Providing a Context
to Motivate NonMajors Into
Computing
Mark Guzdial
College of Computing/GVU
Georgia Institute of Technology
Story



Georgia Tech and the challenge of teaching
programming to everyone
Contextualized CS as a way of motivating the
study of computer science
What does contextualized CS look like?



Introduction to Media Computation as an example
Assessment results from three different CS1’s
Where we’re going next
Georgia Tech

Georgia Institute of Technology


28% female overall
Colleges: Computing (11% female undergrad),
Engineering, Ivan Allen (Liberal Arts), Sciences,
Management, Architecture
Universal requirement of computing

Georgia Tech requires a course in computing of
every student.


Consider computing a necessary prerequisite in a
technological institution.
Only a single course has been available:
CS1321 Introduction to Computing
Developed for our majors
 Uses Scheme with the text How to Design Programs

Universal access to computing

In 1961, Alan Perlis argued that computer
science should be part of a liberal
education.


Explicitly, he argued that all students
should learn to program.
Consider a contrast with Calculus
Calculus is about rates, and that’s
important to many.
 Computer science is about process,
which is important to everyone


The best uses of technology are going to
come from outside of computer science.

If we want computing technologies to
become useful, they have to get out of our
hands.
How close are we to being
able to teach everyone CS?

Not very:
CS1 is one of the most despised courses for nonmajors

At many departments, CS retention rates are lower than the
rest of campus


Drop-out rates in CS1 near 50% at many institutions


At Georgia Tech: 65% for 1995 cohort, vs. 73% for
Engineeering
At Georigia Tech: 28% WFD (Withdrawl, F, or D rate)
typically; but with some spikes.
Female enrollment in CS has been dropping nationally
Why?

Several recent studies and books
claim that CS instruction tends to
dissuade anyone but white males

“Tedious,”
“taught without application
relevance,”
“boring,”
“lacking creativity,”
“asocial”
The Challenge

In order to motivate students to care about
computing, we need to make it



relevant,
social, and
creative
Contextualized CS1

During Spring 2003, Georgia Tech offered three
different CS1 courses:


CS1321: Traditional CS1 (n=127, in section studied)
COE1361: Computing for Engineers (n=75)


Same conceptual content as CS1321 but in MATLAB
and Java with data-first applications.
CS1315: Introduction to Media Computation (n=121)

Introduction to programming and computing concepts,
using media manipulation as a context.
Comparing the three courses
CS1
(CS1321)
COE
(COE1361)
MediaComp
(CS1315)
Learning
Objectives
Programming
Language
Majors
Students as
software
developers
Tool builders
Scheme
MATLAB &
Java
Tool modifiers Python
(Jython)
All, but
designed for
CS majors
Engineering
Management,
Liberal arts,
Architecture,
Biology
Assessment

Led by Ph.D. student Andrea Forte




With Lauren Rich (Undergrad) and Rachel Fithian
(MS HCI)
Surveys in all three courses: Beginning, midterm,
and final survey
Interviews with students in CS1 and MediaComp
Demographic data
Start of term survey:
What is Computer Science?

The audiences for these classes come in with
different preconceptions.
Course
Don't
Know
How
ProgramComputers ming
Work
Scary/
Required
Difficult Class
Web/
Internet
CS1
4.9%
6.1%
51.2%
8.5%
7.3%
0.0%
MediaComp
4.7%
7.0%
48.8%
10.5%
15.1%
8.1%
COE
11.4%
18.2%
45.5%
2.3%
4.5%
0.0%
Particular Focus: Introduction
to Media Computation

121 students in Spring 2003,
with 300 planned for Fall 03 and 450 for Spring 04


2/3 female in Spring 2003 MediaComp
Focus: Learning programming and CS concepts
within the context of media manipulation and
creation

Converting images to greyscale and negatives,
splicing and reversing sounds, writing programs to
generate HTML, creating movies out of Web-accessed
content.
Motivating the Computing
As professionals, these students will often
the use the computer as a communications
medium.
 All media are going digital,
and digital media are manipulated with
software.
 Knowing how to program, then, is a
communications skill.

