Date: Sat, 22 Dec 2001 05:45:49 -0800
Mr. Rod Welch
rowelch@attglobal.net
The Welch Company
440 Davis Court #1602
San Francisco, CA 94111 2496
..
| Subject: | Comments on SDS documents |
Rod,
This comes in a lump because that is how I read the documents – in a lump.
I don't intend for you to respond to everything, only those that seem to
want further discussion.
..
These are miscellaneous notes and comments from the pages about SDS.
..
Plato, Aristotle, and the Bottom Line:
A definition of luck: where opportunity meets preparation.
..
..
High Cost of Medical Mistakes
..
Asilomar Redux
"During Q&A, Mr. Koppelman said the project would have been more successful
if he had hired project managers who tell the 'truth'."
..
I continue to hear statements from management that are so outrageous that I
have to ask myself repeatedly:
The reference to "case law" brings up an objection I have to the legal
profession’s approach to documentation, which is carried over in many other
settings: There is never an update or reference to the original work. The
originating documents are never refactored. The result is that the only way
to know the current state is to start at the beginning and retrace the
entire history. I have seen corporate policy manuals that were handled this
way: never issue an update to a policy, just add changes as needed. Often
there is no single source of policy information and retracing the history or
consulting the oral tradition is the only way to know what the current state
is.
..
"The legal record of a case is organized by a structure to facilitate
access, so that participants can find relations and correlations quickly."
..
I accept that the record is organized, but doubt that this supports finding
relations and correlations quickly. The ability to organize the record and
records pertaining to a case are on of the major uses of outliners such as
MaxThink and Ecco in the legal profession.
Specialized tools such as CaseMap and its related products
(www.casesoft.com) are an attempt to meet the need for a means of organizing
the material if not the record of a case.
..
Is the PMBOK accessible online anywhere, or is this an PMI members only
document?
..
Executive Mindset, Solving Obstacle to Leadership
The characterization of the functions of leadership are good. I think that
one problem we have is that management functions are poorly understood: a
major function of management should be to remove barriers to production, but
that does not seem to be understood.
..
"Clearly, when people speak there is always an intent to lead the listener
toward some understanding that is usually aimed at achieving specific follow
up action."
..
Terry Winograd terms this "CfA (Communication for Action)" and states,
correctly, that it is only one type of communication that is important in
the analysis of systems. His work on language / action is interesting. He
built a computer program to manage such communications (much as IBIS does
for argumentation, but the state diagram is more rigid). I don't know how
well that worked and where he stands today. His web site has some
interesting papers. He is currently at Stanford, I believe.
..
I will save for a fuller discussion my ideas on "Communication is
Understanding". Briefly, however:
..
A permanent record does not equate directly to understanding although it
does reduce drift substantially. While correct follow up action is the
ultimate indication of understanding, we need short cycle feedback as well.
Correct paraphrasing is one method that has proven successful ("let me see
if I understood what you said ..."). There is also a concept called an
"Admin (administration) Scale that sets out the hierarchy of priorities from
vision down to orders and how they need to align for an organization
actually to do what it claims it is doing. There is a whole treatise to be
written on this topic. This was sparked by the introduction of "organic
structure" as the Admin Scale is a major (essentially unknown) one.
..
"'Understanding' is so critical in construction that specialists are used,
called "architects," to figure out the right connections, and they inspect
the actual work against original objectives set out in a data base, called
"plans and specifications," to ensure those connections are correct so that
the building will stand up under the loads it will encounter."
..
This statement is, unfortunately, utopian. What is the reality? Based on my
experience this view of building is not a whole lot more accurate in
practice than is "process" in software development -- hardly anything ever
gets done the way the theory says that it should.
..
I suggest that the term "metrics" has a more precise definition than the on
you are using in many contexts. In process and software a "metric" is a
countable event, not just a means of gauging progress.
..
"Dr. Landauer describes a study showing human mental 'understanding' changes
as a function of new information."
..
This is a major reason for refactoring and for attaching new insights to
original ideas. There is often a fine line between new insight and memory
drift.
