Knowledge Mapping

This is long and fragmented, but it may rove to be of interest to some
of you on this list. Howard

- - The original note follows - -

Path:
psuvm!news.cac.psu.edu!howland.reston.ans.net!europa.eng.gtefsd.com!uunet!utcsri
!dgp.toronto.edu!wright
Newsgroups: alt.architecture
From: [email protected] (Robert M. Wright)
Subject: Knowledge Mapping
Message-ID: <[email protected]>
Organization: CSRI, University of Toronto
Date: 15 Oct 93 20:50:46 GMT
Lines: 136

Hello out there

Let me introduce myself first. I am Robert Wright a Professor in the School of
Architecture and landscape Architecture University of Toronto. I am a member of
the Centre for Landscape Research,
head of applied projects, at our School. For the last 7 years the Centre has
been developing software
for decision making in Environmental Planning and Design. We use SGI (silicon
graphics) machines as a platform and have developed some software for
interactive and real-time modeling of urban design issues dealing with
integrating GIS and CAD formats into a system which allows users to explore 3-D
models and interactively test ideas out which influence the physical
environment. We have funded ourselves through real projects with public agencies
and work with a network of other universities (Harvard, MIT, ETH Zurich) just to
name a few. We have made our software available to them free and have fostered
some good exchanges. Several design and planning schools use our software for
teaching and research purposes.

Our main software (unstable and continually evolving) is called PolyTrim (TRIM -
meaning Toolkit for Interactive Modeling) we have released several programs into
the public domain work which can be accessed by others. I will list the ftp
location at the end of this mail message, so that if nothing else your team can
know about them and look at them if it might be of interest to you. At present
they run only on SGI machines. I am personally an IO (idiot operator) with no
real programming skills, all of the programming has been done by students (none
trained officially in computer science) but our main developer Rodney Hoinkes is
now doing a Doctorate of Design at Harvard and continues to work with us
sporadically through the year and during the summers.

So much for the formalities.

I am presently on a six month sabbatical doing some soul searching and trying to
organize a conceptual framework for myself that will bring into line the
teaching, research and professional implications of our work. I am speculating
on the impacts to our discipline (landscape architecture) that technology and
our current inability to build significantly on our knowledge base in the face
of complex trans-disciplinary issues facing us in contemporary culture. My
speculations are as follows;

1. There are now major structural changes occurring in our economy and social
institutions, e.g., the shift to 'knowledge based' economies and cultures.
2. As a culture we are generating increasing amounts of information but are
losing the ability to manage, interpret and act on relevant information.
3. Static disciplinary boundaries and institutional frameworks are now
impediments to the developing of critical knowledge and creative approaches
needed to solve complex problems.
4. The educational needs and skill sets of practitioners are rapidly changing
due to the increasing specialization of knowledge and the subsequent fast pace
of technological development.
5. Career paths of young practitioners are becoming increasingly fragmented,
long term experience in a single job setting is becoming rare and opportunities
for long-term traditional professional based apprentice-mentor training are
diminishing.
6. The is an increased need to access extra-disciplinary knowledge and to engage
in meaningful trans-disciplinary activities.
7. There are increasing demands by social institutions on professional, applied
disciplines to justify their existence and relevance related to perceived social
needs.
8. There is a diminishing amount of mutually beneficial knowledge being
exchanged between our educational institutions and professional practitioners.
9. Institutions and the public are demanding more involvement in the design
process of which they bear the cost and consequences of design action.
10. We need to build more dynamic forms of dialogue and interaction that share
knowledge and experience in participatory design processes.
11. There is an overall reduction in the amount of capital available for
creative initiatives that would respond to the issues identified above.

I believe part of the solution is to develop better techniques or approaches to
'knowledge mapping' which I define as follows; A knowledge map is an
interactive, open system for dialogue that defines, organizes and builds on
the,
intuitive, structured and procedural knowledge used to explore and solve
problems. It exposes the process of knowledge structuring that leads to
proposed
solutions A knowledge map is trying to capture and makes accessible to others
the experience, methods, processes and judgments used by a person or groups
about a given intent. It is an active technique for making knowledge
representable, explicit and transferable to others.

I have perceived of this system starting of using a notational structure
borrowed from diagramming techniques in our own field (the idea of conceptual
graphs in AI (nodes and relationships)) and by using an organizational
structure
in what I concieve as four spaces (Dialogue spaces, Construct-Relation Spaces,
Operational Spaces, and Interpretive Spaces). The system in its most robust
form
would have to be multi-representational and interactively on-line so that other
could interact, inherit and build on the maps. I have talked to people in the
Mcluan Culture and Technology Program, Faculty of Education etc., here at U of
T
and their seems to be (despite the obvious difficulties and complexities)
interest in the idea. At present I have done analog versions using file folders
(to represent the spaces and have found the technique useful to me at least. I
am using the idea of a knowledge map to help construct knowledge maps of
courses
I am formulating, as a critical tool to look at landscape architectural
knowledge. It seems to work for me. I have tried to think of systems
(multi-media etc.,) that might let me grasp even in some awkward way how I
might
start to chip away at the idea. But to be frank I am disappointed in what I
have
seen in multi-media (canned middle management slide shows for marketing).
Rodney
has started to develop a Hyper-Card like tool (SGI based) that links into our
software (called TRIMCard) to allow connection between diagramming ideas and
the
links necessary to access to multi-media representations that would allow users
to navigate a diagram, understand the relationships implied and access such
things as windows for navigating real-time abstractions like a 3-D model of a
city etc., to explore ideas. At present I am trying to stand back from the
technology a bit and to develop a stronger conceptual framework for the idea. I
know that this will be a long term proposition 5-10 years etc., but what the
hell!

So lets cut to the chase, am I crazy?, does the idea have any merit? (as badly
and as naively as I described it) could anyone point me in some fertile
directions (people, references)? Don't hesitate to tell me to go away? I think
I have started to work on something that is, I think! important. I am starting
with fellow members of our Centre and with people across the university to se
up
a trans-disciplinary working group to share ideas. I have tried to formulate
some of my ideas in a paper for Landscape Journal, I don't want to bore anyone
or distract anyone (too much) from their own work. At least I would like to
establish a starting dialogue if anyone feels it is appropriate.

As Promised here is the ftp address for our stuff.,

Yours truly,

rob

-------------------------------------------------------
Current Software Releases for SGI Machines
available via anonymous FTP:

explorer.dgp.toronto.edu (128.100.1.129)
in pub/sgi

Program Version Description
-------------------------------------------------------
CLRview v1.50 - Real-time visualization
for CAD and GIS data sources
CLRpaint v1.01 - 24bit Paint/Image Processing
program (handles most formats)
JPEG tools v4 - JPEG viewer/translator/library
PhotoCD tools v0.5 - PhotoCD viewer/translator/library
-------------------------------------------------------
Partial thread listing: