Information Shadows

By far one of the coolest reads I’ve seen. Thanks for that.

I too really enjoyed this, but I’m having a bit of a difficult time understanding what an information shadow really is.

• For all things, if has a unique, pervasively accessible identifier ( UPC / Thing taxonomy / Binomial Nomenclature / etc. ) it has an information shadow.

Taking the Tom Coates paraphrase:

Paraphrasing Yahoo!’s Tom Coates, first we learn to digitally point at a thing’s information shadow, then we can glue information handles to it. Once the shadow has handles, we can grab and throw the information around.

So that would be saying “This book has a UPC” ( pointing, above ). The Amazon API would be like “gluing handles” because now I can allow programmatic access to the object as its set of properties. The “information shadow” is the “mentions” of that object, in sum.

Thus flickr picture “YYYY” is a unique ID and it can be mentioned in its tags “Rome coliseum ancient archietucture” …and…can be mentioned in my blog post “I was just in rome and (insert pic )” AND can be part of the “Best of Rome set” and thus the information shadow of YYYY is, effectively, the commentary of the object, in aggregate. In theory, across a semantic web, a definitive information shadow for YYYY would pull all allocations, all mutations of the the associated keyphrases, and then provide the ‘total shadow’ ?

Is that how you guys read it?

Because I’m obsessive like ’dat, I tried to re-organize Mike’s argument into something that, to my mind, had a bit more of a logical progression: précis here.