Saturday, May 17, 2014

British Library posts 1.3M scans of book photos

When I visited the British Library a few years ago, I was in awe at the majesty of the place.  Not to mention the double-rainbow I saw while sitting outside the RBC Capital Markets building...

Last year, The British Library's Mechanical Curator project, using facial recognition tools and Flickr hosting, published over 1 million historical scans from books of the 1700-1900s.  It's a fascinating example of big data at work.  Indexing and curating such a large collection would be nearly impossible without crowd-sourcing and complex algorithms to analyze the imagery.

Imagine a world without Google, or any search engine on the internet.  How would you find anything?  What if you had to browse through 200 year-old books to find a URL to a web site, then manually search through each link on a web site to find a photo?

For all of the 30-50 billion or so web pages on the internet, there is still a world outside that contains far more information than will ever be digitized, categorized, or classified.

After visiting London, my retirement dream is to have a small boat moored for a couple of months each year near the British Museum and British Library.  Nothing fancy, as long as it has internet access and a sitting room for reading a book or two...

No comments: