Nick wrote:
...It's only when you have a significant number of people
contributing from different contexts that large-scale stable patterns of classification
begin to appear.
Do they? I've been looking around a few of the 'tag cloud' using sites and I don't find much evidence of this or any other really useful 'groupings' of terms into 'relationship clusters' to help with finding stuff. Pretty chaotic the lot of them. Please point me at a good, example if you know one, anyone.
Looking at flickr for instance- for the tag paris, related tags today are just 3, : francia, rodin, airport . Thats a pretty random bunch at best - there's no meaning behind it apart from the fact that a photo has been tagged with paris and these terms by n people. It does not help me 'expand my query' from paris in any meaningful way. More hit and miss than Google? Tags are VERY specific - too specific to be very useful unless tied into something more formal? (c.f. Brian's earlier post)
cheers
Mike
|