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

Re: [XP] Honest inquiry

Expand Messages
  • Alan C Francis
    For all concerned, I already posted the explanation way back in direct response to Kent. CHeck your archives, folks. A. ... From: Doug Swartz
    Message 1 of 23 , Jun 1 10:40 AM
    • 0 Attachment
      For all concerned, I already posted the explanation way back in direct
      response to Kent. CHeck your archives, folks.

      A.
      ----- Original Message -----
      From: "Doug Swartz" <daswartz@...>
      To: <extremeprogramming@yahoogroups.com>
      Sent: Saturday, June 01, 2002 5:50 PM
      Subject: Re: [XP] Honest inquiry


      >
      >
      > Robert Martin UncleBob wrote:
      > >
      > >
      > > Kent, I'm dying to know what horrible things you said about Chuck and
      LSF.
      > >
      >
      > I directly asked Ann A. what Kent was apoligizing for. She couldn't
      > recall that he was in a "snit" at all.
      >
      > Hmmmm. I want more info. You'd think if my son can download a copy of
      > Star Wars off the net days before its first release, that someone would
      > have a purloined video posted of "Kent in a snit".
      >
      > Doug Swartz
      > daswartz@...
      >
      > 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/
      >
      >
    • Michael D. Hill
      ... Actually, I don t believe you *do* owe me an apology. You were setting me up, of course, and you slightly mis-represented my views, perhaps by being a bit
      Message 2 of 23 , Jun 2 7:36 PM
      • 0 Attachment
        At 06:39 AM 5/31/2002 +0000, you wrote:
        >First, I owe a public apology to Alan Francis and Mike Hill for using
        >them as an example during my closing snit at XP2002. I should have
        >worked harder to find a positive way of saying what I wanted to say.

        Actually, I don't believe you *do* owe me an apology.

        You were setting me up, of course, and you slightly
        mis-represented my views, perhaps by being a bit too
        disinterested in interviewing me to identify them.

        The quote, however, "I'll pay attention to this when
        I see him develop a mathematical formalism that
        expresses Word for Windows.", is accurate, and even
        literal. Furthermore, you were certainly correct to
        infer that I have basically discarded Professor Parnas'
        views, and would need a great deal of convincing to
        change my mind. I have two grounds for dismissing
        his work, one social/personal, and one technical.

        First, I believe the professor demonstrated high
        rudeness, by pre-empting other speakers, by rejecting
        nearly every questioner's sincere attempt to investigate
        his work, by self-advertising throughout the sessions,
        and most of all, by clearly having conducted not even
        the most rudimentary investigation into what XP
        is about. A counter-quote, if you'll allow me: "I just
        don't see the value of having two people sitting
        there staring at one screen." Were I to be invited
        to speak to a conference of Parnas'ists, even as a
        sort of point/counterpoint figure, I genuinely
        believe it would be incumbent upon me to thoroughly
        and thoughtfully interrogate and understand the
        Parnas literature and experience reports. The casual
        rejection I made of his work in semi-private bears
        no comparison with the casual rejection of our work
        that he made in front of our community. It offends
        me.

        Second, far from being a new and exciting direction
        for software development, I found the bulk of his
        ideas old-school in the extreme. While accepting
        every element of the XP critique of its predecessors,
        he didn't validate even a single practice as appropriate.
        Instead, he seemed to merely be repeating the tired
        old saw about how we just weren't taking the necessary
        steps to "get it right the first time". This advice is
        just about as old-school as it gets, if you ask me.
        I mean, seriously, how different is this than the advice
        I've been receiving for the past twenty years? As for
        the mathematical formalisms, maybe you're right, maybe
        they *are* a form of test-driven development. If you make
        the case (in more than a throw-away sentence), I'll
        study it and decide then. For now, I stand by my
        quote.

        As for the setup? Oh baby, you know me well enough
        to know that I wouldn't hesitate to set *you* up in a
        public forum in order to make an important point to a
        large group.

        Anyway, I love you. And I honor you for having the balls
        to apologize. But I deny the need. We cool.

        Seeya!
        Hill


        P.S. I thought the Bache's stuff was pretty cool, tho
        interestingly, Frank Westphal kinda baited me as I was
        looking at their poster, maybe thinking I wasn't gonna
        like it. I *did* like it. *Do* like it. They're not the
        first to suggest driving development using AT's only,
        as I recall: Ward's proposed it, nu?

        P.P.S. The overriding point you were trying to make, about
        openness in the community to new and interesting ideas,
        is well taken. One good paper idea I took away from the
        conference. "XP and the Computable Grail: Why We
        Should Stop Making Rules and Start Making Code."

