CSE503: Software Engineering

David Notkin
University of Washington, Computer Science & Engineering
Winter 2008

Last Updated: 02/22/2008 03:42 PM

Recent changes:

1. Alloy assignment.
2. Feb 19 slides on Lackwit, Daikon, etc. (better viewed in IE right now...another format coming)
3. I am in Washington DC on Monday-Tuesday February 25-26 for the CRA Board meeting.  Tuesday's class will consist of watching a video of Michael Jackson's keynote address from ICSE 1995.
4. Alloy assignment will be (a) in groups, if you want (maximum of three), (b) on the elevator specification, and (c) due March 6.
5. Notes on analysis.
6. Slides on formal methods and model checking.
7. Slides on design history
8. More papers to read listed below.
9. Summary of "open questions" and initial design slides.
10. My research slides and Marius' configuration research slides.
11. Pro/con documents from debates posted below.
12. Send me "votes" on the debate that you watched but did not participate in.
13. Remember: By  January 22, each of you must individually email to us a set of 3-5 bullets on what you find are key open questions about software engineering/software engineering research after the readings, writings, and debates.
14. Most strikeouts removed.
15. cse503@cs.washington.edu mailing list activated.
16. Course has moved to CSE 203.
17. We will only have two groups for the "debate" day -- the second ("crisis") and third ("empirical"); in class on Thursday people will sign up for preferred groups and I'll make the final assignment later Thursday by email.
18. The day of the "debates", and the due day for the document, is shifted to Thursday January 17 -- two days later.  Some other dates have been adjusted as well.

(Last things first) Final Examination:  By University rule, an instructor is allowed to dispense with a final examination at the scheduled time (10:30-11:50AM, March 17, 2008) only with unanimous consent of the class.  If you prefer to have a final examination for the entire class, you must let me know by 10:30AM before the third lecture (January 15, 2008).

Contact information and office hours: notkin@cs.washington.edu , CSE 542, x5-3798.  I prefer not to have specific office hours, but I am happy to have drop-ins whenever my door is open, and I am usually able to set up meetings on quite short notice to answer questions, have discussions, etc., with individuals or groups.  (If the class prefers to have regularly scheduled office hours, I'm happy to do that.)

TA: Marius Nita (marius@cs.washington.edu)

Location and time: CSE 203, 10:30-11:50AM Tuesday and Thursday

Syllabus and required work:  

Is software engineering really "engineering"?  Should software engineers be licensed?  Is there really a software crisis?  What is software engineering research?  How do we evaluate such research?  Has software engineering improved over the past 40 years?  Why do software systems fail so often?  Why are they so costly?  Isn't "agile" or "aspect-oriented" or "end-user" or "types" or something else actually the silver bullet?  What are some "cool" research results, "hot" industrial tools and approaches, etc.?  The goal of this course is to get you to develop your own rational, thoughtful, well-reasoned answers to these questions -- I already have my own points of view, of course!

The lecture schedule is divided into the following topics, which are by no means intended to be a complete representation of the field.  It is also a fluid list -- in particular, if there is a strong feeling among the class that there are other topics we ought to cover instead, I'll look seriously at those ideas.  And the specifics are subject to change, regardless.

• Overview (and debates): January 7, 9, 15
  • Required readings -- for this set of readings, you needn't read each and every page in these documents, but the core arguments and reasoning are essential
    • Fredrick P. Brooks, Jr.  "No Silver Bullet: Essence and Accidents of Software Engineering", Computer, April 1987.  This is also found in Brooks' Mythical Man-Month (25th Anniversary Edition).
    • The 1968 and 1969 NATO Software Engineering Conferences (Naur and Randell, Randell and Buxton, editors, respectively).  These are long documents, but quite easy (and fascinating) to read.
    • The SWEBOK Guide, Chapter 1, "Introduction to the Guide" (found in several formats at the site).
    • David Notkin, Michael Gorlick, Mary Shaw.  "An Assessment of Software Engineering Body of Knowledge Efforts", A Report to the ACM Council (May 2000).
    • De Millo, R. A., Lipton, R. J., and Perlis, A. J.  Social processes and proofs of theorems and programs. Commun. ACM 22, 5 (May. 1979), 271-280.  http://doi.acm.org/10.1145/359104.359106
    • Walter F. Tichy, Paul Lukowicz, Lutz Prechelt, Ernst A. Heinz. Experimental Evaluation in Computer Science: A Quantitative Study. Journal of Systems and Software 28(1):9-18, January 1995.
  • Required work (10% of total course grade)
    • The class will be divided into three two groups, with each group having two parts, and each part having at least two people.   The groups will focus on the following three two topics:
      • Software engineering is indeed in a crisis. pro document, con document
      • The core evaluation of a software engineering research result must be empirical in nature.  pro document, con document
    • One part of each group will produce a 3-5 page document arguing the "for" side of the topic, and the other part of the group will do the same for the "against" side of the topic.  These brief documents are expected to be based on your readings (above and others, if you wish), scholarly, and to the point; they are due by the beginning of class on  January 17.
    • On January 17, we will have three 30 minute debates on each of the topics: in the debate (details of the format still to be decided), the "for" and "against" parts of each group did not will swap positions.
    • By January 22, each of you must individually email to us a set of 3-5 bullets on what you find are key open questions about software engineering/software engineering research after the readings, writings, and debates.
• Process: January 17
• Project area overviews:  January 22
  • No required readings (subject to change)
  • Presentations in class on areas of particular interest for research projects (software configurations, mining software repositories, etc.)
  • Required work
    • Small group research project (50% of total course grade)
      • In groups of 2-4 (which you can form on your own, with my approval), write a roughly one-page proposal for a small-scale software engineering research project. This is due January 29, but discussions (electronic and otherwise) with Marius and me beforehand are highly recommended.
      • All groups will make presentations to the rest of the class on March 13.
      • Write-ups, roughly in the style of a tech report or a conference paper, are due on March 17 at noon.
      • In general, members of a group will share the same grade, although exceptions aren't excluded for appropriate situations.
• Design: January 24, 29, 31, February 5
  • Parnas, D. L. On the criteria to be used in decomposing systems into modules. Commun. ACM 15, 12 (Dec. 1972), 1053-1058. 
  • David L. Parnas. Designing software for ease of extension and contraction. IEEE Trans. Software Eng. SE-5, 2 (1979).
  • Gregor Kiczales, John Lamping, Anurag Mendhekar, Chris Maeda, Cristina Videira Lopes, Jean-Marc Loingtier, and John Irwin. Aspect-Oriented Programming. Proceedings of the European Conference on Object-Oriented Programming (ECOOP), Finland. Springer-Verlag LNCS 1241. June 1997.
  • Erich Gamma, Richard Helm, Ralph E. Johnson, John M. Vlissides: Design Patterns: Abstraction and Reuse of Object-Oriented Design. ECOOP 1993: 406-431
  • David Garlan and Mary Shaw. An Introduction to Software Architecture.  CMU Tech Report, January 1994.
     
• Specification and Analysis: February 7, 12, 14
  • There will be an assignment using Daniel Jackson's Alloy system, due at the beginning of lecture on February 28 (20% of total course grade)
  • Daniel Jackson and Martin Rinard.  Software analysis: a roadmap. ICSE '00: Proceedings of the Conference on The Future of Software Engineering (2000).

     

• Testing: February 19, 21
• TBA: February 26 (Notkin in Washington DC for CRA Board meeting)
• Tools and environments: February 28, March 4, 6
• So, what did we learn?: March 11
• Project presentations:  March 13
  
• The remaining 20% of the course grade will be based on brief summaries/Q&A about 10 of the papers assigned.