Re: [ontolog-forum] Are Classifications nothing more than Indexes?

[William Frank <williamf.frank@xxxxxxxxx>]

Library science is a key source of classification knowledge.  And classification by a given description is what ontology and 'inheritance' are really about, I believe. Ranganathan's faceted analysis, mentioned by John Sowa, was an important part of my breaking away from  'class inheritance hierarchies.'   In 1989, we even used a  2-D icosohedron picture on which to depict as many as 10 facets of a thing.   

I think there are only two things missing in your account below:

1. casting everything as a noun. 

It is generally a good organizational practice to  have at least two fundamental categories of things -- if everything is one meta-category of thing, the category serves no purpose.  More generally, if everything is an X, being an X is a tautology, and it carries no information to say it is an X.  I have found it most useful to cast every thing either as a relation, a role in a relation, or an atomic individual (also recasting relations and propositions as higher order individuals in higher order expressions).    

2.  Organizing everything into domains. 

Domains are the most important way to classify things, but

There are classification schemes that are domain independent.  These are the sort of abstract categories one finds in some 'upper' ontologies, and the kinds of categories found in case grammars and philosophy: agent, efficient cause, ...

(It is not clear if this holds for semantic contentful descriptors.  For example,  if we have a classification of things by their colors, one would think this would apply to all macro-physical things, whether automobiles or stars or birds or paints.   But, supporting your point, astronomers and auto enthusiasts have their own separate color schemes. (Never heard of a star that was British Racing Green.)  Similarly, sizes of sub atomic particles are a very different thing from sizes of dogs.   However, 'size' itself, as a concept, belongs someplace in some classification scheme for concepts.  One that is domain independent.

Wm

On Sun, Aug 11, 2013 at 11:22 PM, Frank Guerino <Frank.Guerino@xxxxxxxxx> wrote:

Hi,

Very simply… Are classifications nothing more than what Librarians call "Indexes"?

As part of my involvement with the IF4IT, I've been involved in a project called NOUNZ that attempts to figure out and represent Taxonomies and Ontologies visually (both with text and/or with interactive pictures) through the creation of an Electronic Library that helps facilitate Knowledge Management.  In doing so, we followed a common Library Management paradigm that drove us to leverage common Library Constructs…

• Master Catalogs (Catalogs of Catalogs)
• Domain Catalogs (Catalogs of things that are specific to homogeneous topic domains)
• Indexes (ways of grouping things)
• Entities (those individual elements that represent nodes in space like media such as books, articles, etc.) 
• Semantic Relationships Between Entities (relationships between entities that have contextual meaning)

In the process of analyzing and trying to provide solutions for each of these common Library Constructs, and specifically in the context of analyzing Indexes, we keep coming to the conclusions that

1. Indexes are nothing more than a "classification" constructs, and
2. Indexes highlight how ineffective hierarchical representations of Taxonomy really are because "Electronic Indexes" can be and commonly are recursive, also "lopping-back" on and leveraging each other.

For example, if we take "Services" as the domain of discussion, we can first index or classify the Services by Type and by Importance…

• Services
• Services by Type (e.g. Strategy Services, Delivery Services, Support Services, etc.)
• Services by Importance (e.g. High Importance Services, Medium Importance Services, Low Importance Services, etc.)

Computer programming allows us to add elements of recursion and feedback to achieve further sub-indexing or sub-classification that does not scale for humans.  For example, the above becomes…

• Services
• Services by Type (e.g. Strategy Services, Delivery Services, Support Services, etc.)
• Service by Type further classified/indexed by Importance (e.g. Strategy Services that are of Critical Importance, Strategy Services that are of High Importance, etc.) 
• Services by Importance (e.g. High Importance Services, Medium Importance Services, Low Importance Services, etc.)
• Service of Importance further classified/indexed by type (e.g. Critical Importance Services that are Strategy Services, High Importance Services that are Strategy Services, etc.)

In our work, we've come to the conclusion that Indexes and Classifications are 100% synonymous.

I'd like your opinions and insights on the topic.  Do you believe this is accurate or not (and if not, why)?

My Best,

Frank

--
Frank Guerino, Chairman
The International Foundation for Information Technology (IF4IT)
http://www.if4it.com
1.908.294.5191 (M)

_________________________________________________________________
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
 

_________________________________________________________________
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    (01)