Stephen Fickas, Professor

Computer Science Dept.
Deschutes Hall
University of Oregon
Eugene, OR 97403

office: (541) 346-3964
fax: (541) 346-5373
email: fickas@cs.uoregon.edu

                               

A Special Request

I am part of a group that is launching a worldwide software engineering contest for undergrad and master level student teams. If you are interested, or know someone who might be interested, please visit the contest site: http://score.elet.polimi.it/.

Research

• Clinical Requirements Engineering. This project makes a case for integration of requirements engineering (RE) with clinical disciplines. The first paper listed below provides an overview, looking at two examples that employ a clinical RE approach: (1) that of introducing email into the life of a brain-injured individual, and (2) introducing digital darkroom tools into my life. The former uses a Brownfield approach by starting with an existing clinical process, cognitive rehabilitation, and then defining an RE process that fits. The latter uses a Greenfield approach that postulates a new clinical RE process that focuses on the problems some of us have using digital darkroom tools.
  Fickas, S. Clinical Requirements Engineering. Invited paper at the 27th International Conference on Software Engineering (Extending the Discipline track), St. Louis, May 2005 (pdf).

  A more general look at a personal RE process that encompasses the clinical view can be found in these two papers (conference and journal versions):

  Sutcliffe, A., Fickas, S., Sohlberg, M, Personal and Contextual Requirements Engineering, 13th IEEE International Conference on Requirements Engineering, Paris, September 2005 (pdf). Awarded best paper at the 13th IEEE International Conference on Requirements Engineering (RE'05).

  Sutcliffe, A., Fickas, S., Sohlberg, M., PC-RE: a method for personal and contextual requirements engineering with some experience, Requirements Engineering, Mar 2006, Pages 1 - 17 (pdf).

  The project web site contains papers related to the email project: http://www.think-and-link.org. There is also a commercial version of the email system available: http://www.coglink.com

Professor Amy Lobben (Geography) and I have an interest in tailoring navigation assistance to the visually impaired. Our early work has focused on being able to build custom tactile maps. A summary in poster form can be found here (NSF Division of Human Resource Development 2007 Joint Annual Meeting). (On the actual poster, the bottom right corner is a tactile map making it an interactive poster!) Our long term interest is in assessments (e.g., computer-based tests) that can be given to an individual to determine their navigation assistance needs. And then deliver custom and adaptable systems to them. See discussion of "personal RE" above.

One of our findings from the Think-and-Link project is that users often state unachievable goals. However, over time some or all may become within reach. We call these deferred goals. They lead naturally into monitoring: we watch for opportunities to adapt the deployed system to meet deferred goals. We have a paper on this topic:

Fickas, S., Robinson, W., Sohlberg, M, The Role of Deferred Requirements in a Longitudinal Study of Email, 13th IEEE International Conference on Requirements Engineering, Paris, September 2005 (pdf).

Below is a baseline paper in the more general area of requirements monitoring.

Fickas, S., Feather, M.S., Requirements Monitoring in Dynamic Environments, Proceedings of the Second IEEE International Symposium on Requirements Engineering, IEEE Computer Society Press, York, England, March 1995 (pdf). Martin and I are quite honored that this paper was awarded "Most Influential Paper from RE'95" at the 13th IEEE International Conference on Requirements Engineering (RE'05).

The following paper links requirements monitoring with GORE (Goal-Oriented Requirements Engineering):

Feather M, S. Fickas, A. van Lamsweerde, C. Ponsard , Reconciling System Requirements and Runtime Behavior, Proceedings of IWSSD'98: 9th International Workshop on Software Specification and Design, Ise-Shima (Japan), April 16-18 1998, IEEE Computer Society Press (April 1998) (pdf).

We have done some preliminary work on the range of interfaces needed to support customization. This is reported below.

