Mike, (01)
The purpose of the Zachman framework was to distinguish and
support the many points of view of different people involved
in the design and development process. (02)
> That kind of makes sense, but I wonder if there is a danger
> of conflating the semantics of the development with the
> semantics of the problem domain. (03)
As Ed observed and I agreed, Zachman's framework embodies
a particular design methodology that is not ideal for all
kinds of applications. But I believe that the underlying
ideas can be generalized to some important principles: (04)
1. Many different kinds of people are involved in using,
developing, managing, and maintaining any large system.
For example, consider a hospital. (05)
2. There is a common core of people, places, things,
and events that involve everybody in the enterprise
(e.g., patients, physicians, medicine, equipment,
schedules, building, air conditioning, ...). (06)
3. But everybody involved in the enterprise has a very
specialized job, and it's impossible to define a single
universal ontology that covers all views in detail
(e.g., the views of the patients, nurses, surgeons,
physicians of every specialty, administrators, janitors,
accountants, shipping and receiving, etc.). (07)
4. All those people have to communicate with one another,
directly or indirectly, but the messages are always
expressed in a narrow task-oriented vocabulary.
No single person in the enterprise is aware of all
the details. Even for the common core, each person
focuses on very specialized aspects for different
purposes. (08)
> The ZF has one box which is clearly labeled "semantics"
> which is where I would expect the problem domain semantics
> to go (the meanings of terms like stocks, bonds, people,
> employees etc.) (09)
First of all, you have to distinguish the columns and the rows.
The columns represent the answers to six question words:
What? How? Where? Who? When? Why? (010)
The rows represent the perspectives of five kinds of people:
Planner, Owner, Designer, Builder, Subcontractor. (011)
The columns are general, and those questions would be
important for any subject of any kind. If anything, you
might add a few more questions. Ramon Lull, a 13th century
poet, philosopher, theologian, and missionary, developed
a methodology (his Ars Magna) with ten question types,
each of which could be answered by something in one of
Aristotle's ten categories: Whether? What? From what?
Why? How much? What kind? When? Where? How? With what? (012)
My major criticism of the Zachman framework is that he
didn't have enough rows. For example, he didn't have a
row for the user of an information system. That is
partially excusable for the reason that every user
has a different point of view. (013)
> I can see that the "Business" row covers answers from real
> business people about aspects of the job they do. (014)
Yes, but what people? Just consider the hospital. Who is
the user? The patient? The physician? The nurse? The
accountant? The visitors? Every one has a different point
of view, and there is no single ontology that can cover
everything that each of them talks about and uses. (015)
> A role in the development of the solution, that is, not
> a role in the steady-state operation of the business (as
> compared with your later example of nurses etc.). (016)
To a certain extent that is true. But as we have all seen,
no computer system is ever "finished". IBM had a euphemism
"functionally stabilized", which they applied to obsolete
products, for which maintenance was discontinued. (017)
In fact, any business enterprise that isn't constantly
evolving and developing is probably functionally stabilized
in the same sense -- i.e., they're obsolescent, and they're
not likely to be around very long. (018)
> I fear that semantics technology is about to go the same
> route, I have already had to listen to techies explaining
> to me how their model now has "semantics" and even that
> they can generate these "semantics" automatically from
> their ever-inscrutable model. (019)
Any notation that anybody uses for any purpose has syntax,
semantics, and pragmatics. I agree that many people throw
those terms around as colorful buzz words without any regard
to their technical senses. (020)
> Surely one needs to capture perspective in the ontology,
> as appropriate for that particular problem domain. Here
> I guess the Why column perhaps comes into its own? (021)
The answer to the question "Why?" is purpose. Every program
has a purpose, and it's important to recognize that. (022)
One of my major complaints about OWL is that its focus
on computational efficiency forces the knowledge engineer
to write OWL for a particular kind of computation for a
particular purpose. (023)
The people who write OWL are *not* writing an ontology.
They are first and foremost *programmers* who are using
a very awkward kind of logic-programming language. (024)
> I would love to see some of the thinking that goes on in
> here being taken into account in future development of
> the standard. (025)
So would I. But what I find so frustrating is that these
ideas are not new. The database people were talking about
conceptual schemas in the 1970s. I developed my first
version of conceptual graphs as a system of logic for
natural language processing. But in order to make my work
more relevant to my employer (IBM at the time), I emphasized
its application to relational databases. Following is my
first published paper on conceptual graphs: (026)
http://www.research.ibm.com/journal/rd/204/ibmrd2004E.pdf (027)
That was published in 1976, and the abstract is just as
relevant today as it was then: (028)
A data base system that supports natural language queries
is not really natural if it requires the user to know how
the data are represented. This paper defines a formalism,
called conceptual graphs, that can describe data according
to the user's view and access data according to the system's
view. In addition, the graphs can represent functional
dependencies in the data base and support inferences and
computations that are not explicit in the initial query. (029)
I wasn't the only one who made such statements in the 1970s
and developed technology to support them. The R & D in AI has
progressed very far since then, but OWL is still mired in a way
of thinking that was recognized as obsolete thirty years ago. (030)
John (031)
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Subscribe/Config: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/
To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx (032)
|