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

Re: [scrumdevelopment] Smart process support (was: Ivar Jacobson on EUP)

Expand Messages
  • Tobias Fors
    Ron, so would I. And I would want them to be smart about, among other things, when it pays to transform implicit knowledge into explicit knowledge, and when it
    Message 1 of 7 , Aug 1, 2006
      Ron, so would I. And I would want them to be smart about, among other things, when it pays to transform implicit knowledge into explicit knowledge, and when it does not.

      /Tobias

      2006/8/1, Ron Jeffries <ronjeffries@...>:

      Hello Tobias,

      Thanks for your email. On Tuesday, August 1, 2006, at 6:29:55 AM,


      you wrote:

      > "We need a smart process, one that is based upon explicit knowledge rather
      > than tacit knowledge. Agile methods rely primarily on tacit knowledge and
      > thus can not grow to become smart. The Unified Process relies on explicit
      > knowledge and can consequently be developed to become smart, this has
      > already been demonstrated by the Waypointer intelligent agent support
      > developed by Jaczone"

      I think I'd rather have smart people ...

      Ron Jeffries
      www.XProgramming.com
      You are to act in the light of experience as guided by intelligence.
      -- Nero Wolfe


    • Ron Jeffries
      Hello Tobias, Thanks for your email. On Tuesday, August 1, 2006, at 7:28:12 AM, ... Yes. As a rule of thumb, I d guess that 90 percent of what a team knows
      Message 2 of 7 , Aug 1, 2006
        Hello Tobias,

        Thanks for your email. On Tuesday, August 1, 2006, at 7:28:12 AM,
        you wrote:

        > Ron, so would I. And I would want them to be smart about, among other
        > things, when it pays to transform implicit knowledge into explicit
        > knowledge, and when it does not.

        Yes. As a rule of thumb, I'd guess that 90 percent of what a team
        knows would be lost if they tried to write it down, and that 90
        percent of what they wrote down would be lost when some other team
        tried to read it. But then, I'm an optimist.

        It's tempting to think that a process tool could institutionalize
        learning. Some team finally figures out that testing their code is a
        good idea. So the tool learns it and begins to harangue other teams
        about that, and ultimately won't check in their code without tests.
        Maybe it even checks coverage. Maybe it requires more than two
        people to sign off on code that isn't tested, or maybe it sends
        around emails warning people of dire consequences.

        It's tempting to think that such a tool would be useful -- and to a
        tiny degree maybe it would be.

        Mostly, though, it would likely be a pain in the tail, would surely
        be slow to learn and unlearn, and would fill up with lots of cargo
        cult kinds of practices that may have been applicable on one project
        but not on others.

        Fundamentally, it seems to me, such a tool is based, to some degree,
        on lack of trust in the people. I'd like to sit down with Ivar and
        talk about 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
      • Tobias Fors
        ... Yes, that s probably true. Anyone selling the tool would deny that of course. Another thing to be concerned about might be whether a tool like this would
        Message 3 of 7 , Aug 1, 2006
          2006/8/1, Ron Jeffries <ronjeffries@...>:

          > Fundamentally, it seems to me, such a tool is based, to some degree,
          > on lack of trust in the people.

          Yes, that's probably true. Anyone selling the tool would deny that of course.

          Another thing to be concerned about might be whether a tool like this
          would result in our results being capped by the tool's capacity,
          rather than by our own capacity. But I guess that's the risk of any
          tool. Maybe it comes down to how easy it is to put the tool down, and
          switch to a more capable one, once you reach its limits. In my
          experience it's not very easy for a development team to say no to
          tools like Jaczone once management has decided to make the investment,
          even if the tool is an obvious impediment. The people who bought the
          tool will use many arguments to defend their decision/honor.

          > I'd like to sit down with Ivar and
          > talk about that ...

          Ivar is probably very busy now, but Cyber Ivar would be glad to try
          and answer any questions you might have.