Sutcliffe, A., Fickas, S., Sohlberg, M., Ehlhardt, L., Investigating the usability of assistive user interfaces, Interacting with Computers, Volume 15, Issue 4 , 1 August 2003, Pages 577-602 ( abstract)

The implication of deferring goals is that eventually you will return to them, and hopefully, achieve them. This may require a change to the deployed system, i.e., an adaptation. For the think-and-link system, we have been able to use a parameterized application, turning on and off features. I am working with Bill Robinson on a more sophisticated adaptation architecture for a navigation assistant. A description of Bill's tool, ReqMon, can be found off his web page: http://www.cis.gsu.edu/~wrobinso/papers/. The navigation-assistant project home page is here: www.go-outside.org.
• Wearable and ubiquitous computing. I co-founded the Wearable Computing Lab in 1995. The lab's web page is here: www.cs.uoregon.edu/research/wearables/. The most recent topics from the lab are listed below.

  • We completed a project to track the elderly as they move around their neighborhood. Our goal was to use a standard cell-phone. The paper below discusses the project.
    Michael YL, McGregor EM, Allen J, Fickas S., Observing outdoor activity using global positioning system-enabled cell phones, Smart Homes and Health Telematics: 6th International Conference, ICOST 2008, Ames, IA, USA, June 2008 (LNCS 5120)

  • The lab is involved with a major application: designing a wearable navigation-assistance device. Our first test user population are those that are currently housebound by a cognitive impairment. The project, called GO (Get Outside), is described here: www.go-outside.org. In summer 04 we completed a two year study of the population in terms of focus groups and weekly-meetings of a special-interest group at an assisted-living facility. We are now working on field studies that center on observing real trips with multi-modal devices. Following papers provide info on this project:
    Sohlberg, M., Todis, B., Fickas, S., Hung, P., Lemoncello, R., A profile of community navigation in adults with chronic cognitive impairments, Brain Injury, December 2005; 19(14): 1249-1259 (pdf)

    Fickas, S., Sohlberg, M., Hung, P., Route-following assistance for travelers with cognitive impairments: A comparison of four prompt modes, Int. J. Human–Computer Studies, Volume 66, Issue 12, December 2008, Pages 876-888 (preliminary version)

    Fickas, S., Sohlberg, M., A Community Safety-Net for the Brain-Injured Traveler, CHI 2005 Workshop, Engaging The City: Public Interfaces As Civic Intermediary Portland, 2005 (pdf)

    There is also a report, for the obstacle-based GORE-ites out there, that lays out a set of goals and obstacles for the Intelligent Transportation System (ITS) area (see section 2 and Appendix A, in particular).
    "The goals of the Mobility Services for All Americans (MSAA) initiative are to increase mobility and accessibility for the transportation disadvantaged as well as the general public and to achieve more efficient use of Federal transportation funding resources. In support of MSAA, this report summarizes the Phase Two (Foundation Research) effort, including providing summaries and user assessments of many past and ongoing human services transportation, coordination, and technology initiatives." (Report link)

    Contact fickas@cs.uoregon.edu for more details about the GO project.

  • I am interested in a wearable device that can complement (or replace) a tactile map used by a person with a visual impairment. Find more info in a recent overview article.

  • We have an interest in middleware layers that support context-aware computing. An early effort in this area is the CAT (Context-Aware Toolkit) project . We are working on extending this to a more heterogeneous environment that includes 802.11, Bluetooth, and mote-based devices.

  • Field-testing UBICOMP apps is sometimes painful. We are working on a simulator that will allow some testing to happen in a virtual environment. The simulator project can be found here: iPaq simulation project .

• Can Software Engineering Save Formal Modeling?. Conjecture: both Java programming and formal modeling are hard. A myriad of tools and techniques have grown around Java programming. Why can't some of those same tools and techniques be applied to formal modeling (in concept if not in identity)? To date, we have looked in some detail at one such technique, frameworks. We were able to lift a Java framework to the modeling level, and then use it to instantiate several application models. Our main focus has been on Promela/SPIN. However, we continue to dabble with the same ideas in FSP/LTSA.

  NEWS FLASH: we have declared defeat in our efforts to bring modern software engineering tools and techniques to a higher modeling level. We are now looking at modeling at the Java level using Jave PathFinder (JPF).

