The hidden side of linked data
Two of my previous posts have raised a few issues that I find very interesting and lead to some more un-answered questions. One comment in particular is what got me thinking.
What happens when information about us on the web is no longer disparate, no longer hidden in the database — locked away by thousands of different privacy policies that non of us read (at least I don’t). We all leave a trail on the web no matter how socially networked we are. Every quick response to a blog entry, every review on Amazon, every rating on eBay; all lead to a clearer “picture” of us . This picture shows more of “who” we are and “what” we do and maybe even “why” we are doing it.
This can be a bit scary. Actually, this IS scary. Do I really want someone to know all the things I have purchased on eBay, all the books I reviewed on Amazon and all the replies I have made in blogs — all conveniently located in one nice little search? Or all nicely linked together with intelligent data embedded as microformats or RDFa.
This information is publicly available and if someone was really keen could go out and scrape this data together. As a matter of fact there are an increasing, yet small, group of people search engines out there already. But these search engines are working like Google works, with mostly unstructured, raw data. The people search engines are tuned to work with raw text and some other key identifiers to build up there profiles. Brute force is pretty much the key.
Microformats, RDFa, RDF, FOAF, Open Social (and more) all are building a more structured web, one where brute force is no longer the only way to find connections and find things. There is a lot of unknowns about these previous hidden connections becoming very prominent and more importantly, easy to find. If a very educated audience of microformatters are surprised by what is connected, what happens when the general web audience starts to see these things apparently “appear” out of thin air?
I don’t think they will be happy, not at all.