Hi:
The CC EA, by leveraging the OMB FEA and
by encompassing the evolving outline of Federal Executive Branch (FEB) Mission
Essential Functions (MEF), is probably the first implementation of a
COSMO-based Federal executable architecture.
For further information on what I suggest
regarding a “general ontology’ for use as the COSMO, below is a
comment I posted to the KM.gov forum a few months back.
The CC EA status briefing I presented to
the CIO Council’s Architecture and Infrastructure Committee (AIC) is at http://web-services.gov/CIO_COUNCIL_BRIEF_JUL05-rr3.ppt.
Slide 6 shows the relationship between the OMB FEA and the broader CC EA.
Slide 7 identifies some of the “operational/executable” aspects of
the CC EA. Slide 8 gives an overview of the CC EA (i.e., enterprise
management) metaschema. Slide 9 provides a simplified Framework Model for
the CC EA. A more comprehensive CC EA Framework (i.e., CC Ontology) model
is at http://colab.cim3.net/forum//ontac-forum/2005-10/bin00002.bin.
Slide 12 provides a mapping of the CC EA approach to the OMG Model Driven
Architecture.
The subsequent more-technical briefing to
the IAC EA SIG is located at http://colab.cim3.net/file/work/SICoP/2005-09-14/IAC-EA-SIG_Aug5.ppt.
The next version, with more detail, at http://colab.cim3.net/file/work/SICoP/2005-09-14/RoebuckSICop2005_09_14.ppt
was presented to the SICoP on 14 Sep 05.
Roy
From: KMWG
[mailto:KM-List@xxxxxxxxxxxxxxxx] On Behalf
Of Roy Roebuck
Sent: Monday, August 08, 2005 1:13
PM
To: KM-LIST@xxxxxxxxxxxxxxxx
Subject: Current Federal Executive
Branch (FEB) Use of Ontology-Modeliing and KM Approach for Enterprise Architecture (EA)
Hello:
I've been subscribing to this list for a
while, but have not actively
participated to date. However, my
current program/project
status currently gives
me the opportunity for more active participation. Therefore, I'd like to give a presentation on my current program to your two fora (i.e.,
CIM3 and KM-List), at your convenience.
I've been presenting the approach
to government and industry EA
fora over the past few weeks, and it
seems a natural progression to present it to the various government KM fora. Your interest could help promote the involvement of the government KM, EA, CC, and other stakeholders in my client's Program.
My current U.S. Government EA project, now completing its last deliverable, is using a Model
Driven Architecture (MDA) four-layer metamodel approach of my design
to develop the enterprise architecture
to enable day-to-day management of the
continuity communications (CC)
capability of the U.S.
FEB's priority mission essential functions.
We are taking an ontology-based approach to the CC EA, treating and managing the resultant EA as a dynamic, online,
distributed/replicated knowledge-base. I would emphasize here that my
background is not IT, but rather functional and enterprise management and
management analysis. In those capacities I became involved in 1982 in IT, Data, and KM fields in relation to my other management
improvement duties. Since then I have served as the "CIO equivalent" of a large
military organization, as an EA for an even larger organization, and then in various IT Management and Management Consultant roles since them.
As a technical overview, consider the
following.
Our MDA M3-layer EA repository technology is an ontology and knowledge management
product using standard SQL/ODBC databases. The M3-layer object model is a superset of the
OpenGroup's object metaschema, focused on my metaschema for enterprise management and engineering (EME). We call the
full M3-layer our Architecture Engine.
The M2-layer contains the EME metaschema and the methodology for populating and tailoring it
to specific EA variant
efforts, resulting in an integrating upper ontology for
EME. The structure of
the EA metaschema consists of
a core of seven physical and
conceptual universe "reference
catalogs" as taxonomies providing a superset of the U.S. OMB FEA Reference
Models; and
seven relation-types focused on
physical and conceptual pairs of associations between catalog entries.
The seven reference catalogs are:
· location (e.g., physical, virtual, conceptual),
· organization (e.g., government, commercial, private),
· organization unit (e.g., offices, programs, projects, positions,
roles),
· function (e.g., executive, production, and support functions encompassing the OMB
FEA BRM, the DoD UJTLS, the industry NAICS),
· process (i.e., process decomposed to procedure, template, metadata, and business
rules, linked to loosely and tightly-coupled applications and their data,
aggregated back up to deployable web services registered through the EA
repository),
· resource (e.g., people, situational and environmental intelligence, funds, skills, materiel, facilities, services,
space, energy, time)
· mission (i.e., quantitative, qualified, and scheduled requirements
for resources, as inputs, controls, outputs, and mechanisms, over the requirement life cycle).
The seven relation types between the
catalogs and their entries are:
· Categorization (to build the Catalogs' content into their respective taxonomies)
· Containment (to show composition and distribution relations between
catalog entries)
· Sequence (to shown chronological dependency between catalog
entries)
· Change (to show the past, present, and possible future state of a
catalog entry, such as a mission)
· Equivalence (to show that one catalog entry is the same as or
rough equivalent to another entry, useful in normalizing vocabularies as in a
thesaurus).
· Variance (to show that one entry in a catalog is the is the foundation of another entry)
· Reference (to show that one catalog entry provides
details or other reference information about another entry.
The process for populating and
maintaining the "EA
as KB" and broader EME
content of the repository is achieved by starting and then iteratively following a 32 step data collection and refinement process as part of a broader strategic management life cycle process designed to concurrently support the performance
management of the enterprise and its functions.
The M1-layer EA framework (as an EA
ontology and a user-specified set of information product designs) leverages and
extends the OMB FEA within the broader EME metaschema, and enables inclusion of the variant "EA
Profiles" in each FEB Department and Agency (D&A). The M1-layer
enables the integration of various divergent EA into the
normalizing "common ground" of the M2-layer EME structure
and life cycle management process. The M1 and M2 layers together provide
a "translation" mechanism for the variant M1 EA of the D&A, while
providing the integrated/normalized/whole-enterprise M2 view of the aggregate EA content.
The M0-layer is the set of populated M1
EA profiles from the D&A, incorporated into commonly available public FEB
information collected and
modeled the M2-layer in the Reference Catalogs (with their standard Relation
Types), to provide both an enterprise-optimal
aggregate EA, and the
enterprise-suboptimal D&A variant EA seen in the context of the broader
FEB EA.
***************************
If you wish to unsubscribe
from the KM listserv,
please contact
john.andre@xxxxxxxx