• Composite System Design. The general theme of this project is a holistic look at the design of socio-technical systems. We argue that many interesting systems have a mixture of human, hardware and software agents, agreeing to cooperate toward some shared goals. A baseline paper that describes our approach is given below.
  Fickas, S., Helm, R., Automating the design of composite systems, IEEE Transactions on Software Engineering, Vol. 18, No. 6, June, 1992 ( pdf)

  A shorter, light-hearted paper that ties circus trains into the mix.

  Fickas, S., Desert Island Column, Automated Software Engineering Journal, Vol. 1, No. 1, 1993 ( html)

  Two current projects have strong ties into the composite system viewpoint.

  • The Think-and-Link project has been an excellent test bed for our ideas. It became clear early on that email is a quintessential socio-technical system. The project has faced the gamut of problems, including (a) hardware: many of our users do not currently have a computer - we have to define a setup for them that works into their living context, (b) software: the email client and email server, and (c) human stakeholders: the email buddies and careproviders. We found that if we failed to integrate all of these, the deployed system would eventually be abandoned.

  • The GO project focuses on community navigation, again, a profoundly composite-system application. On the social side, we have found the need to work with city officials, the local transportation district, and businesses as agents in the system. On the hardware side, we are forced to look at a range of UBICOMP devices/agents, and the software to integrate them into a coherent navigation system. Our current mix of devices includes (a) wireless iPaqs with GPS, (b) sensor networks (mica2 motes, Intel iMotes), (c) consumer devices using Bluetooth (cell phones), (d) RFID tags/readers, and (e) kiosks and public displays. A current project is to instrument, deeply, one interesting urban route within Eugene. Contact fickas@cs.uoregon.edu for details.

• Operational Envelopes. The base conjecture of this project is that engineered artifacts will not work dependably in every possible environment. When a system is designed, cost/benefit analysis should be applied to determine the likely environmental conditions it will face, and then design and implementation can concentrate on those subset of conditions. This results in an operational envelope for the system: inside the envelope, the system is dependable; outside the envelope, dependability bets are off. The following projects explore this concept.
  • Using forward-engineering, we studied the use of model-checking to explore operational envelopes in an onboard, fault-protection system developed at JPL. The following paper presents the details.
    Feather, S. Fickas, A. Razermera, A., Model-Checking for Validation of a Fault Protection System, Proceedings of IEEE International Symposium on High Assurance Systems Engineering, Boca Raton, 2001 (pdf).

  • Using reverse-engineering, we are interested in recovering an operational envelope from a deployed system. The ROPE project is one current example of this approach.

Teaching

Courses I normally teach (notes below):
CIS 170 - Science of Computing CIS 650 - Graduate Software Engineering HC 209 - Can Computers Think? (Clark's Honors College Course)

I am always interested in sharing notes on the courses I teach. Please drop me a line if you have comments or questions about these courses. As an outline, I can give you the general structure of each course.

• CIS 170 (pre-major). I am using Dewdney as the main text (the 66 excursions book). I mix theory with some practice in the form of algorithmic/procedural programming in Java.

  Students generally liked the topics from Dewdney. I want to also give a plug to David Eck's book, The Most Complex Machine. In particular, I used several of the tools from his textbook's web site, and this was a real hit with students.

  Main concern of students is with the Java programming. Some thought I moved too fast through the programming pieces. Their suggestion is to have a separate lab section that concentrates on programming.

  Other problem is the mixture of programming experience across students. Two camps broke out during the quarter: (1) complete novices, and (2) students taking a Java programming course concurrently. My take is that a lab section could help to level the field a bit.

• CIS 650 (grad). I generally attempt to have a theory track, a research track and a practice track in this course. Most recently, the theory track has focused on formal modeling. I currently use FSP/LTSA (and the Jeff&Jeff; book). The research track focuses on papers in the field. The practice track varies, some years looking at tools, and other years, methodologies.
• HC 209 (Honors College, generally non-science majors). I use Kurzweil's The Singularity Is Near as textbook. Then supplement with many outside readings, e.g., Chinese Room Argument, Bill Joy's The Future Does Not Need Us, Hegel - a guide for the perplexed, etc.

