[10]                                home                        [12]

 

 

 

Key questions on Common Upper Ontology

 

 

4/19/2004 8:44 PM

 

 

 

A comment about the following:

 

<quote>

 

Some funding has been provided for knowledge representation languages, but very little for actual content. Results such as WordNet and CYC have been found useful to some extent, but they have great difficulty achieving wide use, not only due to technical limitations, but mostly because each such ontology is the product of a small isolated group with little input from the wider community. I will be happy to discuss this problem with anyone who is willing to pursue the question. There are sociological as well as technical barriers that have to be addressed.  Those who are only rewarded for doing things that are new and different will have little incentive to participate in large collaborations.

 

</quote>

 

There is, we suggest, an assumption that Knowledge Representation as computer ontology is without question something that can be done and done completely if only adequate resources are provided.

 

The promise of AI, and its ups and downs, often was justified on a similar assumption.  The assumption was that if only enough money and time was provided then the goals of artificial intelligence, which where not always precisely defined, would be meet. 

 

The arguments we have made, following the arguments made by Penrose and others, is that the computer does not have the non-locality and entanglement properties that one sees in natural emergence.  We will not repeat that argument here, as one is either interested or not in this line of thinking.