Seminar 1

(Older edition)

Chapter 7 - Identifying needs and establishing requirements

This chapter mainly discusses requirements and data gathering. A requirement is a statement about an intended product that specifies what it should do or how it should perform, as stated in 7.3. There are traditionally two different kinds of requirements in software engineering that have been identified; functional requirements and non-functional requirements. Functional requirements tell us what a system should do, and the non-functional requirements tell us about the constraints on the system and its development. These two requirements hold more specific types of requirements however, such as data requirements, environmental requirements and user requirements. 

There are several kinds of different data-gathering techniques. Each one has its pros and cons. Questionnaires, interviews,  focus groups and workshops, naturalistic observation and studying documentation are some examples of those techniques.

Chapter 13 - Asking users and experts

This chapter focuses on interviewing techniques. How to develop and plan an interview is explained, and some simple steps one can use to conduct a better interview experience for the interviewee are presented. The steps are in order; introduction, warmup session, main session, cool-offperiod and closing session.

Interviews can be structured, unstructured or semi-structured. Structured interviews are similar to questionnaires in that the questions are predefined and is often used when the study's goals are clearly understood. Closed questions are typically used in structured interviews. It's important that the same questions are used for each participant for a standardized study. Unstructured interviews resemble normal conversations and use more open questions. It is advisable to have a clear interview agenda to stay "on topic". A great benefit with this kind of interview is a richer data and often new ideas never considered pop up. On the other hand a very time-consuming interview and more difficult to analyze, hence a proper structure and preparation is required for maximum efficiency. Semi-structured interviews combine aspects of both previously mentioned interview types and pose both closed and opened questions. The questions are preplanned and the interviewer uses probing to obtain more relevant information. It's important here not to phrase questions which expects particular answers.

Questionnaires have the advantage to be distributable to a much larger number of people. However, the questions need to be carefully thought out  to ensure that the data can be analyzed efficiently. People are also not as prone to complete questionnaires as they are answering questions asked by an interviewer.

Our group chose the ferry route Slussen to Allmäna Gränd since we thought that it would be easier and nicer for everyone to conduct interviews on the ferry rather than a train or bus. We will go to Slussen and take the ferry to document the surroundings and to hopefully find a clear and distinct focus group. We will perform a naturalistic observation to help identify and establish our requirements by collecting qualitative data when observing the ferry population. The next phase is to come up with an interview template, preferably a strucutred or semi-structured one. If the focus group happens to travel in smaller groups the interviews could be done as smaller workshops, but it could dilute the interviews too much.

How should we establish our requirements in a manageable way?
What kind of interview or questionnaire is best suited to collect more data?