Initially I didn't really want to deal with a semantic meta language for the web.  I've been hearing about the "semantic web" for quite some time now, and (honestly) I am getting a bit fatigued by it. Having read the quick overviews I decided to go in and read the chapters provided by Pierre Levy. I have to say that it is quite interesting.  Semantics isn't really my thing - don't get me wrong, I would like to like semantics (I find the study of semantics fascinating) but I haven't had enough exposure to be fully conversant in it.  The second chapter dealt more with topics that I have worked on before, including information organization, librarianship and knowledge management.  Chapter 3 I will be reading this weekend (and after that maybe read some of this week's Change11 stuff).

There is one problem, I see, with the current state of classifications: they need to be learned and applied. Folksonomies change - I mean look at my change11 posts.  My initial inclination was to use ChangeMOOC as the tag for my posts, until I realized that gRSShopper wasn't picking up ChangeMOOC and I needed to write #Change11 as the tag. For a time frame I used both, and now I just use the one that gRSShopper picks up. Folksonomies adapt, but taxonomies need to be learned and applied and people generally don't often want to do that unless they are professionals in records/information management and it's their job to do so (do I as a blogger want to categorize all my blog posts to a "t"? not really - I do just enough to get by).  Machine translation is still imperfect, so asking machines to auto-classify is not going to yield good results - so where do we go from here?

I don't disagree that a meta-language, some classification universals, can't be beneficial to all of us, but who does it? What's the "benefit"? and how do you prevent junk classifications, #like #the #people #who #hashtag #every #single #word #in #their #twitter #post ? - just some thoughts :-)