Loading ...
Sorry, an error occurred while loading the content.
 

Re: Direct evidence, XP reduces the cost-of-change curve...

Expand Messages
  • Lee
    ... === How then can you explain the modules passing the units tests if they had a serious flaw? Secondly, would this flaw have been missed if you doing 100%
    Message 1 of 32 , May 1, 2001
      --- In extremeprogramming@y..., Jiva DeVoe <jiva@o...> wrote:

      > But, all of this is not the truly impressive part. Last week, as we
      > were reviewing things, we discovered a serious design flaw in the
      > system which went back, clear to the VERY first modules we wrote
      > way-back-when we started. These modules were absolutely CORE to the
      > rest of the system, and were used in several other modules VERY
      > intimately. BUT... we decided... it had to go.

      === How then can you explain the modules passing the units tests if
      they had a serious flaw? Secondly, would this flaw have been
      missed if you doing 100% pair programming???? What about the
      process of refactoring? Wouldn't this too have surfaced the flaw?


      > We paired for some of the day, and worked alone during other
      parts. We also did NOT write *as many* *new* tests as we should have
      that day. We did write some, but not as many as we could have. We're
      fixing the latter there this week. However, I just wanted to share
      that with you guys as some PROOF that actually doing these things DOES
      reduce the cost of later changes.

      === Pair programming is essential as is writing tests before coding.
      While some of your practices implemented portions of XP, I'm not
      quite certain the full intent of XP was implemented in your
      project. You admittedly did some NON-pair programming, which lends
      me to believe that refactoring also did not occur. The fact
      that your newly developed passed tests is great, but is also
      suspect since there were not an equal proportion of new tests
      developed for the new code being written.

      On a different note, some of the data you provide does lend some
      added credibility to the XP process and its overall
      effectiveness. I'm sure the community thanks you for the
      information.

      Lee in Las Vegas
    • Ron Jeffries
      ... You didn t get it from me, son ... Ronald E Jeffries http://www.XProgramming.com http://www.objectmentor.com
      Message 32 of 32 , May 3, 2001
        Responding to azami@... (11:23 PM 5/3/2001 +0000):

        >Or I'm just contrarian because there's some French blood in me...

        You didn't get it from me, son ...

        Ronald E Jeffries
        http://www.XProgramming.com
        http://www.objectmentor.com
      Your message has been successfully submitted and would be delivered to recipients shortly.