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

RE: [XP] Re: Agile Rentschian Thinking

Expand Messages
  • Mike Beedle
    ... Grady: I agree, Gould s latest work is outstanding and important. Together with The Origins or Order by Kauffman, I think they will take the
    Message 1 of 48 , Jul 11, 2002
      Grady Booch wrote:
      > Mike Beedle wrote:
      > > Have you read any of the works Kauffman, Holland, Prigogine, Axelrod,
      > > Axtell, Gessler, etc.?
      > [egb> ] I've read some of these authors (and upon your recommendation just
      > ordered a pile from Amazon). Axelrod I have found to be quite vacuous; I
      > don't know of Axtell. On the evolutionary front, I'm more of a
      > fan of Gould and especially his latest, "The Structure of Evolutionary
      > Theory."


      I agree, Gould's latest work is outstanding and important. Together with
      "The Origins or Order" by Kauffman, I think they will take the understanding
      of what life is (more Kauffman), and how it evolves (more Gould), to
      a whole new unprecedented level. Also, in his book Investigations,
      Kauffman explores the expanded role of self-organization in the Universe,
      the mind, and human interaction and exposes new world-view that is
      bound to

      Btw, Axelrod is considered one of the pillars of social understanding
      using Complexity Science through his classics:

      Complexity of Cooperation and
      Evolution of Cooperation.

      Axtell and Epstein also give some of the best explanations as to why
      human organizations always work from the bottom up:

      Growing Artificial Societies:
      Social Science from the Bottom Up

      The above are all classics in the understanding or human organization
      from the perspective of Complexity Science.

      Grady Booch wrote:
      > By fragile, I mean that the conditions that make a self-organized
      > system a) exist at all and b) be resilient to any reasonable degree
      > is c) a wonderfully and fearfully balanced thing.

      I would prefer to use the word stability for what you describe above
      not fragility.

      And I would call the things that can spin the system into undesirable
      states vulnerabilities.

      Do agile projects and methods have vulnerabilities? You bet. To say
      the contrary would be foolish, but I wouldn't use the word "fragile"
      as an adjective to describe agility or self-organization. I would simply
      say that they have different stability conditions and ranges and
      that they have vulnerabilities, as any system does.

      As I see it, XP, Scrum or RUP projects have different stabilities and
      vulnerabilities. Which have more and how do these vulnerabilities affect
      the system is a measure of their fragility.

      Grady Booch wrote:
      > Mike Beedle requested:
      > > Can you enlighten us with specific references?
      > [egb> ] The folks at NASA's Astrobiology Institute
      > (http://astrobiology.arc.nasa.gov/institute_new/about.html) note "We are
      > discovering both the fragility and robustness of life, as we
      > investigate the history of mass extinctions on our planet (including the
      > extinctions taking place today), the subtle alterations in
      > climate triggered by volcanic eruptions and human industry, and
      > the destruction of our protective shield of ozone." Also check
      > out Sole's study on the complexity and fragility of
      > food webs at http://complex.upc.es/~montoya/PRSfragil.pdf. John
      > Doyle at CIT
      > observes that "biological systems are robust yet fragile" in
      > http://mitpress2.mit.edu/ICSB2000/doyle.html. You can read much more about
      > his work at http://complex.upc.es/~ricard/. The axiom of differential
      > fragility appears throughout the literature in contemporary ecological
      > management studies, as in
      > http://www.icsu-scope.org/downloadpubs/scope53/chapter19.html,
      > for example.

      Thanks for the references. I'll grok through them.

      Grady Booch wrote:
      > Mike Beedle asked:
      > > Did it feel to you that you needed to have all the extra baggage of
      > > RUP in those days, or was it sufficient that you could produce lots
      > > of quality software in a short amount of time while you were having
      > > fun?
      > [egb> ] In one engagement in particular, I and my pair programmer, over a
      > two week period in the wilds of Canada, had a lot of fun banging out some
      > great code without all that ugly "extra baggage", well, except ...

      > .. we fully followed the essentials of the RUP: architecture first,
      > growing the architecture through incremental and iterative releases, ..

      Really? Did you engaged into an environment phase where you generated a
      new process instance form a process framework? Did you followed this
      generated instance like an ETVX process? Did you sequentially executed
      tasks and 1) capture use cases, 2) created an architecture, 3)
      "constructed" the architecture by releasing the executable releases
      in iterations, etc., 4) evolved artifacts in iterations, etc.