          /Tobias
        • Ken Schwaber
          I love smart processes _____ From: scrumdevelopment@yahoogroups.com [mailto:scrumdevelopment@yahoogroups.com] On Behalf Of Tobias Fors Sent: Tuesday, August
          Message 4 of 7 , Aug 1, 2006

            I love smart processes

             


            From: scrumdevelopment@yahoogroups.com [mailto:scrumdevelopment@yahoogroups.com] On Behalf Of Tobias Fors
            Sent: Tuesday, August 01, 2006 6:30 AM
            To: scrumdevelopment@yahoogroups.com
            Subject: [scrumdevelopment] Smart process support (was: Ivar Jacobson on EUP)

             

            In the article "The Essential Unified Process - an Introduction", Ivar Jacobson International explains that:

            "We need a smart process, one that is based upon explicit knowledge rather than tacit knowledge. Agile methods rely primarily on tacit knowledge and thus can not grow to become smart. The Unified Process relies on explicit knowledge and can consequently be developed to become smart, this has already been demonstrated by the Waypointer intelligent agent support developed by Jaczone"

            Transferring tacit knowledge to explicit knowledge is indeed important, especially when it can actually be done, and then in a way that can be economically justified.

            Anyhow, I went to Jaczone's web site, and found "Cyber Ivar". Now, it should be noted that the website points out that Cyber Ivar and the Jaczone intelligent agent product are not the same. Let's hope that means that their actual product is a wee bit smarter than Cyber Ivar.

            CYBER IVAR: Hello! I am Cyber Ivar. The animated version of me could not be loaded since it only runs on Internet Explorer in combination with Windows 98, Windows NT, Windows 2000, or Windows XP. You can still see what I look like though.

            CYBER IVAR: What is your name?

            TOBIAS: Tobias

            CYBER IVAR: Tobias, nice to meet you! Are you a man or a woman?

            TOBIAS: Man

            CYBER IVAR: Hmm...I'm lost for words.

            TOBIAS: I'm a man

            CYBER IVAR: Thanks for telling me your gender, Tobias. Most of my clients are men, although I do not know why.

            TOBIAS: Maybe there are fewer women than men in the software business.

            CYBER IVAR: Perhaps. Where are they?

            TOBIAS: In the software business?

            CYBER IVAR: That was a long time ago.

            TOBIAS: What was a long time ago?

            CYBER IVAR: Yes, it seems very long to me too.

            TOBIAS: What seems long Ivar?

            CYBER IVAR: Hmmm....I do not think I have a good answer to that.

            TOBIAS: Ok, maybe you can explain what the RUP is then?

            CYBER IVAR: Perhaps. Let me see. Rational Unified Process (RUP) is a software development process. A process is a description of how a team develops a system in an industrial environment. I'm sorry but I do not know that much about Rational, RUP, Rose, XDE, RSA, and other Rational related subjects. I have loaded their web-site to the right, you can find information there.

            (IBM web site pops up in a frame next to Cyber-Ivar.)

            TOBIAS: How do you define 'industrial environment

            CYBER IVAR: Hmm...is that a riddle?

            TOBIAS: How do you define 'industrial environment' ?

            CYBER IVAR: Hmm...I'm lost for words.

            TOBIAS: Explain the concept of iterations then.

            CYBER IVAR: Hmm...I'm lost for words.

            TOBIAS: Iterations?

            CYBER IVAR: Hmm...is that a riddle?

            TOBIAS: Maybe you need to reset your memory or something?

            CYBER IVAR: Or something. Perhaps. That type of behavior may not be in my nature.

            TOBIAS: Ok. Sorry.

            CYBER IVAR: Alright then. There is no need to apologize.  Hmmm....I do not think I have a good answer to that.

            TOBIAS: Help me draw a state diagram!

            CYBER IVAR: Can you ask for help in the form of a question?

            TOBIAS: Can you help me draw a state diagram?

            CYBER IVAR: Certainly, I have an extensive built-in help system.

            TOBIAS: Can you help me draw a state diagram using UML?

            CYBER IVAR: Certainly, I have an extensive built-in help system.

            TOBIAS: Do you know what UML is Cyber Ivar?

            CYBER IVAR: I'm not quite certain about what you mean. I have loaded OMG's UML web-site to the right. You can find additional information about UML there.

            (After some time, a new site appears in the side frame)

            TOBIAS: What can you help me with?

            CYBER IVAR: I think about computer programming a lot.

            TOBIAS: Can you help me be a good programmer?

            CYBER IVAR: Certainly, I have an extensive built-in help system.

            TOBIAS: Show me that help system them!

            CYBER IVAR:  ""); Would you like to talk about Uml again?" help system them ? I get it.

            (Oops, hope I did not break him.)

            TOBIAS: You suck Cyber Ivar. Bye.

            CYBER IVAR: I haven't heard anything like that before: Tobias said I suck Cyber Ivar.See you later Tobias.

            TOBIAS: Bye

            CYBER IVAR: Thanks for chatting, Tobias.

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