..
"This leads to the third key ingredient in communication: a system of
'follow up' linked to original understandings that ensures desired action is
taken, because if needed action is not taken, leadership fails. How is
'follow up' accomplished?"
..
The correct pattern, hardly ever used is that the person giving the order or
direction receives the follow up with supporting evidence of what was done
so that the result of the action can be compared to the intent of the
directive.
..
"Therefore, technology that improves understanding and follow up, makes it
easier to motivate people to accomplish difficult
objectives."
..
True, but there are often other, more powerful incentives at work in the
culture. Witness what happens, for example, when productivity improvements
are perceived as a threat by unions, or the behavior on a "cost plus fee"
contract. The cultural incentives can totally overwhelm personal incentives,
and do so in many if not most cases.
..
Disincentive often comes in the form "what you do speaks so loudly that I
can't hear a word you say," as when management speaks about quality while
ardently resisting even the smallest of changes which might make it possible
to improve quality.
..
Do this often enough and there is simply no credibility left.
..
With regard to disconnection in time making it difficult to perceive the
value of capturing the record: humans have a difficult time associating
cause with effect the further separated they are in time. The immediate
accessibility of the original "cause" helps to reduce the perceived span of
time and helps associate cause and effect.
..
Typical Day Scenario using the SDS Program
"The record shows that Wendy said people want to do a better job, they want
to make more money, they believe that lifting the capacity to think,
remember and communicate is the answer, but they need faith to sustain them
in taking up SDS because this is really using a new craft: Communication
Metrics."
..
I really would like to believe any part of this statement, but it runs
completely counter to most of my experience. I have witnessed parts of this
at some times in some individuals, but the proportion is too small to be
called anything other than exceptions.
..
POIMS
"This biological drive to transfer momentary common sense by telling others
what we believe we 'know,' ..." Look at the book "Virus of the Mind : The
New Science of the Meme" by Richard Brodie (Hardcover - September 1995). I
got it because it addresses such knowledge transfer, but I haven't read it
yet.
..
In the section on "Management Details" there is no place for the
relationships and connections among the various sorts of details that
confront the manager.
..
"In the vernacular: the paperwork never catches up with the real work!"
..
In a schedule driven organization such as the Space Station Project at
Boeing, often the paperwork is not even started until the work is
(purportedly) nearly done. What passes for documentation is in a continual
state of near disaster, and there is no easy way to determine what is
current and what isn't. Determining that a document is complete is
essentially impossible, so we rely on human memory and personal archives for
much of our validation. Determining cross-document consistency is
effectively impossible.
..
I am convinced that there are problems of this sort that are inherent in the
way in which we structure hierarchical organization. This is one of my
current areas of research.
..
"Implementation of POIMS technology starts as a personal schedule,
..."
..
One of the problems I have with all task management systems is that planning
and tracking are not separate. I want to be able to do a fairly complete
breakdown of a plan into tasks. These tasks have interdependencies that mean
that some things cannot be done until their predecessors are complete. Some
plans are only plans, they are not active until I decide to start on them.
..
I want a task management system that will show me only things that are of
interest -- if it can't be done because required predecessors are not
completed, I don't want to hear about it. If the project is not active or is
not in the area of my current focus, I don't want to be bothered with any of
it. No system I have ever seen will do this, and I have been threatening to
build one for too long.
..
"As problems mount, crisis management results in endless cycles of
correcting mistakes, called rework, ..."
..
Fear limits focus, making it harder to maintain perspective and an insight
into the "bigger picture". There is also the phenomenon of trying to take
shortcuts under time pressure, and to do what is easy -- to "look under the
streetlight" because there is light there rather than looking in the dark
alley where you lost the keys for which you are looking.
..
"The problem with keeping a list of "things-to-do" is that the dynamics of
daily management result in there not being enough time to maintain a useful
list."
..
Having attempted this several times, there are a number of problems that
contribute to this:
Thanks,
Sincerely,
Dynamic Alternatives
Garold L. Johnson
dynalt@dynalt.com
http://www.dynalt.com/