Ontologies and SOA KenLaskey v2 1 20060810

Uploaded from authorPOINTLite
Views:
 
Category: Entertainment
     
 

Presentation Description

No description available.

Comments

Presentation Transcript

Ontologies and SOA or Isn’t Discovery a Wonderful Thing?: 

Ontologies and SOA or Isn’t Discovery a Wonderful Thing? Ken Laskey co-editor SOA Reference Model

Talk about SOA and you invariably talk about discovering things : 

Talk about SOA and you invariably talk about discovering things

For example, : 

For example, Talk about SOA and you invariably talk about discovering things

Slide4: 

But... How did I know what properties were used to describe the thing I was searching for? How did I know what typical values were applied to the properties?

Slide5: 

According to SOA-RM, structure isn’t enough Within a street address structure, the city name and the street name are typically given the same data type – some variant of the string type. However, city names and street names are not really the same type of thing at all. Distinguishing the correct interpretation of a city name string and a street name string is not possible using type-based techniques – it requires additional information that cannot be expressed purely in terms of the structure of data.

Slide6: 

According to SOA-RM, structure isn’t enough Within a street address structure, the city name and the street name are typically given the same data type – some variant of the string type. However, city names and street names are not really the same type of thing at all. Distinguishing the correct interpretation of a city name string and a street name string is not possible using type-based techniques – it requires additional information that cannot be expressed purely in terms of the structure of data. The semantics of the property must be clear - is it street or city I am providing / you are searching on?

Slide7: 

Again from SOA-RM, There is often a huge potential for variability in representing street addresses. For example, an address in San Francisco, California may have variations in the way the city is represented: SF, San Francisco, San Fran, the City by the Bay are all alternate denotations of the same city.

Slide8: 

Again from SOA-RM, There is often a huge potential for variability in representing street addresses. For example, an address in San Francisco, California may have variations in the way the city is represented: SF, San Francisco, San Fran, the City by the Bay are all alternate denotations of the same city. So not only am I looking for a property of city but I have to be able to consistently interpret the value given to that property.

Slide9: 

Another example... My color vocabulary just deals with basic colors: red, green, blue, yellow, orange, purple

Slide10: 

Another example... My color vocabulary just deals with basic colors: red, green, blue, yellow, orange, purple We agree on a vocabulary including the property of color.

Slide11: 

Another example... My color vocabulary just deals with basic colors: red, green, blue, yellow, orange, purple We agree on a vocabulary including the property of color. You search for a shirt in a nice mauve.

Slide12: 

Another example... My color vocabulary just deals with basic colors: red, green, blue, yellow, orange, purple We agree on a vocabulary including the property of color. You search for a shirt in a nice mauve. You find nothing because my property value vocabulary doesn’t include the same values as yours.

Slide13: 

So what is needed for the distributed world of SOA-RM?

Slide14: 

So what is needed for the distributed world of SOA-RM? Clearly defined indication of the vocabulary from which properties originate

Slide15: 

So what is needed for the distributed world of SOA-RM? Clearly defined indication of the vocabulary from which properties originate Clearly defined indication of the vocabulary from which property values originate

Slide16: 

So what is needed for the distributed world of SOA-RM? Clearly defined indication of the vocabulary from which properties originate Clearly defined indication of the vocabulary from which property values originate Eventually, clearly defined mechanisms for mediating between clearly defined vocabularies

Slide17: 

If not ontologies, then what?