  Students like the Kurzweil book, although many think he is on drugs. One thing that I find interesting is that majority of students share Kurzweil's optimistic view of the future and technology. Seem to believe that people (scientists, government) won't let things get out of hand. Maybe it's just me, but I don't have such a rosy view.

---

Other

Collaborations. I benefit from the collaborators I work with, and would like to acknowledge their contribution.
• I work with Amy Lobben through the Spatial and Map Cognition Research Lab. Amy and I also just started a collaboration with Margaret Sereno through the Lewis Center for NeuroImaging. We are currently working with tactile maps for the vision impaired.
  

• I work with Alistair Sutcliffe and Gerhard Fischer on HCI issues related to both the Think-and-Link project and the GO project.

• I work with Martin Feather , Bill Robinson and Axel van Lamsweerde on issues related to requirements monitoring and GORE.

• I am a member of the IFIP 2.9 Requirements Engineering working group , and gain much insight from interactions there.

• My participation in the Think-and-Link and GO projects has brought me many interesting colleagues outside of computer science. This includes McKay Sohlberg (cognitive rehabilitation), Laurie Elhardt (education/training), Bonnie Todis (long term, qualitative, experimental methodologies), and a host of staff and residents at the assisted-living centers in Eugene.

• I work with Bob Hall on the challenges faced by software engineering in the brave new world of open systems. We jointly held a workshop on the topic: Workshop on Requirements Engineering and Open Systems (REOS) .

• I work with Holly Arrow (Psych) and John Orbell (Poli Sci) on a project to insert task-oriented agents into a (human) group process focusing on clubs. A paper appearing in the Workshop for Knowledge-Based Electronic Markets (AAAI 2000) gives a bit of background on this work: Join the Club .

A glimpse of outside life. The Eugene Water Board shutdown a public fountain in mid-summer. I worked to get it turned back on.
I am claiming victory: third installment of the saga , as reported by columnist Karen McGowen. I am sure there is some moral to the story in here.

My other life as a surfer. I grew up in San Diego, and have to say I miss Blacks. Thinking of starting a club: surfers for software engineering :)



My other life as a musician. I took up the tenor sax in 2005. In summer of 2007, I attended Jazz Camp West. It was good. The clip below is of an open mike session (I'm in front with cap). The three of us sax players (Mike Trottier, me, Dave Needham) met and threw our little combo together at camp. Bruce Williamson (faculty member) helped to arrange the Miles tune, Freddie Freeloader, for us. The pianist, Frank Martin was faculty. The bass player, Stuart Fishman, and the drummer, Daisuke Chihara, were two advanced students (i.e., ringers) we asked to join.


Our open mike performance

You may note that I did not solo during first part. It was hard enough to agree to even be on stage! I did do a little 4 bar solo when we started "trading fours" with the drummer. It was interesting that all three of us thought we screwed up when talking afterwards. But looking at the tape, I think it turned out ok :)

eBay/PayPal scams. I've noticed a hard-to-prevent scam using eBay and PayPal in conjunction. If you are a seller and use PayPal, it is almost impossible to stop the buyer from getting the item and then saying they did not make the purchase. Typically they say the credit card was used without their authorization. The buyer contacts PayPal and PayPal issues a "charge back". Your sales money flows back to the buyer. And the buyer keeps the item. The buyer has 45 days from date of purchase to pull off this scam. Suggestion: don't use PayPal on eBay. I know it's a pain, but use a USPS Money Order.

Joining one of my labs. If you are a student interested in working in one of my labs, contact me by email: fickas@cs.uoregon.edu. If you would like a lab tour, contact one of my lab managers:

Wearable Computing Lab Andrew Fortier (andrew@cs.uoregon.edu) Sensor Network Lab Zebin Chen (zbchen@cs.uoregon.edu) Think-and-Link Lab Jason Prideaux (jprideau@cs.uoregon.edu)