I’ve been learning a lot this week, taking the time to work through a lot of files, and continuing to wonder whether there really are some “fundamental underlying principles that can be shared among various perspectives”. John’s comment might suggest that the communication between pet owner and vet illustrates this possibility in practice.
When I discovered the first edition of the Booch book in 1991, at the UCSB bookstore (Booch got a Master’s degree in Electrical Engineering from UCSB), I was instantly fascinated.
I was not then a “programmer” – but I was very excited about the philosophical fundamentals that seemed so well-illustrated in this book. I had been influenced by writers like Herbert Simon (Sciences of the Artificial) and was strongly attracted to hierarchical and taxonomic models. Booch’s comments and diagrams nailed it for me – and many of his points were strongly resonant with ideas I had absorbed years previously from Conceptual Structures.
His first four chapters go over the basics of design philosophy in ways I felt were very seminal – so I was happy to see this new PDF, of the third edition.
For ease of study and engagement, I transcribed the entire PDF to a series of word.docx files, and put them on line here:
http://originresearch.com/docs/booch/
More or less what I am trying to do with these documents is synthesize from them a minimal set of “basic fundamentals” – that might be compared or correlated with the guiding principles JS teaches, and maybe the basic ideas from Barry Smith and others in the Applied Ontology book. It might be very tough to agree on the details – maybe impossible – but there might (?) be some very obvious basic fundamentals (“what is hierarchy?” “what is abstraction”? “what is a class?” “what is an object?” “why is a taxonomy totally context-dependant?”) that could form a kind of “logic-kernel” or seed from which (or towards which) a lot of semi-independent and overlapping approaches might branch or converge….
Sample chapter (“Classification”) in word.docx:
http://originresearch.com/docs/booch/booch_CH04.docx
These first four chapters are like a high-end “Cog Sci 101 for Engineers”
- Bruce Schuman
From: ontolog-forum-bounces@xxxxxxxxxxxxxxxx [mailto:ontolog-forum-bounces@xxxxxxxxxxxxxxxx] On Behalf Of Ravi Sharma
Sent: Wednesday, June 03, 2015 3:59 PM
To: [ontolog-forum]
Subject: Re: [ontolog-forum] The "qua-entities" paradigm
On Wed, Jun 3, 2015 at 3:25 PM, John F Sowa <sowa@xxxxxxxxxxx> wrote:
On 6/2/2015 10:46 AM, Bruce Schuman wrote:
> "Object Oriented Design, With Applications", by Grady Booch (1991)
> http://www.nicolasdanino.com/tongji/OOAD_Booch_3rd_Edition.pdf
>
> The book is rich with charming illustrations on fundamental issues
> of hierarchy and abstraction.
That's fine. I'll point to a few of them to illustrate some general
points. See page 46 (or p. 70 as Adobe counts) for a picture of a cat
from the point of view of the pet owner and a veterinarian. One views
it as a purring fuzz ball, and the other sees a collection of parts.
> Do these principles illuminate the argument for a revolutionary
> simplification - consistent across disciplines and industries
> and cultures?
Since some people play both roles, the two views aren't contradictory.
But it's difficult to entertain both at the same time or to draw
coherent inferences from a mixture of the two descriptions. That is
not a revolutionary observation. But it shows that the goal of
a single universal, consistent foundation is too simplistic.
The notion of encapsulation on page 51 (or 76) shows a way to
accommodate both views: simplify (or underspecify) the ontology
in order to ignore details that are treated differently in each view.
That is what Schema.org does.
Modularity (diagram on p. 54 or 79) is a related way of ignoring
troublesome details. The detailed ontologies of different modules
may be inconsistent with one another. But they can interoperate
by passing messages that use terms at the underspecified level.
When they ignore details, the pet owner and the vet can communicate.
Fundamental principle: heterogeneity and diversity are essential.
John
--
Thanks.
Ravi
(Dr. Ravi Sharma)
313 204 1740 Mobile