In my spare time I'm working on InfoAgent for Squeak.
I intend to provide automatic and "intelligent" support for information retrieval on the Web.
Including:
Tracking changes of arbitrary Web-Documents
MetaSearch engine that can be configured to "tap" any Search-engine on the web
Bookmark management
Seamless integration of these features into Scamper (see improve Scamper Project).
I derived modifications of some of the HTML base classes for extraction of urls from web pages. Reusing this code will make it easy to build a webspider. I´m planning to work on this when I´m finished with InfoAgent.