I've been submitting syllabi for classes to the powers-that-be, recently. The last two (the Human System Dynamics class I taught at Olin, and a "How Power Works" class) I submitted because I thought they'd be fun, but I think I've found one that would be much more useful.
It's a Computational Tools class, for to-be researchers in sustainable development (or really anything at the intersection of social science and earth science). The idea is to give an introduction to all the tools (or the subset I know about) that can be at a researcher's fingertips, from an understanding of XHTML to extract data from webpages to the basics of inversion theory. Simultaneously, the class provides a
semester-long education in the thought processes behind programming, computation, and modeling, using the various tools as both examples and building blocks.
Here are all the topics I've come up with, and I'm interested in any other ideas!
unix and shell scripting
techniques in programming: working with data files, data structures, algorithms, ideas of modern languages
useful Matlab toolboxes
techniques in network theory: metrics, algorithms
techniques in inversion theory
system dynamics: elements, insights
agent based modeling
xhtml for web data collection
algorithms from numerical analysis
source control and project management systems
spectral analysis and tools from signal processing
data collections and organizations to know
LaTeX and beamer
using the econ cluster
tools from ``Structure and Interpretation of Classical Mechanics''
insights from nonlinear dynamics and chaos
If it's not designed for people in my program, I would add the following:
OLS, GMM, VAR
In addition, there would be an ongoing project, where each student to chooses a field (like climate or child health) and expands a collective wiki by finding good data sources, modeling tools, and techniques used in that field.