Bug vs bug?
My friend Chris has a great question:
> Did some reading on the issues in building the connection between items that have multiple definiations. Such as the word bug. Is it a software bug or a insect? Do you think it may be easier to take a clean slate approach and do it right versus trying to smash something into XML such as RDF?
My response is as follows:
Good question. If it's just the string "bug", then context is king. Through statistical analysis, you can determine to some relatively high degree that bug is a software problem or an insect.
Now, on the web, "bug" can also be a Resource (see WWW Architecture for what a Resource is). If it's a Resource, it has a URI. Therefore, it's not "bug", but http://example.org/bug. Now, that's globally unique. Now that the name is unique, it's easy to all say things about it without being ambiguous.
Of course, you can say that http://www.example.org/bug is a software problem and I can say it's an insect. But this is where ontologies come in, for example. They let us formally describe things in the world, such that we can easily determine if think the thing is two different *types* of things. But at least we're talking about the same thing, even though our perceptions of it are completely different.
The important thing to note is that there's a big difference between wide scale web searching, like what Google does, and knowledge integration, which the semantic web does.
> Did some reading on the issues in building the connection between items that have multiple definiations. Such as the word bug. Is it a software bug or a insect? Do you think it may be easier to take a clean slate approach and do it right versus trying to smash something into XML such as RDF?
My response is as follows:
Good question. If it's just the string "bug", then context is king. Through statistical analysis, you can determine to some relatively high degree that bug is a software problem or an insect.
Now, on the web, "bug" can also be a Resource (see WWW Architecture for what a Resource is). If it's a Resource, it has a URI. Therefore, it's not "bug", but http://example.org/bug. Now, that's globally unique. Now that the name is unique, it's easy to all say things about it without being ambiguous.
Of course, you can say that http://www.example.org/bug is a software problem and I can say it's an insect. But this is where ontologies come in, for example. They let us formally describe things in the world, such that we can easily determine if think the thing is two different *types* of things. But at least we're talking about the same thing, even though our perceptions of it are completely different.
The important thing to note is that there's a big difference between wide scale web searching, like what Google does, and knowledge integration, which the semantic web does.