Stephen Downes

Matthias Melcher discusses my recent criticism of one of George Siemens's posts. In that post, I express the concern, with some justification, that the language is too vague to be descriptive. His explanation is that the vagueness is because a word like 'knowledge' has several senses. But we can be more precise about other words, such as, say, 'concept'. I offer a comment at the end of the article suggesting that the vagues occurs because the entitities we discuss when talking about networks are sub-symbolic. So there is not a nice neat correspondence between the physical properties of a network and the semantical properties of, say, a sentence.