def clearRed(picture):
for pixel in getPixels(picture):
setRed(pixel,0)
def greyscale(picture):
for p in getPixels(picture):
redness=getRed(p)
greenness=getGreen(p)
blueness=getBlue(p)
luminance=(redness+blueness+greenness)/3
setColor(p,
makeColor(luminance,luminance,luminance))
def negative(picture):
for px in getPixels(picture):
red=getRed(px)
green=getGreen(px)
blue=getBlue(px)
negColor=makeColor(255-red,255-green,255-blue)
setColor(px,negColor)
Use a loop!
Our first picture recipe
def decreaseRed(picture):
for p in getPixels(picture):
value=getRed(p)
setRed(p,value*0.5)
original
Used like this:
>>> file="/Users/guzdial/mediasources/barbara.jpg"
>>> picture=makePicture(file)
>>> show(picture)
>>> decreaseRed(picture)
>>> repaint(picture)
Recipe to Increase the Volume
def increaseVolume(sound):
for sample in getSamples(sound):
value = getSample(sample)
setSample(sample,value * 2)
Using it:
>>> f="/Users/guzdial/mediasources/gettysburg10.wav"
>>> s=makeSound(f)
>>> increaseVolume(s)
>>> play(s)
>>> writeSoundTo(s,"/Users/guzdial/mediasources/louder-g10.wav")
Generalizing Algorithms


We talk about algorithm complexity later in the
course, after the media is done.
We talk about different approaches to the same
problem, where the criteria might be aesthetics or
correctness, instead of speed or size


For example, generating greyscale
During the media, we point out similar themes in
different functions.

We refer to them as “sub-recipes”
Scaling the picture down
def copyBarbsFaceSmaller():
# Set up the source and target pictures
barbf=getMediaPath("barbara.jpg")
barb = makePicture(barbf)
canvasf = getMediaPath("7inX95in.jpg")
canvas = makePicture(canvasf)
# Now, do the actual copying
sourceX = 45
for targetX in range(100,100+((200-45)/2)):
sourceY = 25
for targetY in range(100,100+((200-25)/2)):
color = getColor(getPixel(barb,sourceX,sourceY))
setColor(getPixel(canvas,targetX,targetY), color)
sourceY = sourceY + 2
sourceX = sourceX + 2
show(barb)
show(canvas)
return canvas
Scaling the picture up
def copyBarbsFaceLarger():
# Set up the source and target pictures
barbf=getMediaPath("barbara.jpg")
barb = makePicture(barbf)
canvasf = getMediaPath("7inX95in.jpg")
canvas = makePicture(canvasf)
# Now, do the actual copying
sourceX = 45
for targetX in range(100,100+((200-45)*2)):
sourceY = 25
for targetY in range(100,100+((200-25)*2)):
color = getColor(getPixel(barb,int(sourceX),int(sourceY)))
setColor(getPixel(canvas,targetX,targetY), color)
sourceY = sourceY + 0.5
sourceX = sourceX + 0.5
show(barb)
show(canvas)
return canvas
Recipe for halving the
frequency of a sound
def half(filename):
source = makeSound(filename)
target = makeSound(filename)
This is how a
sampling synthesizer
works!
sourceIndex = 1
for targetIndex in range(1, getLength( target)+1):
setSampleValueAt( target, targetIndex,
getSampleValueAt( source, int(sourceIndex)))
sourceIndex = sourceIndex + 0.5
Here are the
play(target)
return target
pieces that
do it
Compare these two
def copyBarbsFaceLarger():
# Set up the source and target pictures
barbf=getMediaPath("barbara.jpg")
barb = makePicture(barbf)
def half(filename):
canvasf = getMediaPath("7inX95in.jpg")
source = makeSound(filename)
canvas = makePicture(canvasf)
target = makeSound(filename)
# Now, do the actual copying
sourceX = 45
for targetX in range(100,100+((200-45)*2)):
sourceIndex = 1
for targetIndex in range(1, getLength( target)+1): sourceY = 25
for targetY in range(100,100+((200-25)*2)):
setSampleValueAt( target, targetIndex,
color = getColor(
getSampleValueAt( source,
getPixel(barb,int(sourceX),int(sourceY)))
int(sourceIndex)))
setColor(getPixel(canvas,targetX,targetY), color)
sourceIndex = sourceIndex + 0.5
sourceY = sourceY + 0.5
sourceX = sourceX + 0.5
play(target)
show(barb)
show(canvas)
return target
return canvas
Both of them are sampling

Both of them have three parts:
Setting up the objects: Source and
target
 A loop where samples or pixels are
copied from one place to another





To decrease the frequency or the size, we
take each sample/pixel twice
In both cases, we do that by incrementing
the index by 0.5 and taking the integer of
the index
Finishing up and returning the result
Making it better: Averaging/blurring to
smooth the result
There are
serious
algorithm
issues here, but
it’s not sorting.
Using your personal pictures
And messin’ with them
Relevance through Data-first
Computing


Real users come to a user with data that they care about,
then they (unwillingly) learn the computer to manipulate
their data as they need.
MediaComp works the same.
We use pictures of students in class demonstrations.
 Students do use their own pictures as starting points for
manipulations.



Some students reversed sounds looking for hidden messages.

