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

RE: [scrumdevelopment] Agile 80-20 Rule vs. RUP Risk prioritization

Expand Messages
  • Mike Beedle
    ... William: Great article. It is interesting to discuss the relationships of (hyper)productivity, nonlinearity, connectedness, communication and
    Message 1 of 9 , Aug 31, 2004
      Message

      >Although it seems like everything follows an 80/20 rule (or something

      >similar) , that is an exageration.  When applied to
      people, this
      >becomes problematic because the suggestion that 20% of the
      people do
      >80% of the work impiles, roughly, that 80% of people are
      below
      >average. This seems strange when qualities like intelligence
      and
      >strength are known to follow a Normal (Bell curve)
      distribution.
      >
      >Although I cannot pretend to understand the meaning
      of this I suspect
      >that some people are the "glue" that make teams work.
      These very
      >productive people are adept at building on the work of others
      and ,
      >therefore, appear to have superhuman
      productivity.

      William:

      Great article.  It is interesting to discuss the relationships of
      (hyper)productivity, nonlinearity, connectedness, communication
      and collaboration.

      Non-linearity is *very apparent* in my experience.  What I see
      it on every day development is that the ratios are, in some cases,
      as high as 100-1, but most typically > 5-1 from the highest performer
      to the lowest performer. 

      The first cause of non-linearity simply appears to be
      "intrinsic" to people -- a combination of social skill, development
      skills, tenacity, will power (?), passion for the what they do, etc.

                              (Cumulative)
                              Iteration 1      Iteration 2  Iteration 3 Iteration 4
      Worker 1  Items/day     1               2               3                 4
      Worker 2  Items/day     2               4               6               8
      Worker 3  Items/day     3               6               9               12
      Worker 4  Items/day     3               6               9               12
      Worker 5  Items/day     3               6               9               12
      Worker 6  Items/day     4               8               12              16
      Worker 7  Items/day     4               8               12              16
      Worker 8  Items/day     8               16              24              32
      Worker 9  Items/day     10             20              30             40
      Worker 10 Items/day     12            24              36             48

      Avg:                            5.0             5.0             5.0             5.0
      Ratio of last 30%:  30/50~60%   60/100~60%   90/150~60%  120/200~60%

      Fig 1. Non-linear people evolving "linearly in time". (Example,
      using "fake data".)

      The second cause of non-linearity is how some workers "learn"
      to be more productive over time .


                              (Cumulative)
                              Iteration 1      Iteration 2  Iteration 3 Iteration 4
      Worker 1  Items/day     1               2             3         4
      Worker 2  Items/day     2               4               6               8
      Worker 3  Items/day     3               6               9               12
      Worker 4  Items/day     3               6               9               12
      Worker 5  Items/day     3               6               9               12
      Worker 6  Items/day     4               8               12              16
      Worker 7  Items/day     4               8               12              16
      Worker 8  Items/day     8               17(9)        29(12)       44(15)
      Worker 9  Items/day     10             22(12)      36(14)       52(16)
      Worker 10 Items/day     12            24             36             48

      Avg:                            5.0             5.01            5.3             5.8
      Ratio of last 30%:  30/50~60%   63/103~61%   101/161~62%  144/224~64%

      Fig 2. Non-linear people evolving "nonlinearly in time" with 2
      nonlinear workers.  (Example, using "fake data".)


      Nonlinearity is certainly puzzling.  It would be great to have
      a controlled "experiment" to study the productivity ratios
      among developers.

      Does anyone know of such a study?  (Involving say 7+ people.)

      (Btw, there Laurie Williams conducted several studies on "pair programming"
      that showed an increased productivity for pairs, but that's
      a little bit different.)

      - Mike

       
      -----Original Message-----
      From: William Nichols [mailto:William.nichols@...]
      Sent: Tuesday, August 31, 2004 5:43 PM
      To: scrumdevelopment@yahoogroups.com
      Subject: Re: [scrumdevelopment] Agile 80-20 Rule vs. RUP Risk prioritization


      The 80 -20 rule has occured in many contexts. Pareto  characterized
      the ownership of property (80% of the land owned by 20% of the
      population) He also observed that about 80% of the population lived in
      about 20% of the metro areas. However, he never called it the 80/20
      law. Other similar distributions have been observed, 80% of web links
      point to 15% of web pages; 80% of citations go to only 38% of
      scientists, 80% of Hollywood are connected to 30% of actors.

      Barabasa showed that such  extreme distribitions occur when the
      underlying distribution follows a power law, e.g. N(x)=x^z. (where
      usually 2<z<3) When a network grows via preferential attachment, e.g.
      (the probability of moving to a specific city is weighted by the
      current population) the result will be a power law distribution. Over
      time, small differences will be exagerated by the preferential
      attachment. Herb Simon actually discovered this mechanizm in the 1950s
      when he explained the size distribution of business firms. The
      characterization of an 80/20 rule is based on the heuristic that 80%
      of the problems will  be traced to 20% of the fault and so on. The
      sizes of cities follow this distribution as do the words used in the
      English language (see Zipf). The procedural interconnectivity of large
      software systems also typically follow this sort of distribution, as
      do the distribution of software faults among those procedures.

      Although it seems like everything follows an 80/20 rule (or something
      similar) , that is an exageration.  When applied to people, this
      becomes problematic because the suggestion that 20% of the people do
      80% of the work impiles, roughly, that 80% of people are below
      average. This seems strange when qualities like intelligence and
      strength are known to follow a Normal (Bell curve) distribution.

      Although I cannot pretend to understand the meaning of this I suspect
      that some people are the "glue" that make teams work. These very
      productive people are adept at building on the work of others and ,
      therefore, appear to have superhuman productivity.


      ----- Original Message -----

      All,
      ..snip

      The first question, is my guess on the Deming source of the 80-20
      rule correct?
      The second question, is my definition of the 80-20 rule correct?
      Lastly, is there general agreement about my conclusion that the 80-20
      rule and RUP Risk Prioritization work well with each other?

      Thank you,

      Mike Van, PMP


      To Post a message, send it to:   scrumdevelopment@...
      To Unsubscribe, send a blank message to: scrumdevelopment-unsubscribe@...


      Yahoo! Groups Sponsor
      ADVERTISEMENT






      Yahoo! Groups Links

      To visit your group on the web, go to:
      http://groups.yahoo.com/group/scrumdevelopment/
       
      To unsubscribe from this group, send an email to:
      scrumdevelopment-unsubscribe@yahoogroups.com
       
      Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.

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