ontolog-forum
[Top] [All Lists]

Re: [ontolog-forum] Just What Is an Ontology, Anyway?

To: "[ontolog-forum]" <ontolog-forum@xxxxxxxxxxxxxxxx>
From: Phil Murray <pcmurray2000@xxxxxxxxx>
Date: Thu, 29 Oct 2009 20:11:47 -0400
Message-id: <4AEA2F43.8000004@xxxxxxxxx>
Steve --    (01)

I doubt that anything you have ever said has wasted anyone's time, let 
alone my time.    (02)

I distinctly remember one of your early presentations on the HyTime 
Architectural Forms at Hypertext 89 in Pittsburgh. I was overwhelmed, 
but you certainly didn't waste my time.    (03)

SN> Phil, I have the feeling that you're (very kindly) saying that we 
should
 > use screwdrivers to drive screws, and hammers to drive nails.    (04)

Yes, that's certainly part of my intent. Every once in a while I raise 
my hand cautiously in an attempt to point out that the work done in many 
different projects and documents by the members of this forum is not 
even close to reaching its full possible impact because that work is not 
seen as part of what we do every day in order to create value.    (05)

By that, I *do* mean that    (06)

 > "connecting knowledge with reality" is one kind of task (involving
 > "knowledge" and "reality") and expressing correspondences between
 > subjects of conversation is something else entirely.    (07)

With apologies to members of this forum for straying off the narrow path 
of discussions of ontologies, let me give a simple example from recent 
personal experience -- in this case, watching my wife's frustration at 
being unable to open a file on her new laptop:    (08)

The non-geek computer user really has no way at all of knowing that in 
Windows environments, you can't open a file unless you first set up 
associations between software applications and the types of files they 
produce (or are used to view). At least not without hours of searching, 
a prolonged discussion with experts, or intense interaction with more 
experienced users.    (09)

Of course, some of those correlations are available out of the box. But 
your initial novice-level experience successfully opening some files may 
lead your to think that this process happens automatically ... and that 
may increase your misunderstanding and frustration when your encounter 
new types of files. This is exactly what happened to my wife. 
(Everything worked to her satisfaction on her old computer, because I 
had set up the file associations for her.)    (010)

Most of the time, you'll get a message that asks you to pick an 
application to open a particular file type, but that will probably be 
the first time a novice is exposed to the idea of file types.    (011)

A useful representation of this practical knowledge might connect 
"statements" like    (012)

* Windows will open a file automatically only if it has a record of the 
application used to open that type of file.
* You can make (or change) those associations using Windows' "Default 
Programs" controls.
* Sometimes Windows will fail to open a file when you click its name or 
click a link to that file.    (013)

There are ways to express these connections both with ontology tools and 
with Topic Maps. But, IMHO, they're really clumsy. They're an extremely 
poor match for how we communicate with each other, for how we talk about 
reality, for how we believe we share knowledge, and for how we work.    (014)

There are applications out there that address some aspects of this level 
of communication -- that is, connecting several [unambiguous] 
descriptions of reality. For example, the Knowledge Management 
Institute's Compendium is an argumentation-support system that enables 
people to make evaluations of important statements about work (or policy 
or ...). Users can also use such tools to say "This is *not* important" 
or "This is *not* connected to (or influenced by) that."    (015)

My quarrel with such tools is that they, too, are inadequate for the 
challenges of representing meaning in ways that are helpful to knowledge 
workers. They don't scale. They're difficult to use because they don't 
leverage existing knowledge. And they are disconnected from representing 
knowledge in the rich, interlinked but essentially static ways that are 
characteristic of ontologies in general and of Topic Maps in particular.    (016)

I don't think that's vacuous, unless you assumed that I meant that we 
can share complete and perfect understandings of reality with others. 
Nobody over the age of 12 believes that's true. Representating 
statements about reality and the relationships among them in a 
structured, systematic way can be *sufficiently* close to shared 
understandings (in most cultural and work contexts) that it can be 
highly effective in solving a problem or enabling knowledge workers to 
see new possibilities for creating value. Most of the time. The 
representation is functional. My wife would surely have loved to have 
that knowledge at hand.    (017)

A graphic ("circles and arrows") representation of those statements and 
their relationships might be even more effective. But don't get me 
started on that.    (018)

(And if you communicate with Sam Hunting, tell him I said hello.)    (019)

    Phil    (020)

Phil Murray
Chief Knowledge Architect
---------------------
The Semantic Advantage
"Turning Information into Assets"
phil.murray@xxxxxxxxxxxxxxxxxxxxx
401-247-7899
Blog: http://semanticadvantage.wordpress.com
Web site: http://www.semanticadvantage.com    (021)


_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/  
Config Subscr: 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 join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J
To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx    (022)

<Prev in Thread] Current Thread [Next in Thread>