Skip to content

Commit

Permalink
Update README to reflect most recent developments.
Browse files Browse the repository at this point in the history
  • Loading branch information
fako committed Nov 18, 2014
1 parent e7261db commit b40e139
Showing 1 changed file with 9 additions and 11 deletions.
20 changes: 9 additions & 11 deletions README.md
Original file line number Diff line number Diff line change
@@ -1,22 +1,20 @@
HIF
===
Data Scope
==========

The Hyper Information Framework is innovative software meant to empower everybody to execute complex search queries through divers media. Think about search queries like: “which websites name at least three experts from field X”, “which historical events took place at date X”, “what are iconical images for X in culture A, B & C” and “which people are known for similar reasons that person X is known for”. Without HIF such queries are almost impossible to do.
Data Scope is innovative software meant to empower everybody to execute complex search queries through divers media. Think about search queries like: “which websites name at least three experts from field X”, “which historical events took place at date X”, “what are iconical images for X in culture A, B & C” and “which people are known for similar reasons that person X is known for”. Without Data Scope such queries are almost impossible to do.


Innovation
===
==========

HIF divers radically from current search solutions in three ways. First of all the search happens semi-automatic, which enables users to combine different sources on the web. These mashups of information are most interesting when they make smart use of crowd sourced information, like information from Wikipedia. HIF delivers surprising results in this way. The current search engines deliver only the things that people expect to find.
Data Scope is radically different from current search solutions in three ways. First of all the search happens semi-automatic, which enables users to combine different sources on the web. These mashups of information are most interesting when they make smart use of crowd sourced information, like information from Wikipedia. Data Scope delivers surprising results in this way. The current search engines deliver only the things that people expect to find.

Secondly HIF breaks with the trend to improve results implicitly by tracking online user behavior. The belief behind HIF is that the “advanced search” screen should respond to the times. That's why HIF enables users to set search preferences explicitly. This not only gives more control to the user, but also improves privacy.
Secondly Data Scope breaks with the trend to improve results implicitly by tracking online user behavior. The belief behind Data Scope is that the “advanced search” screen should respond to the times. That's why Data Scope enables users to set search preferences explicitly. This not only gives more control to the user, but also improves privacy.

HIF is also innovative because it aims to become public property that people can freely use, modify and distribute. Online communities and companies should get the opportunity to modify their search technology as they see fit. The status-quo in search is a one-size-fits-all approach
Data Scope is also innovative because it aims to become public property that people can freely use, modify and distribute. Online communities and companies should get the opportunity to modify their search technology as they see fit. The status-quo in search is a one-size-fits-all approach


Development
===
===========

Currently HIF is in development. The models and “input” part of the system is more or less finished. The “processes” and “output” part of the system need improvements, refactoring and (unit)tests. Also HIF is currently written as a Django module, while it should probably be a project in itself with a Django client.

The next step is to write a frontend on the visual-translations “service” to demonstrate the use of the Hyper Information Framework.
Currently Data Scope is under full development. It features a visual-translations algorithm that answers: “what are iconical images for X in culture A, B & C”. A people-suggestions algorithm that answers: “which people are known for similar reasons that person X is known for”. And a city-celebrities algorithm that answers: "which well known people are associated with location X". I'm still tweaking these algorithms as well as building frontends for them in my ds-webapps repo. I want to improve on test coverage and documentation.

0 comments on commit b40e139

Please sign in to comment.