Organization

From: John Conover <john@email.johncon.com>
Subject: Organization
Date: Sun, 8 May 1994 23:26:35 -0700 (PDT)

Excerpts from recollections about experiments in the electronic
mediated group dynamics of intellectual teamwork.

1) Concerns:

        1.1) concerns about mounting evidence that the electronics
        markets were assuming properties/behaviors of a
        NLDS. (Chaotic/fractal.) (Power law, exponential divergence.)

                1.1.1) Implications:

                        1.1.1.1) Not clear what the "prediction
                        horizon" means to strategic marketing and
                        planning.

                        1.1.1.2) Application of Markovian
                        process/statistics to team performance metrics
                        and market analysis may be inappropriate.

                        1.1.1.3) Require very dynamic response to
                        changing market environment that does not
                        close options, etc., and respond to market
                        dynamics in a quick fashion. The decision
                        making must be largely in the team itself.

                1.1.2) Possible Solutions:

                        1.1.2.1) Use semi-autonomous, cross
                        functional, "self-directed" team concept to
                        attempt to integrate the market, management,
                        and intellectual team into a single
                        environment. Would require electronic
                        mediation to accommodate the large volumes of
                        information and communications in a timely
                        fashion. Would require electronic querying to
                        collate and distribute the information.

                        1.1.2.2) Require an "intellectual SWAT team
                        concept."  (Compatible with "Kiasan" process.)
                        Improvisation is an important issue.

                        1.1.2.3) He who does the most the quickest
                        with the least, wins. (Optimal game strategy.)
                        Linear programming should be applied to this
                        issue.

        1.2) Concerns about the game theoretic issues of the
        intransitive nature of determination of priorities within the
        team. Insincere voting in the determination of priorities is a
        related issue.

                1.2.1) Implications:

                        1.2.1.1) If the team was to be
                        "self-directed," it is not clear the means by
                        which resolution of priorities could be
                        established. This is the traditional domain of
                        the levels of upper management, (usually
                        called conflict resolution-which is related to
                        the prisoner's paradox.)

                1.2.2) Solution:

                        1.2.2.1) Coalition seems to fix (Re: works by
                        Stephanie Forest)

                        1.2.2.2) Build "electronic coalition machine."

                        1.2.2.3) Possibly, use NASA's "Group advice,
                        One Man Decisions" as a means of determination
                        of priorities.

2) Implementation: (provides audit trail to counter insincere voting.)

3) Later additions:

        3.1) RCS as documentation control, but not clear whether ISO
        9K accepts.

        3.2) Gnats for sales tracking/program management tracking
        issues-possibly use the "trouble" program.

4) Retrospect:

        4.1) Should have done electronic "Robert's Rules" to formalize
        use of qt.

        4.2) Routing of email (for approvals) in ISO 9K.

        4.3) Structured query (eg., use \subject{...} of TeX.

        4.4) Construct a file on demand.

        4.5) Allows micro-management.

        4.6) Add forms to mail, (like ATT Forms) so that mail could be
        used as a general interface to content databases.

        4.7) Qt should have been programmed as a perl script to allow
        common middle ware to context and content
        databases. (ctreeperl, btreeperl, etc.)

        4.8 It would probably be of some value to apply information
        theoretic issues to the above.

--

John Conover, john@email.johncon.com, http://www.johncon.com/