Hello everyone,
Raymond Yee, who teaches the "Working with Open Data" course at the
School of Information, will facilitate a discussion about "Teaching
Analysis and Presentation of Open Data" next week, on Thursday 5 June
from noon to 1pm, at our usual meeting place: 200C Warren Hall.
Here are Raymond's readings for this group (written by him, hence the
first-person in italics):
/As a way of jumpstarting a discussion on the topic of teaching open
data analysis, I'd like to briefly share my experiences of teaching
"Working with Open Data". Here's a write-up no the 2013 edition of the
course.//
/
http://blog.fperez.org/2013/05/exploring-open-data-with-ipython.html
/I've not written up an analysis of the 2014 course yet. But you can
find the course materials for the course at/
http://is.gd/wwod14
/Specifically, the final projects can be found at/
https://github.com/working-with-open-data-2014/project-organization/wiki/Working-with-Open-Data-2014-Projects
/I mention Software Carpentry below, so it'd be good for us to take a
look at:/
http://software-carpentry.org/faq.html
Our discussion will focus on these questions (a mix suggested by Steve
and Raymond):
(a) What technical issues, if any, are presented by open data that
differ from copyrighted, restricted, or sensitive data sets used in some
campus research?
(b) What kinds of technical / infrastructural resources could be made
available to instructors to facilitate teaching and focus class/student
time on topics and activities of core value in the context of a
course/project?
(c) What is suggested by the following brainstormed list of areas for
improvement for a next version of "Working with Open Data" --
improvements Raymond would try to make if more time and resources were
available:
* expand the implementation of bCourses/Canvas API to make it easier
to do more frequent quizzes and automated grading
* more video, screencasts to enable students to revisit my lectures
outside (and inside the classroom)
* having students make screencasts
* teach students how to use the whole range of the Census API and
how to process bulk census data
* basic front end web development skills, including JavaScript
* Firmer integration of software carpentry and introductory Python
review
* Statistics training
* Test driven development
* Python outside of the IPython Notebook
* Systematic overview and coverage of the open data intellectual
property regime and movement
* Formal writing and presentation skills
* Working Real world clients
* Coverage of scaling up computation
(d) In general, what can be done to lower the intellectual and technical
barriers in working with the combination of open data, open source
software, and scalable on demand computational infrastructure?
See you next week for what promises to be a very interesting discussion,
Steve
--
=====================
Steve Masover, IT Architect
IST - Research Information Technology
maso...@berkeley.edu
