Featured Online Identity Social Classification The Social Web

How we created a metadata profile

My tagcloud
This is me, these are words extracted from every 850+ post ever published on hypernarrative. This is meta data; data about data. It tells you what hypernarrative is about, and since it’s kind of a personal blog it also tells you something about me. These words are related to the things I care about.

The size emphasizes its frequency and thus importance. Sure, most of these tags are general words. The value is not in the individual tag but in the collection of tags.

The last years semantic classification in the form of tags got really big. Not mainstream, but big. Why? Because it is a raw classification, you don’t need to know classification structures to order data the same way scientists used to know to order species (how many legs does this creature have?). Tags are free and easy, I can tag objects the way I like, there are no rules and you don’t need any knowledge.

In the last years we have created a new layer to information on the web, we tagged hyperlinks, photos, video’s, presentations and articles. Almost every new website gives us the option to tag. Tags are the new categories. Categories are closed, tags are open and infinite.

My semantic profile
But what if we trace tags back? Tags tell you something about the tagger. What if you would trace back all my web accounts and the tags I have used. If you collect this information in a database you have a semantic layer about me. I have a semantic profile, and this is valuable.

It shouldn’t be difficult the come up with an application that can use this. For example you could make a social network that automatically connects you to other people that have similar interests through the tags they have used on Flickr (what they have seen for real) on YouTube (what they made) on Delicious (what they like) on blogs (what they care about) and more.

It would create a cross websites social network that doesn’t need an invitation system. It just connects you to minds alike based on what you have done already. Maybe it’s already out there.

Link to the picture on Flickr

5 replies on “How we created a metadata profile”

I also see the danger of communities becoming too boring this way. I mean if you’re only connected to likeminded…
But maybe this ain’t true because for now I’m only experiencing advantages of tags and social networks in general.

Good point by Renier. Same goes for the personalized Google-results. I don’t want to read what Google thinks that I want to read; I want to discover what is out there, independent of who I am.

Thing is that the more personalized content the become, the less possible is the chance to discover something completely new. I stumbled upon tens of interesting artists by just trying them. If I would only follow my advices, I would keep running in the same musical circle.

There’s a beautiful paradox here: we want to get (out of the endless mass of information) relevant content, but we also want to know what’s out there (independent of what we personally consider relevant, since this scope will be too limited).

I think you’re both right. We don’t like personalized content. We like the non intrusive ‘you might like’ personalization, but we don’t like a fully personalized service.

I like the recommendations Amazon makes. I don’t like a bookstore filled with only the books that Amazon thinks I should read.

It also frightens us, there is this computer that knows who you are based on your prior behavior. Even when this computer is right (there will be a time a computer can probably think up your next step) we are still frightened by it. Too much personalization takes away the feeling of freedom.

The signal to noise ratio, what we like is somewhere in the middle.

Comments are closed.