Re: [oor-forum] OOR Architecture & API Workshop-XI - Tue 2012.03.20

[Todd J Schneider <todd.schneider@xxxxxxxxxxxx>]

Leo,

Your assertion may well be true. But is it worth the
effort? Is there sufficient value in having a graphical
representation available when there may be alternate
mechanisms that satisfy the requirement?

For the OOR that requirement is to able find an ontology
that meets a user's (system/project/whatever) requirements.
The ability to first identify candidate ontologies and
then be able to investigate the candidates in details to
see if they'll meet the user's requirements. Perhaps
this could be met be one or more queries against an
OOR instance.

Todd

"Obrst, Leo J." ---04/03/2012 03:23:57 PM---Todd, Every language can be represented graphically, though a given ontology language may not have i

From: "Obrst, Leo J." <lobrst@xxxxxxxxx>
To: OpenOntologyRepository-discussion <oor-forum@xxxxxxxxxxxxxxxx>, John F Sowa <sowa@xxxxxxxxxxx>
Date: 04/03/2012 03:23 PM
Subject: Re: [oor-forum] OOR Architecture & API Workshop-XI - Tue 2012.03.20
Sent by: oor-forum-bounces@xxxxxxxxxxxxxxxx

---

Todd,
 
Every language can be represented graphically, though a given ontology language may not have implemented such a visualization. Here is my contrived graph of a Cyc Meld _expression_:

From: oor-forum-bounces@xxxxxxxxxxxxxxxx [mailto:oor-forum-bounces@xxxxxxxxxxxxxxxx] On Behalf Of Todd J Schneider
Sent: Tuesday, April 03, 2012 10:52 AM
To: John F Sowa
Cc: OpenOntologyRepository-discussion
Subject: Re: [oor-forum] OOR Architecture & API Workshop-XI - Tue 2012.03.20


John,

While I can't disagree with your suggestions, from an
OOR perspective the issue I attempted to raise was
more pragmatic in nature. That is, we assumed that
every OOR representation language module would have
a way to graphically represent an ontology. This is
an incorrect (implicit) assumption. We'll need to
modify the requirements to reflect this.

Some of the notions of a foundational logic you listed
may prove useful in specifying OOR UI requirements.

Todd

John F Sowa ---04/03/2012 10:42:41 AM---On 3/20/2012 1:11 PM, Todd J Schneider wrote: > We discovered we had an implicit assumption that the

From: John F Sowa <sowa@xxxxxxxxxxx>
To: Todd J Schneider <todd.schneider@xxxxxxxxxxxx>, OpenOntologyRepository-discussion <oor-forum@xxxxxxxxxxxxxxxx>
Date: 04/03/2012 10:42 AM
Subject: Re: [oor-forum] OOR Architecture & API Workshop-XI - Tue 2012.03.20


---





On 3/20/2012 1:11 PM, Todd J Schneider wrote:
> We discovered we had an implicit assumption that the common notions
> of 'class/concept' and 'relations/properties' would
> be valid for all representation languages.

That is why I prefer to use a foundational logic with a minimum
amount of terminology. Then all other terms can be defined,
explained, or related to one another through the basic terms.
You don't even have to agree on a specific logic or notation.
Just use the following minimal set of terms, which can be
related directly to many different logics.

Minimal set:

1. Boolean operators.

2. Quantifiers (some & every, or the symbols for them).

3. Relations or predicates.

4. Sets.

Translations:

1. Use 'monadic relation' for every occurrence of property, attribute,
characteristic, feature, facet, concept, or type.

2. Use 'dyadic relation' instead of role.

3. Treat a class as a composite of a type and a set. Then you can
clearly distinguish whether you mean the monadic relation that
specifies the type or the set of all instances of that type.

4. Instead of using the modal terms necessary, mandatory, required,
or obligatory, state some constraints that must be true.

Something is necessary, mandatory, required, or obligatory
if it is implied by the constraints. It is possible,
optional, or permissible if it is consistent with the
constraints.

John
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/oor-forum/
Subscribe: mailto:oor-forum-join@xxxxxxxxxxxxxxxx
Config/Unsubscribe: http://ontolog.cim3.net/mailman/listinfo/oor-forum/
Shared Files: http://ontolog.cim3.net/file/work/OOR/
Wiki: http://ontolog.cim3.net/cgi-bin/wiki.pl?OpenOntologyRepository

_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/oor-forum/  
Subscribe: mailto:oor-forum-join@xxxxxxxxxxxxxxxx 
Config/Unsubscribe: http://ontolog.cim3.net/mailman/listinfo/oor-forum/  
Shared Files: http://ontolog.cim3.net/file/work/OOR/ 
Wiki: http://ontolog.cim3.net/cgi-bin/wiki.pl?OpenOntologyRepository     (01)