Why do portals all
look alike?
By
Brad Neuberg
Introduction
Your final project
is to design
and implement iris, an exercise in information visualization.
Iris is not supposed to be an acronym (if you insist, you
could try ``Information Reification through Interactive Software''),
but rather stands for the topic of a small set of data that you'll
be visualizing.
The data
present information about 150 instances of flowers, fifty each
of three different species of iris. Each instance is represented
by a record containing five items: sepal length, sepal width,
petal length, petal width, and the genus and species of the instance.
In the database included in this file, each instance's record
occupies one line containing comma-separated fields.
If you are
interested, you can find out more about this database here. While the iris
data often used for exploring algorithms for automatically analyzing
and classifying data, this is not what we'll be doing
with it. Instead, you will be designing and implementing ways
to allow the user to visualize this data to gain a better understanding
of it.
Using the
general user interface techniques and information visualization
techniques discussed in class and in the readings, your task is
to develop an interactive system that lets the user gain the best
understanding of the data that you can make possible. You can
choose techniques such as detail and context views (e.g., as developed
for SDMS), dynamic queries (e.g., as implemented in the MovieFinder),
or any of other techniques (not necessarily restricted to those
covered in class and in the readings). No matter which set of
techniques you choose, however, you will need to develop a user
interface that exploits these techniques.
Note that
you are free to ``hardwire'' the data set into your system. While
a ``real'' visualization system would enable you to load external
data sets, you've already written enough file I/O and parsing
code this semester, that a reprise wouldn't be valuable. You will
demo your version of iris to us at the end of the semester,
so you should start thinking about how to show off its functionality,
emphasizing user interface design and implementation.
Your
submission will consist of your java code directory, which should
contain your .java and .html files, and a README file. The README
file should include any information needed to understand your
code beyond the comments in the other files. and explanations
of the choices you made in implementing their functionality. Please
justify these choices by appeal to material in the course text,
or (optionally) by referring to Smith and Mosier, Guidelines
for Designing User Interface Software or NASA Goddard
User Interface Guidelines, or by explicit reference to any
additional reading that you have done.
Comment
2 added by Anthony Gumbs*
on
Mon 12/14/1998
Please
start early! You cannot use any late days for the final project.
Please read
Shneiderman Chapter 15 carefully. You may also wish to look at
some of the cited references.
While you
do not have to read or write files as part of this project, you
may find yourself getting bored looking so much at the same single
dataset. You may also discover that this dataset does not show
off some of what you have done as well as other data. Therefore,
you are welcome to either implement a self-describing data file
format and let the user specify datasets at runtime, or to simply
include multiple hardwired data sets in your program. (Again,
note that the emphasis here is on the user interface, not parsing.)
