Tuesday, November 10, 2015

Internet Archive to The Commons

Image from "The Lepidoptera of the British Islands...."
Yesterday I received a very interesting email from the Annelida newsletter system. I made me aware of the huge amount of information that is available through Flickr's Internet Archive. Over a while they were populating this area with images of book pages. Here a little bit of background from their website:

The Internet Archive is best known for its historical library of the web, preserving more than 400 billion web pages dating back to 1996. Yet, its 19 petabytes include more than 600 million pages of digitized texts dating back more than 500 years. What would it look like if those 600 million pages could be “read” completely differently? What if every illustration, drawing, chart, map, or photograph became an entry point, allowing one to navigate the world’s books not as paragraphs of text, but as a visual tapestry of our lives? How would we learn and explore knowledge differently? Those were the questions that launched a project to catalog the imagery of half a millennium of books.

A Yahoo research fellow at Georgetown University, Kalev Leetaru, extracted over 14 million images from 2 million Internet Archive public domain eBooks that span over 5 centuries of content, compiling more than 14 million high resolution images spanning nearly every topic imaginable. Each image includes detailed descriptions, including the subject tags of the book it came from and the text immediately surrounding it on the page. The latter is especially powerful, as it allows to keyword search 500 years of images, instantly accessing particular topics or themes. Searching for love yields a myriad images of cherubs and courtship, while mortis (death) offers a glimpse into the early modern period’s fascination with the subject. A search for bird offers a vividly colorful showcase of the world’s bird species, while searching for telephone traces the invention’s history from its introduction as an electric novelty to its widespread adoption.

It is very easy to search through this vast amount of data. Here are some links to give you an idea:

As you can see the search can be easily narrowed down through modification of the hyperlink (after the "text=" section). 

Happy browsing.


h/t Geoff Read

No comments:

Post a Comment