.soc.surrey.ac.uk/1/3/1.html
(you can just skim the math)
**Recommended (i.e., entirely optional) reading and resources:**
- http://en.wikipedia.org/wiki/Conway%27s_Game_of_Life
Some Possible Topics for Discussion
-----------------------------------
- Segregation:
Should the Schelling model interest social scientists?
- Segregation:
Is the Schelling model a cellular automaton?
Why or why not?
- Optional:
I welcome discussion of the Wedding Ring model (RG ch. 6).
Alan Isaac
==================
**Reminder:**
The Segregation Exercise is due by midnight tomorrow.
I was expecting more discussion and questions by now.
**Reminder:**
Timely and substantive participation in the discussions is
a graded activity. Please participate in **discussion**.
Posting a summary of a reading *begins* a discussion.
Assignments
------------
- Segregation Exercise:
Finish up and turn it in by midnight tomorrow.
Turn in your report as a PDF.
You may also turn in a .nlogo file if you feel
it is necessary. If you do, here are
some lessons from the previous assignment.
- Always put your name and the date at the top
of your source files.
- Fully comment all procedures.
Make your comment lines reasonably short,
using multiple lines for long comments.
Put comments on their own line when that reduces clutter.
- Always attribute borrowed code (e.g.,
by adding a comment with the author's name).
- Avoid hardcoding numbers when possible.
- Distinguish between procedures and reporters
(``to`` vs. ``to-report``): only reporters
can use the ``report`` keyword.
- When you work with patches, work with
them directly (instead of working with
their coordinates)
- Most important: bring the error messages that
you encounter to the list for discussion,
so that you do not have to struggle on your own.
- Spend **lots** of time on your project.
By the end of this week, you should have refined your model
outline and implemented a skeleton where not all the
procedures are empty.
Readings
--------
**Required reading:**
(Emphasis: the use of Behavior Space.)
https://subversion.american.edu/aisaac/notes/experimental-design.xhtml
(Discuss in the *Experimental Design* thread.)
- Reread the documentation for Behavior Space:
http://ccl.northwestern.edu/netlogo/docs/behaviorspace.html
Graduate students should also skim:
- http://www.palgrave-journals.com/eej/journal/v34/n4/full/eej200826a.html
(you can just skim the article, which you will need to access
through the AU library, unless you want to read an earlier version:
http://andromeda.rutgers.edu/~jmbarr/NYComp/BarrEEA.pdf)
**Recommended (i.e., entirely optional) reading and resources:**
http://en.wikipedia.org/wiki/Verification_and_validation_%28software%29
Some Possible Topics for Discussion
-----------------------------------
What are the uses and difficulties of this NetLogo facility?
- Segregation:
What are the lessons of the Schelling model
for social scientists?
- Course Project:
Recount your implementation difficulties and frustrations.
Alan Isaac
Week 5: Wednesday
==================
**Reminder:**
The Segregation Exercise is due by midnight.
**Reminder:**
Spend lots of time on your project.
By Friday, you should have refined your model
outline and implemented a skeleton where not all the
procedures are empty.
Readings
--------
**Required reading:**
- Finish reading the Lecture Slides on Experimental Design
https://subversion.american.edu/aisaac/notes/experimental-design.xhtml
Graduate students should also skim:
- nothing today
**Recommended (i.e., entirely optional) reading and resources:**
- Sofware Testing
http://en.wikipedia.org/wiki/Software_testing
Assignments
------------
- Learn how to get data out of a simulation using file operations:
https://subversion.american.edu/aisaac/notes/netlogo-intro.xhtml#open-a-file
Try the following in any model (e.g., the segregation model or your own model).
During ``setup``, open a file ``temp.csv``.
Each tick, write one number on one line of the file.
(This could just be the value of ``ticks``, or something more interesting.)
At the end of the simulation, be sure to close the file using ``file-close``.
Now, try to open the file you created in your favorite spreadsheet.
Please discuss your experiences in our discussion group.
- Next, try to set up a Behavior Space experiment.
Use table output, and try to read the output file into your favorite spreadsheet.
Please discuss your experiences in our discussion group.
Some Possible Topics for Discussion
-----------------------------------
Reminder: timely and substantive participation in the discussions is
a graded activity. Please participate in **discussion**. Posting
reading summaries *begins* a discussion.
- Experimental design: experiences and difficulties with BehaviorSpace.
- Segregation model: continue ongoing discussion.
Discuss whether this should interest social scientists.
Is this a cellular automaton?
Alan Isaac
Week 5: Thursday
==================
**Reminder:**
The Segregation assignment was due last night.
**Reminder:**
Spend lots of time on your project.
By tomorrow, you should have refined your model
outline and implemented a skeleton where not all the
procedures are empty. Graduate students should post
their skeleton program to the Subversion repository.
**Reminder:**
Timely and substantive participation in the discussions is
a graded activity. Please participate in **discussion**.
Posting reading summaries *begins* a discussion.
Assignments
------------
- Course Project:
Implement your two-page outline as a (possibly empty) collection of procedures.
Add procedure bodies one at a time, testing (!) as your go.
- Course Project:
Decide on your outcome measures for your experimental design.
Readings
--------
**Required reading:**
- [kelton-1999-dse]_
http://www.informs-sim.org/wsc99papers/004.PDF
(Do not worry about the math.)
- Kleijnen et al. (2005)
A User's Guide to the Brave New World of Designing Simulation Experiments
http://ideas.repec.org/p/dgr/kubcen/20031.html
sections 1 & 2
Graduate students should also skim:
- the rest of Kleijnen et al. (2005)
**Recommended (i.e., entirely optional) reading and resources:**
- nothing today; work on your project!
Some Possible Topics for Discussion
-----------------------------------
- Experimental Design:
Post your experiences and difficulties with BehaviorSpace.
- Segregation Model:
Continue ongoing discussion.
- Course Project:
Share your experiences, difficulties, questions, etc.
- Testing:
R&G chapter 6 emphasize the importance of testing
as part of the process of writing code.
What kinds of tests are you finding helpful?
Which seem more trouble than they're worth?
Alan Isaac
==================
**Reminder:**
The ``First Approach to the Template Models`` is due
next Wednesday by midnight.
**Reminder:** Your course project is due June 26!
For the next two weeks, spend **lots** of time on your project.
By today, you should have refined your model outline
and implemented a skeleton where some procedures are not empty.
By today, graduate students should commit this skeleton to our
Subversion repository.
**Reminder:**
Timely and substantive participation in the discussions is
a graded activity. Please participate in **discussion**. Posting
reading summaries *begins* a discussion.
Assignments
------------
Start preparing this for submission next Wednesday.
You will want to make good progress on this exercise
over the weekend. Allow 10 hours for completion.
- Course project: code and paper are due June 26!
Readings
--------
**Required reading:**
(Reread, wiwth emphasis on the use of Behavior Space.)
Graduate students should additionally read:
(Reread; master the details.)
**Recommended (i.e., entirely optional) reading and resources:**
http://www.demographic-research.org/Volumes/Vol17/3/
Some Possible Topics for Discussion
-----------------------------------
Please post difficulties encountered in implementing
this assignment.
- Course Project:
Please post difficulties encountered in implementing
a syntactically correct "skeleton" version of your model.