In this section, we look at the Artefacts of PEAF. The
artefacts deal with the structures required for Structural and Transformational
description. The WHAT.
Interestingly it is the thing that most frameworks and
ontologies focus on, which when you see the other quite important things like
Methods, Culture and Environment might make you wonder how much value they
It is understandable however, that people concentrate on
Artefacts as they are very “tangible” - the things being produced and consumed
- but also because many frameworks come from IT departments and IT people, and
IT people love datamodels, ontologies and Meta-models, and those types of
things are reasonably easy to produce. Methods, Environment and Cultural things
are much harder to produce and therefore tend not to be.
The artefacts defined in PEAF constitute the EA Meta-model
and some of its content such as Principles.
There are many Meta-models that already exist (for the
reasons defined above) and the lack of a Meta-model has never really been a
reason why EA initiatives fail. For this reason, PEAF does not concentrate on
defining a complete and detailed Meta-model (no point reinventing the wheel).
However, PEAF does defines some important things that are usually missing and
also exposes a higher level structure that people can use to understand how the
detailed Meta-models they wish to use fit and relate.
Unfortunately, this is not an easy task because all the
myriad Meta-models out there do not share any common backbone or structure
(such as MACE and MAGMA). Even if you can figure out which parts of which
Meta-models you wish to use, integrating them into a whole is also fraught with
difficulty, again because there is no common meta-meta-structure (such as MACE
This is exacerbated by Tool Vendors not providing facilities
for the importation and separate management of different Meta-models from
different sources along with the functionality to then present and manage and
use them as a coherent whole.
This whole problem area is an area that Pragmatic is currently working on, driving
Meta-model providers to structure their Meta-models around a Structural
Ontology (MACE), a Transformation Ontology (MAGMA), all expressed at different
levels of abstraction (Enterprise Context, Contextual, Conceptual, Logical,
Physical, Operational) and driving Tool Vendors to think about how to then use
them and allow people to use and manipulate them.
This will be a long and hard road (some say impossible), but
one that will reap fantastic benefits for Enterprises globally. These 2 things
(doing the impossible and fantastic rewards for customers) are the things that
mean it’s something that Pragmatic has a
passion to do.
In the meantime, Pragmatic
will continue to map existing Meta-models to MACE and MAGMA to provide
Enterprises with, at least, a starting point to understand how they relate and
therefore how to adopt them.
Do you use multiple
What Meta-models do you
Do they have limitations?
How do you integrate them?