The semantic web work is interesting because it finds classification and taxonomies based on metadata in the html. You use informative “div” tags for instance to provide context / categorization of the content on a web page. In order for this to work, you need a (set of) taxonomy(ies) and a controlled vocabulary. You also need the rules for how you use both. All of this comes together to build an Ontology for web data.

I see the semantic web fitting into a communication and collaboration environment as one layer – a fairly high layer for structured information. The semantic web layer also has a high overhead for creation of the layer and for maintenance of information. This is why you need a 747 full of Librarians to implement it.

Folksonomies and Social Software operate at a lower easier less-formal layer. No ontology. No data dictionary. No controlled vocabulary. What you gain is very easy sharing of information, the chance for social discovery (“he is interested in this too?”) and the ability quickly find and sort based on a few key ideas. Clay Shirky did an interesting analysis of the tags for Ontology is Overrated Summer Remix posting. He found that it only took 10 people to establish what the top 3 tags are for the post. People are very good and interpreting information and drawing out the key themes. We should be, we have been asked to do this since 4th grade Folksonomies and Social Software rely on the wisdom of the masses to build their structure – not the expertise of the few – or the 747 full of librarians.

Thanks for the comment.

Jim

my question would be, is Folksonomy/social tagging, the first answer for idea written down by mr Berners-Lee

http://www.sciam.com/article.cfm?articleID=00048144-10D2-1C70-84A9809EC588EF21&ref=sciam