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

Hard Data on XP vs. Traditional Methodologies

Expand Messages
  • davehane
    Greetings XP rs: I am current engaged in non-XP work for a client that doesn t know that they are in the software development business. Delusional issues
    Message 1 of 9 , Mar 31, 2003
    • 0 Attachment
      Greetings XP'rs:

      I am current engaged in non-XP work for a client that doesn't know
      that they are in the software development business. Delusional
      issues aside, it is clear to me that they would benefit greatly from
      apply XP or XP-related principles to their development.

      Making such an observation from the sidelines is easy. Convincingly
      demonstrating that "better mousestraps" are available may be quite
      another story.

      What I am looking for is hard data that compares and contrasts XP vs.
      a heavy weight methodology. My assertion is that the client will be
      able to develop higher quality software in less time if they adopt XP
      or a cousin. Barring hard data, does anybody have strong anectdotal
      evidence they've used to help sell XP projects?

      Thank you, David Hane
    • Ron Jeffries
      ... As you look further, you ll find that there are very few studies on applying any software process. That s because to make a comparison you have to do the
      Message 2 of 9 , Apr 1, 2003
      • 0 Attachment
        On Tuesday, April 1, 2003, at 9:39:28 AM, Harald Svensson wrote:

        > I am looking from results of applying XP, but I am finding very few papers
        > about it. If someone could provide me with empirical data from applying XP,
        > I would really appreciate it.

        > Second question, why are there so few studies regarding applying XP? XP has
        > been around for a couple of years and at least a little more research
        > should have been done to see the effects of applying XP.

        As you look further, you'll find that there are very few studies on
        applying any software process. That's because to make a comparison you have
        to do the project twice. With the same people. Except that they have to
        start out in the same state both times. And lobotomies are expensive and
        imprecise.

        In short ... comparisons are lacking in our business because they are
        almost impossible to do.

        You can find lots of anecdotal evidence of teams that perceived improvement
        after changing their process. And you can learn what's right for you by
        experimenting.

        Ron Jeffries
        www.XProgramming.com
        The greatest mistake we make is living in constant fear that we will make one.
        -- John Maxwell
      • Pollice, Gary
        Actually, not quite true. You don t have to use the same people, but you do have to have enough trials to have a randomized representation of the population.
        Message 3 of 9 , Apr 1, 2003
        • 0 Attachment
          Actually, not quite true. You don't have to use the same people, but you do
          have to have enough trials to have a randomized representation of the
          population. And, you're right, Ron, it's expensive. I'm teaching a class now
          on empirical software engineering and we are running a study/experiment on
          the efficacy of test-first. I hesitate to call it TDD because something that
          can be done in a couple of hours will have minimal design to it. But we can
          get some results on the effect of test-first on different characteristics of
          development.

          No one study is enough. There are a couple of other studies, such as the one
          from Karlsruhe, which was unable to claim any conclusion for TDD, Laurie
          Williams did some work on pair programming, and has been doing some studies
          on TDD.

          There are a couple of other things that make it hard to get results. First,
          in order to see that you're doing better, you need to have data about how
          you've done. Many, if not most, organizations have not done a good job of
          measuring their existing process. Secondly, when they do, they often are not
          willing to make their information public.

          --Gary

          -----Original Message-----
          From: Ron Jeffries [mailto:ronjeffries@...]
          Sent: Tuesday, April 01, 2003 8:54 AM
          To: extremeprogramming@yahoogroups.com
          Subject: Re: [XP] Results from applying XP?


          On Tuesday, April 1, 2003, at 9:39:28 AM, Harald Svensson wrote:

          > I am looking from results of applying XP, but I am finding very few
          > papers
          > about it. If someone could provide me with empirical data from applying
          XP,
          > I would really appreciate it.

          > Second question, why are there so few studies regarding applying XP?
          > XP has
          > been around for a couple of years and at least a little more research
          > should have been done to see the effects of applying XP.

          As you look further, you'll find that there are very few studies on applying
          any software process. That's because to make a comparison you have to do the
          project twice. With the same people. Except that they have to start out in
          the same state both times. And lobotomies are expensive and imprecise.

          In short ... comparisons are lacking in our business because they are almost
          impossible to do.

          You can find lots of anecdotal evidence of teams that perceived improvement
          after changing their process. And you can learn what's right for you by
          experimenting.

