| Project Description |
An RSS aggregator that applies per-user classification to RSS entries. |
| Current Project Version |
0.9.0 |
| Project Maturity |
Alpha |
| Project License |
Apache License 2.0 |
| Compatible GigaSpaces XAP Version |
GigaSpacesXAP6.02 |
| Project Captain |
[~jottinger@techtarget.com]. |
| Contributors |
[~jottinger@techtarget.com], Owen Taylor. |
Project Sitemap
Project Homepage
Features and Capabilities
BrightReader is a trainable RSS aggregator; Each user can train the system to his or her own preferences, and should be able to choose what the system should highlight. It scales out three aspects of the application:
- Data. It currently uses the space for all data, including users, feeds, entries, and training data.
- Feed reads. As feeds become stale, a listener is notified for each feed, meaning that the entire grid is potentially involved in reading feeds.
- Rating. As new entries come in, each one is rated according to the user's training data. This is a computationally expensive operation, and is the primary reason the application needs a grid: it would be fairly easy, with a large number of feeds, to overwhelm the rating queue.
There are still some changes to put in, of course. The two most important ones:
- Use Spinn3r as a source of entries rather than having BrightReader read each feed itself.
- A JSF UI is included as a module, but it's unsatisfactory; I'd prefer to use Struts2 or GWT for the interface.
