Effective Reduction of Avoidable Complexity
in Embedded Systems Experiment

Call for Participation

You are invited to participate in a series of short experiments to explore new measures of complexity in software models and to evaluate quality, productivity, and modeling tool usage outcomes in the context of complexity.

If you are interested in participating, please provide your contact information and answer a few questions about your professional experience.

Submit Information

Research Purpose

Question 1

What metrics in software models can be used to identify avoidable complexity? 

Question 2

How do novice and expert software modelers introduce complexity when designing a system, and how may this complexity be automatically detected and removed?

Procedure

Participants will remotely log in to participate in the experimental session. The experiments will be conducted in two phases.

Phase 1

Phase 1 will consist of participants operating under timed "test-taking" conditions (approximately 2 hours) to evaluate and/or modify specific software models.

Phase 2

Phase 2 will consist of participants operating under timed "test-taking" conditions (approximately 4 hours) to begin with a provided problem description and then to develop a software model.

Free Training

Prior to Phase 1, our research team will provide each participant with approximately 1 day of free training on the use of a software modeling tool called SCADE, by ANSYS, and instructions on how to install it. A set of duplicate 1-day training sessions will be scheduled to enable participants to choose one of the sessions. They have been tentatively set for

  • Monday, July 27
  • Wednesday, July 29
  • Monday, August 3

As a condition of providing the free training, the SCADE tool vendor, ANSYS, has asked for the contact information of all participants so that they may conduct training follow-up beyond our short training evaluation survey. 

Confidential Exercises 

Each training session will be followed by a single day of confidential exercises as part of the research experiment. They have been tentatively set for

  • Tuesday, July 28
  • Thursday, July 30
  • Tuesday, August 4
Stress Level

Stress will be minimal; we will inform the study subjects that they may find some of the exercises difficult or frustrating. That is perfectly normal. It will not be taken as a negative judgment of their abilities, and it will have no adverse effect of any kind on evaluation of their status as students or working professionals.

Evaluation

Our research team will separately and confidentially evaluate the participants' activities, measures, and modeling results for data recording and determining quality and productivity of the tasks.

Participants

We are soliciting up to 70 participants of 18 years or older, from two different communities:

  • students studying computer science at a college or university
  • industry and government professionals from the software development domain

We would like a mixture of participants who do and do not have background in software/systems behavioral or state transition modeling.

Benefits of Participation

We anticipate that all participants will first gain reinforced knowledge of software modeling and a related tool set (via the 1-day prerequisite training provided by the research team). Second, we anticipate that participants will better appreciate the notion of software complexity and the dimensions of avoidable complexity. We will also share information with the participants after the experiment on examples of avoidable complexity and actions to better avoid such complexity within their work.

Confidentiality of Results

Data will be coded (de-identification) and secured on a secure server; identification records will be maintained separately from experimental results to allow researchers to trace experimental subject participation. Participants taking part in the experiment will not have access to other participants' results. We will remind participants that this study may help them improve their ability to conduct software modeling in terms of quality and productivity.

Disclosure

As a condition of being accepted as a participant in the training and exercises, you agree that your contact information will be provided to the SCADE tool vendor, ANSYS, for followup beyond our short training evaluation survey.  You also commit to participating in the exercises that follow the 1 day of free training.