          Ron Jeffries
          www.XProgramming.com
          The greatest mistake we make is living in constant fear that we will make
          one.
          -- John Maxwell


          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/
        • Harald Svensson
          I am looking from results of applying XP, but I am finding very few papers about it. If someone could provide me with empirical data from applying XP, I would
          Message 4 of 9 , Apr 1, 2003
          • 0 Attachment
            I am looking from results of applying XP, but I am finding very few papers
            about it. If someone could provide me with empirical data from applying XP,
            I would really appreciate it.

            Second question, why are there so few studies regarding applying XP? XP has
            been around for a couple of years and at least a little more research
            should have been done to see the effects of applying XP.


            Regards
            Harald
          • Mark Stiggs
            I don t get the question. Are you saying they don t know they re in the software development business but they have a heavyweight methodology? That can t be
            Message 5 of 9 , Apr 1, 2003
            • 0 Attachment
              I don't get the question.
              Are you saying they don't know they're in the software development business but they have a 'heavyweight' methodology? That can't be right.
              Or are you saying that they have no methodology but you're going to present both a lightweight approach and a heavyweight approach. Why wouldn't you just start with a lightweight approach and go from there?

              davehane <davehane@...> wrote:Greetings XP'rs:

              I am current engaged in non-XP work for a client that doesn't know
              that they are in the software development business. Delusional
              issues aside, it is clear to me that they would benefit greatly from
              apply XP or XP-related principles to their development.

              Making such an observation from the sidelines is easy. Convincingly
              demonstrating that "better mousestraps" are available may be quite
              another story.

              What I am looking for is hard data that compares and contrasts XP vs.
              a heavy weight methodology. My assertion is that the client will be
              able to develop higher quality software in less time if they adopt XP
              or a cousin. Barring hard data, does anybody have strong anectdotal
              evidence they've used to help sell XP projects?

              Thank you, David Hane



              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/




              ---------------------------------
              Do you Yahoo!?
              Yahoo! Tax Center - File online, calculators, forms, and more

              [Non-text portions of this message have been removed]
            • Ron Jeffries
              ... And thirdly, measuring the process leads almost inevitably to improving it, if there s anyone in the room with a brain and interest in quality -- and there
              Message 6 of 9 , Apr 2, 2003
              • 0 Attachment
                Around Tuesday, April 1, 2003, 9:01:56 AM, Pollice, Gary wrote:

                > There are a couple of other things that make it hard to get results. First,
                > in order to see that you're doing better, you need to have data about how
                > you've done. Many, if not most, organizations have not done a good job of
                > measuring their existing process. Secondly, when they do, they often are not
                > willing to make their information public.

                And thirdly, measuring the process leads almost inevitably to improving it,
                if there's anyone in the room with a brain and interest in quality -- and
                there probably is. Drawing the team's attention to any important metric is
                the first step, and often the only necessary step, to improving it. The
                processes available amount to a menu of techniques for improving metrics.

                Need fewer defects? Try more testing, try test-driven development, try
                pair programming, try inspections, try code reviews, try design by
                contract. Any of these will help. Paying a little more attention will
                help. It's all good. Some's maybe better, quite possibly it's mostly a
                matter of taste and preferred style.

                Ron Jeffries
                www.XProgramming.com
                You are closing your eyes to a situation you do not wish to acknowledge.
                --Professor Harold Hill
              • Harald Svensson
                I see your point, and yes prefferably you should do the exact same thing but using two different processes, however, some general indications of improvement
                Message 7 of 9 , Apr 2, 2003
                • 0 Attachment
                  I see your point, and yes prefferably you should do the exact same thing
                  but using two different processes, however, some general indications of
                  improvement could surely be concluded, and that´s always a start. You just
                  have to be sure that you do not state something that you can not stand for.

                  I welcome even somewhat questionable studies, present results etc.


