ANY BLOG. Desire and Design

ANY BLOG is a programming experiment that aims at the design and creation of
consistent blogs by relational means, regardless of content. It implements
different programming concepts –crawling, indexing, ranking, and API
interaction– widely used by search engines and social networks.
 
Any Blog constantly updates an Index of posts from Tumblr, a micro-blogging
platform and social network. The Index is modified upon API requests and a
personalized web-crawler. Images, videos, texts, hyperlinks, and their metadata
become part of the Index when a blog-seed from Tumblr posts or reposts any
content.
 
The consistency of a new blog is double articulated by a cluster of blog-seeds
and by a unique repost-threshold: (1) In principle, every Tumblr blog has its
own consistency by the population of posts it holds together, and a cluster of
blog-seeds sketches a world of potentials from which new consistencies can be
designed. The desired blog, rendered in a cluster, is ultimately defined by the
user. (2) The repost-threshold is the customized procedure that sets the
conditions in which a posts from the cluster may be part or not of the new blog.
It neglects content by considering only its metadata –’likes’, number of times
it has been reposted, and popularity rank of its origin.
 
Finally, an operational index is created so all the Tumblr API requests needed
for the new blog are implemented.
 
As of 31.03.2014 the project runs with:
Blog-seeds – 40
Posts in Index – 7.000
New Blogs – 2