Skip to main contentCal State San Bernardino
>> [CNS] >> [Comp Sci Dept] >> [R J Botting] >> [New Bibliographic Items] >> newb0515 [Blog/News] || [Purpose] || [Notation] || [Copyright] || [Site Search] || [Bibliography Search]
Mon May 16 15:19:05 PDT 2005

Contents

    GoldMohanLayzell05

    1. Nicholas E Gold & Andrew M Mohan & Paul J Layzell
    2. Spatial Complexity Metrics: An Investigation of Utility
    3. IEEE Trans Software Engineering V31n3(Mar 2005)pp203-212
    4. =EXPERIENCE METRICS 30 COBOL TECHNICAL LoC
    5. Spatial Complexity Metrics allow for the distance between references to identifiers.
    6. Two out of three such metrics are no better than Lines of Code at predicting complexity on this sample.
    7. Sample is of single file COBOL programs, many versions.
    8. A metric worth study: Douce Basic(1999). Sum[each call] (number of lines of code from call to definition)

    ErdogmasMorisioTorchiano05

    1. Hakan Erdogmas & Maurizio Morisio & Marco Torchiano
    2. On the Effectiveness of the Test-First Approach to Programming
    3. IEEE Trans Software Engineering V31n3(Mar 2005)pp226-237
    4. =EXPERIMENT TESTING AGILE TFD TDD
    5. Summarizes previous studies: test-first makes no difference or improves quality and may or may not improve productivity.
    6. This paper has a controlled experiment comparing test-first with test-last development of a series of user stories.
    7. Test-first programmers broke stories into a series of tests and added one test and then added code to make program pass test. Test last coded complete story and then carried out all tests.
    8. Experiment: 40PCs Java JUnit ECLIPSE CVS, 35 3rd year volunteer students, 11 dropped out.
    9. Quality measure via hidden acceptance tests.
    10. Productivity in terms of number of stories per unit time.
    11. Theories: Test-first may tend to increase the number of tests, the quality, and productivity. However more tests anyway may increase quality and productivity.
    12. My translation of their statistics Interesting/significant Results:
      1. quality >= 0.55 + tests/10. NOTE: not a linear equation but an inequality.
      2. More tests increase the minimum number of test passed in the acceptance testing. Other factors can improve quality as well as tests.
      3. productivity= 0.255 + 0.659 tests.
      4. Test-first programmers did more tests on average.
      5. Test-first programmers where more productive and produced the same quality code.
      6. 30% dropped out -- "morbidity"

End