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

RE: [XP] Before and After metrics

Expand Messages
  • Hubert Smits
    John, If you go to the articles section of the Agile Alliance and search for metrics, then you find some info. Not huge but there is material available.
    Message 1 of 6 , Dec 31, 2002
    • 0 Attachment
      John,

      If you go to the articles section of the Agile Alliance and search for
      metrics, then you find some info. Not huge but there is material
      available.

      --Hubert

      | -----Original Message-----
      | From: cpp gent [mailto:cppgent0@...]
      | Sent: 30 December 2002 21:45
      | To: extremeprogramming@yahoogroups.com
      | Subject: [XP] Before and After metrics
      |
      |
      | Does anyone have metrics before and after introducing XP?
      |
      | The metrics can be anything, although I'd appreciate enough
      | detail so I can
      | replicate them. Any thoughts you have concerning which
      | metrics were most
      | effective and least effective would be greatly appreciated as well.
      |
      | TIA,
      | John
      |
      | _________________________________________________________________
      | MSN 8 limited-time offer: Join now and get 3 months FREE*.
      | http://join.msn.com/?page=dept/dialup&xAPID=42&PS=47575&PI=732
      4&DI=7474&SU=
      http://www.hotmail.msn.com/cgi-bin/getmsg&HL=1216hotmailtaglines_newmsn8
      ishere_3mf


      To Post a message, send it to: extremeprogramming@...

      To Unsubscribe, send a blank message to:
      extremeprogramming-unsubscribe@...

      ad-free courtesy of objectmentor.com

      Your use of Yahoo! Groups is subject to
      http://docs.yahoo.com/info/terms/
    • cpp gent
      From: Hubert Smits ... The single article under the metrics category is excellent, thanks. A search for metrics or metric
      Message 2 of 6 , Dec 31, 2002
      • 0 Attachment
        From: "Hubert Smits" <hubert.smits@...>
        >If you go to the articles section of the Agile Alliance and search for
        >metrics, then you find some info. Not huge but there is material
        >available.

        The single article under the metrics category is excellent, thanks. A search
        for "metrics" or "metric" returned nothing however.










        _________________________________________________________________
        Add photos to your e-mail with MSN 8. Get 3 months FREE*.
        http://join.msn.com/?page=features/featuredemail&xAPID=42&PS=47575&PI=7324&DI=7474&SU=
        http://www.hotmail.msn.com/cgi-bin/getmsg&HL=1216hotmailtaglines_addphotos_3mf
      • Paul Hodgetts
        ... We gathered a set of relatively simple metrics, including elapsed time, developer effort, defects, and code quality, for two similar projects, one before
        Message 3 of 6 , Dec 31, 2002
        • 0 Attachment
          John wrote:

          > Does anyone have metrics before and after introducing XP?
          >
          > The metrics can be anything, although I'd appreciate enough detail so I can
          > replicate them. Any thoughts you have concerning which metrics were most
          > effective and least effective would be greatly appreciated as well.

          We gathered a set of relatively simple metrics, including elapsed
          time, developer effort, defects, and code quality, for two similar
          projects, one before and one after XP. We wrote up the results as
          a paper, which was eventually published in Extreme Programming
          Perspectives (ISBN 0-210-77005), Chapter 30, Extreme Adoption
          Experiences of a B2B Start-Up. The paper is also available at
          Agile Logic's web site (http://www.agilelogic.com) in the resources
          section.

          I feel the most effective metrics are to track actual task effort
          put into the project. If possible track it both as initial
          development, and then as the result of rework due to defects or
          other causes. These metrics allow some overall cost comparisons,
          and tend to show pretty clearly that a team doing XP tends to
          deliver better product initially and saves total effort over the
          long-term.

          In my experience, it will appear at first that the tasks take more
          effort under XP, since the initial total effort is more. This is
          especially true if tracked in terms of individual developer-hours
          due the doubling of developer-hours when pairing. But the rework
          effort turns out to be dramatically less, and the resulting total
          project effort is less.

