OpenRank
From Wikipedia, the free encyclopedia
 |
This article or section may contain original research or unverified claims. Please improve the article by adding references. See the talk page for details. (September 2007) |
|
|
This article or section is written like an advertisement. Please help rewrite this article from a neutral point of view. Mark blatant advertising which would require a fundamental rewrite in order to become encyclopedic for speedy deletion, using {{db-spam}}. (December 2007) |
OpenRank is an open source ranking algorithm that is designed to be able to rank items such as web pages, photos, books, etc in a way that the ranking tries to predict users' interests and tastes.
The OpenRank project was started in 2007 in an attempt to create a new and improved may of ranking websites and other items in a multi consumer/producer environment. Based on the idea that the power that comes from an effective and powerful search method is not something that should rest in the hands of the few. OpenRank attempts to freely place the best technology, in the hands of the many, to the benefit of all.
Contents |
[edit] Limitations of current best solution PageRank
In its most crude description PageRank uses the number of links to and from a website to give the page a rank. This rank is then used in tools such as the Google search tool. To some this is a fundamentally flawed concept, from a past time, when the internet was simpler. In the book The Long Tail Chris Anderson highlights that popularity is no longer the best way to rank an item's importance to an individual, and that personal taste is now king. OpenRank takes into account the tastes and views of the user to give a weighting to the links to the item and the items itself
[edit] Uses
Whilst a better search engine is an ideal goal for OpenRank, there are many other smaller scale uses of OpenRank. Example of use would be: The way to locate images on the photo sharing website Flickr (as of May 2007) is based on tags. Using a better search method that knew what kinds, and types, of image you like, it would be able to return images that would interest you.
[edit] The Phases of development
- Phase 1 - What should OpenRank be? (Current Phase)
Defining and discussing what factors should be considered in making the best process for ranking items and content
- Phase 2 - Design
Working out how OpenRank can be created
- Phase 3 - Creation
Creating of OpenRank
- Phase 4 - Let get searching
Implementing OpenRank in the real world
- Phase 5 - Continual refinement
[edit] Phase 1 What Should OpenRank be?
OpenRank should be able to
- Rank any item.
- Take into account the personal tastes of the user. The person tastes of the user should not need to be specifically stated (though the user would need to be able to see what information is being stored on them - see openness). Many users produce content similar to that which they are searching for. If someone blogs about The Moon then we can imply that they are interested in The Moon (unless their post was about how boring The Moon is).
- Take into account the pier group of the user so that recommendations from other users can play a part
- Be abuse-proof. In a completely open process the owners of content should be able to view and simply understand how the process works. Currently where the ranking of items such as website is known by the few, they can get a significantly higher rank. By opening the process to everyone, everyone is on a level playing field and the ranking cannot be fooled.
- Be completely open. Any user should be able to see what information is being recorded about them; They should be able to update, alter, and add to this information. If people don't like the results it is returning they should be able to rework the process to improve it. People creating content should have access (see abuse proof point above)