                  Harald


                  At 09:01 AM 4/1/03 -0500, you wrote:
                  >Actually, not quite true. You don't have to use the same people, but you do
                  >have to have enough trials to have a randomized representation of the
                  >population. And, you're right, Ron, it's expensive. I'm teaching a class now
                  >on empirical software engineering and we are running a study/experiment on
                  >the efficacy of test-first. I hesitate to call it TDD because something that
                  >can be done in a couple of hours will have minimal design to it. But we can
                  >get some results on the effect of test-first on different characteristics of
                  >development.
                  >
                  >No one study is enough. There are a couple of other studies, such as the one
                  >from Karlsruhe, which was unable to claim any conclusion for TDD, Laurie
                  >Williams did some work on pair programming, and has been doing some studies
                  >on TDD.
                  >
                  >There are a couple of other things that make it hard to get results. First,
                  >in order to see that you're doing better, you need to have data about how
                  >you've done. Many, if not most, organizations have not done a good job of
                  >measuring their existing process. Secondly, when they do, they often are not
                  >willing to make their information public.
                  >
                  > --Gary
                  >
                  >-----Original Message-----
                  >From: Ron Jeffries [mailto:ronjeffries@...]
                  >Sent: Tuesday, April 01, 2003 8:54 AM
                  >To: extremeprogramming@yahoogroups.com
                  >Subject: Re: [XP] Results from applying XP?
                  >
                  >
                  >On Tuesday, April 1, 2003, at 9:39:28 AM, Harald Svensson wrote:
                  >
                  > > I am looking from results of applying XP, but I am finding very few
                  > > papers
                  > > about it. If someone could provide me with empirical data from applying
                  >XP,
                  > > I would really appreciate it.
                  >
                  > > Second question, why are there so few studies regarding applying XP?
                  > > XP has
                  > > been around for a couple of years and at least a little more research
                  > > should have been done to see the effects of applying XP.
                  >
                  >As you look further, you'll find that there are very few studies on applying
                  >any software process. That's because to make a comparison you have to do the
                  >project twice. With the same people. Except that they have to start out in
                  >the same state both times. And lobotomies are expensive and imprecise.
                  >
                  >In short ... comparisons are lacking in our business because they are almost
                  >impossible to do.
                  >
                  >You can find lots of anecdotal evidence of teams that perceived improvement
                  >after changing their process. And you can learn what's right for you by
                  >experimenting.
                  >
                  >Ron Jeffries
                  >www.XProgramming.com
                  >The greatest mistake we make is living in constant fear that we will make
                  >one.
                  > -- John Maxwell
                  >
                  >
                  >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/
                  >
                  >
                  >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/
                  >
                • George Dinwiddie
                  ... Have you looked through the archives of this list? Have you looked at the articles published on http://www.agilealliance.com/articles/index ? - George --
                  Message 8 of 9 , Apr 2, 2003
                  • 0 Attachment
                    Harald Svensson wrote:
                    > I see your point, and yes prefferably you should do the exact same thing
                    > but using two different processes, however, some general indications of
                    > improvement could surely be concluded, and that´s always a start. You just
                    > have to be sure that you do not state something that you can not stand for.
                    >
                    > I welcome even somewhat questionable studies, present results etc.

                    Have you looked through the archives of this list? Have you looked at
                    the articles published on http://www.agilealliance.com/articles/index ?

                    - George

                    --
                    -------------------------
                    George Dinwiddie
                    agile programmer for hire
                    Baltimore/Washington area
                    gdinwiddie@...
                    -------------------------
                  • Harald Svensson
                    No I had not, thank you very much for telling me. Harald
                    Message 9 of 9 , Apr 3, 2003
                    • 0 Attachment
                      No I had not, thank you very much for telling me.


                      Harald


                      At 11:15 AM 4/2/03 -0500, you wrote:
                      >Harald Svensson wrote:
                      > > I see your point, and yes prefferably you should do the exact same thing
                      > > but using two different processes, however, some general indications of
                      > > improvement could surely be concluded, and that´s always a start. You just
                      > > have to be sure that you do not state something that you can not stand for.
                      > >
                      > > I welcome even somewhat questionable studies, present results etc.
                      >
                      >Have you looked through the archives of this list? Have you looked at
                      >the articles published on http://www.agilealliance.com/articles/index ?
                      >
                      > - George
                      >
                      >--
                      > -------------------------
                      > George Dinwiddie
                      > agile programmer for hire
                      > Baltimore/Washington area
                      > gdinwiddie@...
                      > -------------------------
                      >
                      >
                      >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/
                      >
                    Your message has been successfully submitted and would be delivered to recipients shortly.