Groups are available here.
(due Feb 24)
OK, pointing out the problems in something is one thing, but improving it is another. In this phase of the Design Challenge, we’ll try to do just that: to take something we’ve critiqued and try to suggest how to improve it.
For better or worse, we’ll do this with the Design Challenge 1 results. I am making this assignment before seeing the results, but judging from the early results, there will be plenty of interesting ideas to consider.
- Schedule:
Tuesday, February 14th – partner requests
Wednesday, February 15th – partner assignments
Friday, February 17th – group work day
Friday, February 24th – redesign reports due
Working in Groups:
You must do this assignment with an assigned group. The idea is that you have a conversation/discussion to come up with the ideas. We will try to create groups of size 3. Groups of 2 will be allowed.
If you would like to request your partner/group, send email to the TA (mcorrell@cs.wisc.edu) before noon on Tuesday, February 14th. We will assign everyone a partner/group on Wednesday, February 15th. (We should have the list in class).
We encourage you to work with someone who comes from a different part of the University than you. However, we understand that this takes extra effort.
To make it easy for you to meet with your group, we will use the class time slot on Friday, February 17th for groups to meet. You can meet wherever you like (the classroom is a convenient place), or you might choose to meet another time. But, at least you know there is an hour that you all are available to meet.
If your group has 3 people, you must consider 4 designs. If your group has 2 people, you must consider 3 designs.
The Re-Design Task
Pick 3 or 4 (one more than your group size) designs that have been submitted. You may pick the Professor’s design as one of your designs. You may pick the designs you submitted (but if each person submitted one design, your group will need to consider at least one design that it did not create. It is best if you choose a diversity of designs.
For each design, summarize its good and bad points. Think about these in the various ways we’ve discussed in class: what kinds of questions does it help answer? how does it utilize the perceptual system? who might be able to make use of it? How might it scale to larger data sets?
Now come up with something “better.” Of course, you need to decide what better means (what kinds of questions might it be good at answering, how might it scale, …). Ideally, when you improve a design, you make it better in multiple ways. Ideally, it doesn’t get any worse – but if you make a tradeoff (it gains X but loses Y), be sure to articulate it, including why you think its a good tradeoff.
The intent is that this is a redesign – that you are suggesting changes to one of the existing designs in order to make it better. However, where re-design ends and new-design starts is gray. If your new design is radically different than the ones you picked initially, you have a little more responsibility in describing it in sufficient detail.
Each group must create two “new” (or re-designed) designs. You need to describe them in detail, including the changes that you made to the original design and why you made them. Be sure to explain why you think each is “good” (e.g. solves some problem, makes use of perceptual skills, looks nice, …). Provide sketches. Describe interactions or movement.
We are open to you using novel methods to communicate your ideas with us. Your handin must include some written document with pictures in it (either as Word or PDF, or even as a web page). But if you want to communicate your design using a video, or an interactive prototype, or … If you are thinking of some alternative thing, let us know so we can figure out a hand-in mechanism. (this was inspired by me thinking about DoodleCast and ScreenChomp)
What To Turn In
Your group must turn in a single document (in either Word or PDF form. if you have a web page, print it to PDF as well as send us a ZIP file with everything we need to look at it).
In this document, you must:
- Discuss 3 or 4 (one more than your group size) designs. Please include a description/sketch of the design (re-using the picture from the posting is OK).
- Present TWO redesigns. While the original intent is that you pick 2 of the ones you’ve evaluated, it could be that you pick one of them and modify it in two very different ways.
You will turn this document in by 4pm on Friday, February 24th. Send it by email to the TA (mcorrell@cs.wisc.edu). Late assignments will be accepted, but may be penalized.
How we will evaluate it
We will look to see that you have thought through the rammifications of the designs you picked – applying ideas from the class, along with your creativity. We will look to see that you have gained some insights on how these designs solve problems and may be effective presentations (or not).
We will assess how good your resulting design is, as well as how well it improves on the starting points. Taking a mediocre design and making it a little better (or a good design and making it only slightly better) is not as good as taking a mediocre design and changing it so it is more effective, or taking design that is really good for one thing and finding a way to adapt it to be good for something else as well. The preference is for transformations from good to great.
There is a tradeoff with this: you can make a good design great, or make a mediocre design acceptable. With bad designs, there is more room for improvement, so you might make more improvements. We will value both absolute quality of the results and “amount” of change. Especially since a small change to a good design might be challenging because it has to be sure not to damage what was already there.
We will assess how good your reasoning is. If you came up with a great design without telling us how or why, we may think you just got lucky. Explain your rationale for the choices you have made.
Roughly, the grading will be:
- 20% Presentation Quality (how clearly does your hand-in articulate what it is supposed to articulate)
- 30% Evaluation of Previous Designs (how well does it show insight into the problem and design space, how well does it make use of concepts from class, …)
- 50% Quality of re-designs (Note that we do not separate “good arguments” from “good designs.” Your reasoning for your design is as important as the design itself. Part of a good design is explaining why it is a good design. Put another way, we are really evaluating the rationale for your design, not just the design itself- you need to tell us why your design is good). Again, we will be looking for the application of ideas from class, your ability to see what problems the tool might be useful for, … Note: it is better for you to point out the issues/downsides in your design than for us to think of them ourselves.
The expectation is that this should be a page or so (with pictures) for each of the existing design, and maybe 2 or so pages for each of the re-designs.
There is a Piazza page for discussion of this assignment.