Comments on: MetaPerspectives, Metaphysics, and Metadata – Semantic Web as Nonsense

By: Sachbak

Sachbak — Sun, 09 Sep 2007 08:27:14 +0000

Personally, I think that since the point is agreeing on semantics, we will copy the mechanisms we use in the real world to the world of the semantic web.
When you think of it, when we express certain concepts in a conversation, we effectively negotiate their meaning, through spoken natural language. When we agree on the meaning – we can than gain some value from the conversation (pass a meaningful message).

The same goes for software agents in the semantic web – they talk about URIs (nouns) and agree about them using ontologies. The decision as to what is to be agreed upon is basically programmed into the software. You can say that the human prejudice is programmed into the software agents.

For example, I can program a computer program that contacts another agents using RDF and ontology. I can then choose to give more weight to properties and definitions given by your ontology as opposed to any other ontology, because *I* (the human me) trust *you* (the human you) better than I trust another ontology creator.

By: Josh

Josh — Sat, 08 Sep 2007 23:43:28 +0000

Cody,

I happen to largely agree with you. However, as I am sure you are aware, language is an accepted bastardization of what is in our minds. Reasoning is not enough. Reasoning may fall prey to cultural or religious factors.

If the Semantic Web is meant to fulfill the vision of a “machine-readable” internet, I think we have a shot at that. I just don’t know how useful it will be to me without an additional layer of decision making that allows me to impose myself upon It.

Great comment. Thank you.

Josh

By: Cody Burleson

Cody Burleson — Sat, 08 Sep 2007 23:00:50 +0000

If I have a concept in my mind and you have a concept in yours, unless we share a mind, that concept is different in each instance thereof.

That is true, while it exists only in our minds alone. As soon as we share it, however, we have to use language and all language is a set of concepts which must ultimately be resolvable to primaries or concretes. In a similar way that we reason against statements in a language, computers can reason against the same or similar statements. By “reasoning”, I am not talking about intellegence in a “human mind” sense. I just mean the simple ability to infer new meaning based on the context of statements and on statements that have already been made.

I think that the term “Semantic Web” was never meant to express one web that is semantically understood through the brilliant reasoning of some global system or anything like that. It simply means – a web that is now semantically enabled; a new “dark” web that is serving machines as much as the humans 9on our behalf, of course).

By: Josh

Josh — Fri, 07 Sep 2007 18:48:28 +0000

I responded to you on the Digg page. You are saying a semblance of what I am saying. The Semantic Web with all capital letters needs a Semantic Ontology. We can have semantic webs, but not an uber Semantic Web. Until all entities agree upon an ontology, we will not have Ontology. We will not have Semantic Web. I think we can have semantic webs within closed environments, but in the wild? Tell me what a “happy meal” is on the Semantic Web? We need additional layers, compromise, INSTANCES of the Semantic Web: semantic webs.

By: James

James — Fri, 07 Sep 2007 17:36:55 +0000

I said this on the Digg page for this entry, but it does more good to post feedback to the source. The Semantic Web doesn’t rely upon a single ontology driven by any single entity. The Semantic Web relies on ontology agreements. The more entities that agree on which ontologies they work with, the greater the strength of the Semantic Web will be.