Gold University of Minnesota M. Skip to main content.University of Minnesota. Home page.
 

What's inside.

Upcoming Meetings

Previous Presentations

Calendar of Events

Become a Sponsor

Sponsors

TwinSPIN News

Related Links

Contact Us

Join our Mailing List

 

Twin-SPIN Home

 
 
UMSEC: University of Minnesota Software Engineering Center
 
Twin-SPIN
Twin Cities Software
Process Improvement Network
 

Adding Bits of Precision to Use Cases

October 3, 2002

Location: Electrical Engineering/Computer Science Bldg, Room EE/Csi 3-180

Thursday, 03 October, 2002
5:45-8:00 p.m. at The University of Minnesota
5:45 start for networking, 6:15 start for meeting
Electrical Engineering/Computer Science Bldg
Room EE/Csi 3-180
Minneapolis, MN

Abstract:

Sometimes a few sentences or paragraphs are sufficient to describe usage, but sometimes more details are needed.  Alistair Cockburn introduced the notion of "levels of precision in functional requirements".  In a subsequent wiki discussion, he wrote about 1-bit through 6-bit precision corresponding to increasing information content.  In his scheme for use cases, 1-bit precision names the goal, 2-bit adds the main scenario, 3-bit adds failure conditions, 4-bit adds failure actions, 5-bit adds data descriptions, and 6-bit adds user models.

In the spirit of agile methods and "just enough" written communication, user activities that are well understood could be marked by XP user stories (1.5-bits of precision?), while in cloudier areas (e.g., high complexity) some of the misunderstanding risk should be managed with descriptions of higher precision (i.e., more details).  The project team should decide how much information to write for each user need.  Sometimes this decision will be incremental as discussion reveals unanticipated complexity or misunderstanding.

This presentation will describe and illustrate by examples eight information candidates that you can use to increase the precision of your use cases.  Each candidate can be used alone or in combination to supplement your current practices.

  • valid input alternatives
  • alternative system actions
  • embedded post-conditions
  • invariant case conditions
  • exception conditions - the unless clause
  • abstract conditions - e.g., request is valid
  • class constraints
  • dependencies between conditions
These candidates are included in a specification style called Precise Use Cases.   A medium size example of this style for a library management system can be found at http://www.StickyMinds.com.  The paper upon which this presentation is based can also be found on http://www.StickyMinds.com.

About the Speaker:

David Gelperin (dave@livespecs.com) is currently CTO & President of LiveSpecs Software in Minneapolis, MN.  He has more than 35 years experience in software engineering with an emphasis on software quality, verification, and test (SQVT) including software process engineering.

Dave has been a SQVT consultant/mentor and instructor (20 yrs), quality support manager (5 yrs), verification lead (2 yrs), project lead (2 yrs), and programmer (5 yrs).   He has consulted for both commercial and in-house software development organizations.

Dave cofounded Software Quality Engineering (www.sqe.com, www.stickyminds.com) -- the leading provider of software quality information, worldwide -- in 1986 and catalyzed the launch of STQE (Software Test and Quality Engineering) magazine.

In addition, Dave chaired the development of both ANSI/IEEE standards on software testing - 829 on software test documentation and 1008 on software unit testing.

Dave is chief architect of:

  1. the LiveSpecs Reference Model for Requirements Activities
  2. the Better Requirements Platform(tm) & Product Explorer(tm)
  3. Precise Usage Models(tm) and Precise Use Cases(tm)
  4. a Testability Support model
  5. High-Impact (tm) technical reviews
  6. the Systematic Test & Evaluation Process (STEP (tm)) test methodology and
  7. the Unique Cause & Pre-emptive Debugging test strategies.
David received a Ph.D. (1973) and MS (1970) in Computer Science from the Ohio State University, after majoring in math at Carleton College (1964).

 
The University of Minnesota is an equal opportunity educator and employer.