        +--------------------------------------------------------------
        |Michael Hill
        |Software-> Developer, Consultant, Teacher, Coach
        |Lifeware-> Egghead, Romantic, Grandpa, Communitarian
        |<uly@...>, <Hill@...>
        | www.objectmentor.com, www.pairprogramming.com, www.junit.org
        +--------------------------------------------------------------
      • Piergiuliano Bossi
        Sorry for being a little late on this thread, but my son is born on Saturday and I was pretty crazy and busy! I agree with you: applying XP is not a matter of
        Message 3 of 23 , Jun 3 12:23 AM
        • 0 Attachment
          Sorry for being a little late on this thread, but my son is born on Saturday and I was
          pretty crazy and busy!

          I agree with you: applying XP is not a matter of checking boxes about the practices.
          During the panel about "The 13th Discipline or the Missing Practice(s)?" there was a
          discussion about using ideal time versus complexity units in order to handle team
          velocity and estimations. As you know, we took another approach, based on Francesco
          Cirillo's ideas (see the green book "Planning XP" and
          <http://www.communications.xplabs.com/contents/paper2001-2.html> for more information),
          that is to track down estimates and efforts in small chunks of real time (30 minutes
          period, or rather 25 minuts of work + 5 minutes of daydreaming), called "pomodori"
          (tomatoes in Italian).

          I guess that in order to improve we have to constantly assess the value of our
          practices. This is why I don't believe in concept like "ideal time" and I do prefer
          working only with real time in terms of pomodori. Maybe this is not simpler at a first
          glance, because it requires discipline; but I'm convinced that it gives an incredible
          amount of feedback, with which you can effectively manage the process. Moreover, this
          makes my life easier and it permits to handle my time with more serenity.

          I think that we should clearly constantly work in order to make applying XP simpler,
          but I also think that there are no real shortcuts. For this reason I guess we should
          always focus on the values, instead of the practices. Every little or big thing that we
          do should be evaluated in terms of values. Is it simple? Does it give me feedback? And
          so on ...

          Our current project is the 4th tracked down this way. It is a financial application
          that redistributes market and portfolio data to the so-called "personal bankers" of a
          major financial institution. The project began in September 2001 and it has reached the
          21st iteration during these days. We have approximately 26 KLOC and 400 classes. It's
          not a big project (I guess it will end in a couple of months) but I guess it is
          meaningful somehow. The most important thing is that through tracking we are learning
          to estimate and to manage velocity changes. It is a sloooooow process, but it works.

          Giuliano

          kentlbeck wrote:

          > First, I owe a public apology to Alan Francis and Mike Hill for using
          > them as an example during my closing snit at XP2002. I should have
          > worked harder to find a positive way of saying what I wanted to say.
          >
          > Here's what set me off. The Bache's had a poster about how they got
          > along without unit tests-- http://www.xp2002.org/atti/Bache-Bache--
          > Onesuiteofautomatedtests.pdf. My reaction was to say, "Interesting. I
          > wonder if there is something different about their situation, or if I
          > have been hanging on too tight." They reported that they took a lot
          > of grief for not writing unit tests. Now, they might just have been
          > overly sensitive, but it's also possible that checking off all the XP
          > boxes has become more important to some people than continuing to
          > learn better ways of developing.
          >
          > I would love to make XP simpler. Natural emergent systems get by with
          > far fewer rules than we have, so I'm sure there are opportunities to
          > improve XP by eliminating or consolidating practices.
          >
          > I'm not sure this ended up any more positive than my original black
          > hat statement. However, I do owe Alan and Mike an apology.
          >
          > Thoughts?
          >
          > Kent
          >
          > 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/
        • Ron Jeffries
          ... From your excellent note, I m not clear whether you have studied Parnas ideas and/or used them. In fact I have, though not recently, and I ll report what
          Message 4 of 23 , Jun 3 4:26 AM
          • 0 Attachment
            Around Sunday, June 2, 2002, 10:36:31 PM, Michael D. Hill wrote:

            > As for
            > the mathematical formalisms, maybe you're right, maybe
            > they *are* a form of test-driven development. If you make
            > the case (in more than a throw-away sentence), I'll
            > study it and decide then. For now, I stand by my
            > quote.

            From your excellent note, I'm not clear whether you have studied
            Parnas' ideas and/or used them. In fact I have, though not recently,
            and I'll report what I got from them.

            My intention here is to express that
            - Our forebears deserve our respect
            - Parnas' ideas are well worth knowing
            - Parnas' ideas are /not/ a form of test-driven development
            - Ancient ideas may be XP compatible
            - Yet XP is different, perhaps in a discontinuous way

            Our forebears deserve our respect.

            Dr Parnas deserves our respect, as do all the giants who have
            gone before us, for the work and the contributions he has
            made, even if he wasn't courteous enough to study our work
            before speaking at our conference. If we want to reject his
            work, let's be more courteous than he was, and study it first.