How often do students use their second week of CS1 on their own
data?
How does that change the students’ relationship to the material?
They started doing this in the second week


In COE course, 30% of the students reported at midterm
that they had used some of the course content.
Rough overview of Syllabus


Defining and executing functions
Pictures
Psychophysics, data structures, defining functions, for
loops, if conditionals
 Bitmap vs. vector notations


Sounds
Psychophysics, data structures, defining functions, for
loops, if conditionals
 Sampled sounds vs. synthesized, MP3 vs. MIDI


Text



Converting between media, generating HTML, database,
and networking
Movies
Then, Computer Science
Computer science as a solution
to their problems

“Writing programs is hard! Are there ways to
make it easier? Or at least shorter?”



“Movie-manipulating programs take a long time
to execute. Why?”


Object-oriented programming
Functional programming and recursion
Algorithmic complexity
“Why is PhotoShop so much faster?”


Compiling vs. interpreting
Machine language and how the computer works
Assignments encourage
learning in social contexts
Homework are all collaborative
 Quizzes are preceded by nearly-identical,
collaborative pre-quizzes
 Two “take-home exams” (programming
assignments) are non-collaborative

Assignments encourage
creativity

For several homeworks, the task is to manipulate
media in some way, but we don’t care what media


For example, creating a collage or building an
animation
Encouraging homework results to be posted to
CoWeb (collaborative website) in galleries
First Homework assignment
Homework 1:
Write a program named hw1 to accept a picture as input,
and change its pixels as follows:
•
•
•
Set the green component to 125% of its current value
Decrease the blue by 25%
Decrease the red by 75%
Solutions shared in the CoWeb
Homework #3: Make a collage with images that
you modify by code only—any images you want
End of term results


Of the 121 students who
started, only three dropped
MediaComp.
When asked “What part of this
class should not change?”
19% of MediaComp
students mentioned the
CoWeb
 20% mentioned
“collaboration” in some
form
WFD Rate
CS1
42.9%
COE
18.7%
MediaComp
11.5%

Midterm survey: “What do you like
best about this class so far?”
Don't like it
/Nothing
Enjoy
Content
Content is
Useful
Traditional CS1
18.2%
12.1%
0.0%
COE
12.9%
16.1%
25.8%
MediaComp
0.0%
21.3%
12.4%
Midterm MediaComp Survey:
“What do you like about the class?”
“I like the feeling when I finally get something
to work.”
 “Very applicable to everyday life.”
 ‘I dreaded CS, but ALL of the topics thus far
have been applicable to my future career (&
personal) plans—there isn't anything I don't
like about this class!!!”

Non-majors surprised to find that they
like contextualized computing


Interviewer: Has this class changed your opinion
of computer science?
MediaComp Student (female): Yes, I’m not
intimidated by it anymore. My mom was so
surprised when I told her I want to be a TA she
almost fell on the floor, ‘cause she’s heard me
complain for years about taking this class and
now I want to go do it to myself again!
Non-major students want to learn
about computing

From an interview with an Engineering student in
traditional CS1:
“I think if anything [should be required] it should be the
second one [course], like Java. A lot of people know that
Scheme isn’t used and so a lot of people want to copy the
homework and get through it. They’re not very
motivated. I’ve heard that Java and C and all those, they
are different, and I’d like to learn C and Java and all that.”
Where we’re going next

This was only the pilot. We should still be skeptical.

Where do these students go next?
Does MediaComp influence how they use computing in
their own major?
 Does it influence any of them to take more CS?


How does it change as it goes out beyond?
Spring ’04: Others teaching it at Tech.
 Already being trialed elsewhere (at Gainesville College)



How well does Python transfer to other languages and
contexts?
Exploring a high school version of the course.
Conclusions


Our classes are driving people away from computer
science.
Part of the problem is that we are too abstracted away
from the relevant problem domains.
We are not Mathematics.
 These students won’t use Calculus every day,
But they will use Computing.
 We have to teach better.



Contextualized CS seems to be leading to more motivation
and better performance.
Hard issues here:


Budgets and teaching loads—”for non-majors?!?”
But it may be our greatest impact as a discipline—
extending CS beyond our borders.
Acknowledgements



Course materials development: Jason Ergle, Claire Bailey,
Ellie Harmon, Toby Ho, David Raines, Joshua Sklare,
Adam Wilson, Andrea Forte, Mark Richman, Matt
Wallace, Alisa Bandlow, Keith McDermott, Derek
Chambless, Larry Olson, Eric Mickley, Lauren Biddle.
Assessment: Andrea Forte, Rachel Fithian, Lauren Rich
Thanks for support



to Vice-Provost Bob McMath and the Al West Fund,
to GVU and the College of Computing, and
the National Science Foundation
For further information



Course CoWeb:
http://coweb.cc.gatech.edu/cs1315
Where we planned the course:
http://coweb.cc.gatech.edu/mediaComp-plan
guzdial@cc.gatech.edu