Homework 1
Due at beginning of class on Tuesday, 16 February 1999.
CLARIFICATION OF HOMEWORK 1:
In part 1 of the homework, you should explain the situation to
your boss and assess the validity of the design premise.
In part 2, you propose a ***plan of action*** to address the two
questions at the end of the paragraph.
There is, of course, no single "correct" answer to this problem,
but there will be some answers that are better than others, and
probably some that will miss the mark.
NOTE: This is an individual assignment, not a team assignment. The Honor Code applies. Work on this assignment is to be done individually.
Real message from a student this year (someone in a previous HCI class here at Va Tech, used with permission):
Over Christmas break I was talking with some people working on a new cellular phone system for Germany. The person in charge of interface design was asking about the strategy for interface development that we discussed last semester. He raised an interesting question which I wasn't really sure how to respond to. The system that they are designing is largely undefined, it's never been done before and they can't really accurately anticipate the types of tasks and actions and information that users are going to need access to on a regular basis. The interface design head argues that this is a reason to model the interface after the underlying system architecture (hardware, software, network). His claim is that in a situation where you don't and can't know how the system is going to be used the most important characteristic of it is an internal consistency. In other words, the interface needs to reflect the object model such that when a user attempts to perform an action that has not been anticipated that the system behaves in the manner that the user expects it to based upon the other tasks they have successfully performed. While I agree that there are certain objects that make as much sense to a user as to a designer (such as an account, in this case, with billing information) some of the object model is designed around technical issues like where's the server and how are we going to communicate with it. In practice this means that much of what he is talking about will mean structuring an interface around a bunch of data structures which aren't necessarily designed to make a lot of sense to the user. I just wondered what you would have to say about that; what do you do in cases where you really can't come up with a list of possible or even probable tasks? If you can't effectively design around the user because you don't really know what the user might do then how can you still take user issues and usability into account?
Your assignment:
Suppose you are a member of the development team. You are not even the designated usability person, because there isn't one in this project. However, you are taking this course and you feel you should write a memo (not more than 400 words) to your boss, the project manager, about this situation.
In particular, address:
- the validity/reasonableness of the design premise (i.e., in absence of information about how the system will be used, model the interaction design on the system architecture for consistency);
- the two questions at the end of the previous paragraph (they are essentially two ways to ask the same question, so you can address them as one question).