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

Re: [XP] Comparing Architectures

Expand Messages
  • Steve Berczuk
    ... I can think of a few of answers to this ;) 1. People tend to know more about what they need from living and working spaces than they know about software.
    Message 1 of 7 , Sep 1, 2005
    • 0 Attachment
      On 9/1/05, Victor <vmgoldberg@...> wrote:
      > When the Gang of Four wrote their book Design Patterns, they were heavily
      > influenced by the architecture of buildings. Since then, the name stuck and
      > now people talk about software architecture and software architects.
      >
      > Of course, careful people are clear in delineating the differences, like
      > software developers don't need to move walls when there is a change in
      > design. Once I was making a presentation about XP and an experienced and
      > bright developer asked me: "Architects and all kind of engineers can do
      > great upfront design, why is that software developers can't?"

      I can think of a few of answers to this ;)
      1. People tend to know more about what they need from living and
      working spaces than they know about software.
      2. People expect less from buildings than they do from software.
      3. (as you say in your original message) There often really isn't a
      whole lot of great up-front design in buildings.

      There are also more constraints in the physical world than the software world...
      I'm working with an architect now to do an addition to my house...
      He's very good at talking what we tell him, translating the
      "requirements" into "needs" and then sketching out ideas. We're going
      to be interating a lot on design possibilities... In an ideal world
      we'd want to live with "First Floor Scheme A" for a while, then maybe
      move a room around when we realize that the laundry room would work
      better if it were 3 feet wider (and we don't really need the adjacent
      closet anyway(*). But we can't do that, so we're going to iterate at
      the design phase. And they will be small changes during construction
      as we discover issues with the foundation, existing structures, etc...

      But in software we can actually show things to people as we go, and
      change them, so why not do that...

      I suspect that "good" up front design is VERY rare in most all types
      of projects... it is just a necessary evil in projects that involves
      bricks, steel, etc...

      Steve


      (*) anyone who lives in a house built before 1900 realizes that adding
      /any/ closets is a good thing, and that this is very hypothetical! ;)

      --
      Steve Berczuk | steve@... | http://www.berczuk.com
      SCM Patterns: Effective Teamwork, Practical Integration
      www.scmpatterns.com
    • yahoogroups@jhrothjr.com
      From: Victor To: extremeprogramming@yahoogroups.com
      Message 2 of 7 , Sep 1, 2005
      • 0 Attachment
        From: "Victor" <vmgoldberg.at.verizon.net@...>
        To: "extremeprogramming@yahoogroups.com"
        <extremeprogramming.at.yahoogroups.com@...>
        Sent: Thursday, September 01, 2005 9:32 AM
        Subject: [XP] Comparing Architectures


        > When the Gang of Four wrote their book Design Patterns, they were heavily
        > influenced by the architecture of buildings. Since then, the name stuck
        > and
        > now people talk about software architecture and software architects.

        I believe that the term "software architect" was in use long before
        the GOF book.

        > Of course, careful people are clear in delineating the differences, like
        > software developers don't need to move walls when there is a change in
        > design. Once I was making a presentation about XP and an experienced and
        > bright developer asked me: "Architects and all kind of engineers can do
        > great upfront design, why is that software developers can't?"

        The best answer is that they can't do it any better than we do. The
        illusion that they can is fostered by a lack of knowledge of exactly
        what it is that these people do.

        An architect in the building trades is the same as a UI
        or useability designer in software development; the actual
        work of designing the building, in the sense of structural
        steel, plumbing, wireing and so forth is done by a structural
        engineer. And the details of how to assemble the building is
        usually done by the general contractor.

        Disaster buildings are not unheard of. The State of Illiniois
        building in Chicago has (or had) notorious livability problems, and
        there was one in Boston where _all_ the windows blew out.
        "Sick building Syndrome" didn't get it's name because it's
        unheard of. It's all too common.

        Challenge the delusion head on.

        [excellent observations snipped]


        > Victor
      • Keith Ray
        See Brand s book How Buildings Learn.
        Message 3 of 7 , Sep 1, 2005
        • 0 Attachment
          See Brand's book "How Buildings Learn."



