Symbol Grounding and Extensible Aggregators
Jon Udell,
Aug 12, 2003
This is a difficult article (you can tell by the fact that even the title needs a little interpretation). But it gets at the heart of a debate that is gradually engaging the entire XML community. In a nutshell: how do you know what the names in XML tags mean? In traditional XML, they are defined in a fixed vocabularity (such as IEEE-LOM). In RDF, they are defined by various name-spaces, which may be mixed and matched. For myself, I fall firmly into the second camp, though I am not dogmatic as to whether RDF is used, just so long as I can define terms on the fly. I will have more on this and will try to find a nice outline. Tim Bray advances, by way of reply, the sceptical line: only humans do semantics, and the namespaces themselves don't matter one whit. Bray is partially right, but for the wrong reasons. For my own part, I have been re-reading Ludwig Wittgenstein recently in this regard. Because, while I think that we can't just define our way into meaningful metadata, it doesn't follow that we must therefore do without meaning. Meaning is use, according to the famous slogan, and while I don't think it makes a great theory of cognition, I think it makes a fantastic theory of metadata.
Today: 3 Total: 3 [Direct link] [Share]