            I guess I'll dig his book out of the cellar -- I think I know
            exactly where it is -- and take another look. If I do that,
            and change my opinion, I'll report back. For now, this is my
            view:

            Parnas' ideas are well worth knowing.

            Parnas' approach, basically, is to use invariants and
            assertions to write programs which are visibly, even provably,
            correct. It is a powerful technique, and one that tends to
            create simple code that is quite functional (in the sense of
            Functional Programming) in style.

            I have used the approach and found that I got very clean and
            reliable code from it. As with any approach to programming
            that is functional in style, I tended to get simple functions
            that hooked together in a kind of clean mathematical way. Nice
            modularity. I remember liking the technique and finding it
            useful.

            This is a technique that is well worth knowing. I'm really
            glad that I have so much mathematics and math style in my bag
            of tricks. There are times when I have encountered a problem
            that was hard to reason about, and using math / proof /
            functional kinds of reasoning has helped me get the job done.

            Parnas' ideas are not a form of test-driven development.

            While I did find Parnas' approach valuable, to my recollection
            it is everything that test-driven development is not. Parnas
            recommends intensive, formal, and ritualistic reasoning about
            the code one is writing. The approach is a way of synthesizing
            a design in a mathematical kind of meta-language, the language
            of assertions and invariants. It is design up front, pure and
            simple.

            It isn't testing, it isn't letting the code participate in the
            design. It's different from what we do, and in my opinion, it
            is, as Hill suggests, "old school".

            Parnas' work is not a form of test-driven development, unless
            tests don't have to be automated or even executable. It is,
            however, an interesting technique to know and to know well.
            Maybe it's an etude.

            Ancient ideas may be XP compatible.

            I'm glad that I know as much about programming as I do, and I
            wish that I knew even more. It gives a kind of richness to the
            way I think about problems, and it gives me a kind of "instant
            access" to solution ideas. I wouldn't want not to have that.

            I know I use those ancient ideas as I work

            Yet XP is different, perhaps in a discontinuous way.

            That said, while I'm glad I know how to do these kinds of
            things, I do far less of them than I used to. My programming
            now is more like a StarLogo Turtle, or a cell in the Game of
            Life. I work locally, and get good results globally.

            I'm richer for the XP way of working, mostly because of the
            people connections, and wouldn't go back. But I get to build
            fewer castles in the air. Sometimes I miss that.

            Ron Jeffries
            www.XProgramming.com
            You do ill if you praise, but worse if you censure,
            what you do not understand. --Leonardo da Vinci
          • Dave Astels
            ... I agree. I talked about this issue from a slightly different angle in my XP For One tutorial last week. For me, the goal is to get better at creating
            Message 5 of 23 , Jun 3 4:46 AM
            • 0 Attachment
              On Fri, 2002-05-31 at 06:53, Dave Rooney wrote:
              > Kent,
              >
              > I don't think this really addresses the issue of making XP simpler, but...
              > In a recent presentation I was asked the inevitable question, "If you don't
              > use all of the practices, is it still XP?" I responded that it's not so
              > much the practices as it is the "spirit" or "mind-set" of the team
              > practicing them.

              I agree. I talked about this issue from a slightly different angle in
              my "XP For One" tutorial last week. For me, the goal is to get better
              at creating software, not at checking off practices.

              Dave
              --
              Dave Astels (dastels@...)
              Coauthor of "A Practical Guide to XP"
              Presenter at XP2002 (www.xp2002.org) of:
              "XP For One" tutorial
              "Refactoring with UML" paper
              CEO & Master Software Craftsman, Saorsa Development Inc.
              www.saorsa.com
              "An Agile Software Studio"
            • Charlie Poole
              Hi all, If you re coming to OOPSLA, stop in on Tuesday night for the birds-of-a-feather session by, for and about XP User s Groups! It s Tuesday Nov 5 from
              Message 6 of 23 , Oct 31, 2002
              • 0 Attachment
                Hi all,

                If you're coming to OOPSLA, stop in on Tuesday night for the
                birds-of-a-feather session by, for and about XP User's Groups!

                It's Tuesday Nov 5 from 7-9pm in meeting room 602.

                Here's the announcement. We'll begin with a series of warm-up
                exercises and then try to answer some of these questions.

                *************************************************************
                Birds Of A Feather: Extreme Programming User Groups

                The Seattle XP User Group is hosting members of other groups,
                or those interested in forming XP groups, to discuss issues
                like:

                * How can groups help their members to learn good practices.

                * What can groups do to improve awareness of XP practices
                in the local development community and to reconcile XP
                with other approaches... or should they do that?

                * What can groups do to increase local opportunities for
                members to practice XP?

                * Are there ways in which groups in different localities
                might work together to support one another?

                ***************************************************************

                Charlie Poole
                cpoole@...
                www.pooleconsulting.com
                www.charliepoole.org
              Your message has been successfully submitted and would be delivered to recipients shortly.