          On Sep 1, 2005, at 8:32 AM, Victor wrote:

          > Reversing the inspirational flow of ideas, there is a tendency to
          > build
          > large spaces with mobile walls, which makes the architecture more
          > flexible.
          > It would be nice if building designers would also make more widely
          > available
          > considerations of esthetics, livability, and comfort. :-)
          >
        • William Pietri
          ... As Keith Ray says, How Buildings Learn is a fantastic book that demolishes a lot of myths about buildings and how they get built. I found it very
          Message 4 of 7 , Sep 1, 2005
          • 0 Attachment
            Victor wrote:

            >Once I was making a presentation about XP and an experienced and
            >bright developer asked me: "Architects and all kind of engineers can do
            >great upfront design, why is that software developers can't?"
            >
            >

            As Keith Ray says, "How Buildings Learn" is a fantastic book that
            demolishes a lot of myths about buildings and how they get built. I
            found it very inspiring.

            Another way to answer these people is to point out that architects only
            do half the job. Nobody wants just a house; they want a place to live.
            We're accustomed to doing a lot of the rest of the work by painting,
            adding furniture, putting in lamps and phones, and maybe remodeling the
            kitchen.

            >My conclusion after finding so many buildings that could benefit from
            >some/much improvement is that software development, when practiced by
            >competent professionals using an agile methodology, has the potential of
            >producing much more satisfying results in terms of functionality and
            >maintainability than what can be expected from traditional architects and
            >civil engineers.
            >

            That's an excellent point.

            I think high tech, between its potential and the history of rapid
            improvements, leads people to expect a lot more than they do with
            traditional materials. People accept that with traditional houses you
            can't just move the doorway six inches to the left, and that cars
            neither fly nor go underwater. But most non-technical people don't have
            a similarly clear model of the current limitations of software and hardware.

            I was just talking with somebody yesterday who thought it perfectly
            reasonable that if her Wifi-equipped laptop gets her internet service
            anywhere on campus, it should work anywhere off campus, too. And she's
            right; in ten years, that's probably how it will work.

            But you're right; rather than getting people to lower their
            expectations, we might as well take our agile methods and see how well
            we can keep up. Maybe if we stir things up enough, somebody will get
            inspired to solve whatever problems are keeping me from getting my jet pack.

            William
          • Jeff Langr
            Richard Gabriel s book Patterns of Software also has some interesting thoughts. For example, referring to the New England farmhouse: The result is rambling,
            Message 5 of 7 , Sep 1, 2005
            • 0 Attachment
              Richard Gabriel's book Patterns of Software also has some interesting
              thoughts.
              For example, referring to the New England farmhouse:

              "The result is rambling, but each part is well-suited to its needs, each part
              fits well with the othersÂ…The inhabitants are able to modify their environment
              because each part is built according to the familiar patterns of design, use
              and construction and because those patterns contain the seeds for piecemeal
              growth."

              Jeff
              author, Agile Java: Crafting Code With Test-Driven Development

              Quoting Keith Ray <keithray@...>:

              > See Brand's book "How Buildings Learn."
              >
              > On Sep 1, 2005, at 8:32 AM, Victor wrote:
              >> Reversing the inspirational flow of ideas, there is a tendency to build
              >> large spaces with mobile walls, which makes the architecture more flexible.
              >> It would be nice if building designers would also make more widely available
              >> considerations of esthetics, livability, and comfort. :-)
            • Fehskens, Len
              ... If you re thinking of the Hancock Tower, nowhere near _all_ the windows blew out. Less than a hundred (still an embarassing number) of the over ten
              Message 6 of 7 , Sep 2, 2005
              • 0 Attachment
                Victor sez:

                >there was one in Boston where _all_ the windows blew out.

                If you're thinking of the Hancock Tower, nowhere near "_all_" the
                windows blew out. Less than a hundred (still an embarassing number) of
                the over ten thousand (10,344) windows popped out, due to flexing of the
                building due to wind loads and thermal expansion. The windows were not
                blown out. The flexing was corrected by the installation of an inertial
                damper.

                http://www.galinsky.com/buildings/hancockboston/
                http://www.glasssteelandstone.com/US/MA/BostonJohnHancock.html

                Just in the interests of factual correctness.

                len.
              Your message has been successfully submitted and would be delivered to recipients shortly.