          The big problem is whether these metrics exist for prior non-XP
          projects, because it seems people's memories tend to recall things
          went better than they did for the non-XP projects. "We could have
          done that without XP." "Maybe, but that non-XP project looks like
          it cost 30% more developer-hours, and delivered two months later."

          Regards,
          Paul
          -----
          Paul Hodgetts -- President, Principal Consultant
          Agile Logic -- www.agilelogic.com
          Consulting, Coaching, Training -- On-Site & Out-Sourced Development
          Java, J2EE, C++, OOA/D -- Agile Methods/XP/Scrum, Use Cases, UI/IA
        • cpp gent
          Here are some links (via google) that have metrics: http://www.iona.com/devcenter/appserv/XP_Orbix.pdf http://fc-md.umd.edu/projects/Agile/Maurer.htm
          Message 4 of 6 , Dec 31, 2002
          • 0 Attachment
            Here are some links (via google) that have metrics:

            http://www.iona.com/devcenter/appserv/XP_Orbix.pdf
            http://fc-md.umd.edu/projects/Agile/Maurer.htm
            http://sern.ucalgary.ca/~milos/papers/2002/MaurerMartel2002d.pdf
            http://sern.ucalgary.ca/~milos/papers/2002/MaurerMartel2002c.pdf
            http://collaboration.csc.ncsu.edu/laurie/Papers/ieeeSoftware.PDF

            I found many more articles with metrics but they focused on software or OOD
            metrics. I'm looking for business-level metrics pre- and post-XP
            introduction.


            _________________________________________________________________
            MSN 8 limited-time offer: Join now and get 3 months FREE*.
            http://join.msn.com/?page=dept/dialup&xAPID=42&PS=47575&PI=7324&DI=7474&SU=
            http://www.hotmail.msn.com/cgi-bin/getmsg&HL=1216hotmailtaglines_newmsn8ishere_3mf
          • cpp gent
            From: Paul Hodgetts ... Thanks Paul, I ll add it to my list of sites... ... Which is exactly my concern and thus my question. ... Do
            Message 5 of 6 , Jan 2, 2003
            • 0 Attachment
              From: Paul Hodgetts <phodgetts@...>
              >The paper is also available at Agile Logic's web
              >site (http://www.agilelogic.com) in the resources section.
              Thanks Paul, I'll add it to my list of sites...

              >The big problem is whether these metrics exist
              >for prior non-XP projects, ...

              Which is exactly my concern and thus my question.

              >In my experience, it will appear at first that the
              >tasks take more effort under XP, since the initial total
              >effort is more. This is especially true if tracked in terms of individual
              >developer-hours due the doubling of developer-hours when pairing. But the
              >rework effort turns out to be dramatically less, and the resulting total
              >project effort is less.

              Do you tell your clients up-front about these trends?

              >I feel the most effective metrics are to track actual task effort put into
              >the project. If possible track it
              >both as initial development, and then as the result of
              >rework due to defects or other causes.

              Initial effort + rework or bugfix time sounds like a good measure. It seems
              to beg the question though: How long do you wait for the bugs to come in
              before you stop counting bugfix/rework time?

              I guess the answer depends on how long the full cycle from release to user
              actually trying the feature to finally generating a bug report. If it's very
              long, then gathering post-XP metrics must wait until at least that long.

              >These metrics allow some overall cost comparisons,
              >and tend to show pretty clearly that a team doing XP
              >tends to deliver better product initially and saves
              >total effort over the long-term.
              Is it strictly that Total effort = intial development + rework or bugfixes,
              or are there other quantities in total effort that you see?


              John



              _________________________________________________________________
              Add photos to your e-mail with MSN 8. Get 2 months FREE*.
              http://join.msn.com/?page=features/featuredemail
            Your message has been successfully submitted and would be delivered to recipients shortly.