Projects

I have a few projects put here, and a few more coming when I get to that point. The first one, in importance if not chronological order, is the JCN Project. What is the JCN Project? It is a framework for Intelligent Web Agents for internet browsing. It involves three elements:

1. The Sidebar
   This is where the agent communicates with the user, via AJAX updates.
2. The Spy
   A Greasemonkey user script sends on load a piece of the current page (URL, Domain, Title, Referrer) to the Agent. This is the grist for the Agent's mill, giving it raw data to work from.
3. The Agent
   This is where it becomes cool. Pieces of code work from the data set that the Spy provides. What I have isn't the cool part -- the most commonly hit domains of the last hour and the most commonly hit domains ever. I'll be writing more. And I'll be putting together a standalone HTTP server so it runs on localhost instead of a central server.
   1. News
      This was the first-working aspect of this program. As currently set, it reads a list of RSS feeds to read, then asks for them through a cache on the server side. There are two main reasons for that:
      1. xmlHTTPRequest on Firefox, the intended platform for this tool, doesn't allow requests on domains other than the one the request comes from, and
      2. By caching the RSS feeds, the program won't batter the server the feed is hosted on.
   2. Bookmarks
      Bookmarks are handled through del.icio.us, a service that already did what I had intended with this tool.
   3. Agents
      xxx

I have a Sourceforge account set up, and I'll have this packaged up and released as soon as I can.