      Sounds kind of heavy for a Smalltalk project. In most Smalltalk projects
      that I have been involved with, there has always been a great deal
      of self-organization driven by the Smalltalk culture practices and

      I am seriously interested in the answer to this question:

      Have you ever been allowed to self-organize in a team to
      deliver the project's goals?

      Grady Booch wrote:
      > [egb> ] Delivering a <product> is much more than delivering a pile of
      > executable code.

      Yes, but delivering an executable typically gets more customer satisfaction

      Is satisfying the customer by delivering executable code top priority for

      It is for me, btw.

      Grady Booch wrote:
      > [egb> ] If it walks like a duck and talks like a duck, one can reasonably
      > conclude that it is a duck. I would argue that in each of these cases you
      > mention, you have injected a new way of working (a process) that was
      > different than the initial conditions: you specified a set of artifacts
      > (stories) and activities (test first) and specific roles (coach). If this
      > does not constitute the injection of a process, what do you call it?

      Installing _practices_ or patterns if you prefer.

      Installing practices is not the same as installing a defined process.

      In agile methods, when we install practices the actual process executed
      are a second order effect -- they are generated by the application of
      the practices.

      So I vehemently disagree, a "new way of working" is _not_ necessarily a
      defined process, it can be a dynamically created self-organized process.

      So it's not a duck is a dove ;-) And zirconias are not diamonds, but
      they look like them ;-)

      Grady Booch wrote:
      > [egb> ] But creating a product requires more energy than simply
      > creating an executable release....

      I agree. And just to clarify, in Scrum we don't self-organize only
      to create executables. We can add _any_ tasks into the Product Backlog.

      But yes, we have a bias towards delivering executable code :-)

      - Mike
    • Laurent Bossavit
      ... Controlled by whom ? Aye, there s the rub. ... Agreed. Here and now, given our skill and knowledge, it would be stupid. Do remember, though, that fairly
      Message 48 of 48 , Jul 25, 2002
        > As such, any meaningful method has to "embrace change" but perhaps
        > where we differ is that I would add "and do so in a controlled manner."

        Controlled by whom ? Aye, there's the rub.

        > if you are building a high rise, it would be infinitely stupid to start
        > with a pile of lumber and some hand tools and expect to be successful.

        Agreed. Here and now, given our skill and knowledge, it would be
        stupid. Do remember, though, that fairly simple biological organisms
        routinely build, with their bare appendages, constructions which are
        to them as a highrise is to us. Termites. Wasps.

        Conclusion : it is not stupid to expect that we might *develop* the
        knowledge and skill to build a highrise starting with a pile of
        lumber and some hand tools.

        Now. Metaphor is nice, but we are not, in fact, building highrises.
        We are building software, a different kind of thing. What transposes
        there from the metaphor, and what doesn't ?

        > To be clear as well, I have problems with the pseudoscientific sound
        > bites in your statement: "embracing change" strikes me too much like
        > the pop business edits of the 80's and 90's. I mean, what's the
        > alternative? "I'm a Luddite."

        I don't see where you get that. Could be a language problem - as a
        freakin' furriner I always double-check this kind of thing.

        Dictionary.com says : "embrace, to take up willingly or eagerly", or
        more interestingly, "to avail oneself of". Thus the alternatives are:
        "to accept change reluctantly", or even "not to avail oneself of the
        possibilities offered by change". Which is where you're leading with
        your "embrace change in a controlled manner".

        Dictionary.com says Luddite is "One who opposes technical or
        technological change". I think one could legitimately oppose some
        kinds of technical change - human cloning is routinely opposed by
        some people who, perhaps, might object to being labeled Luddites.

        I definitely don't think "embrace change" is vacuous. On the
        contrary, reading some Roberto Unger, who in the domain of social
        science says the same thing under the slogan "Plasticity into Power",
        I found more content in the position than I would have expected from
        a principle of software development.


        The greatest obstacle to transforming the world is that we lack the
        clarity and imagination to conceive that it could be different.
        Roberto Mangabeira Unger
      Your message has been successfully submitted and would be delivered to recipients shortly.