thanks. couple of things I have in mind

1. more output formats like JSON, HTML, Doc, RSS, ATOM

2. allow programming in loops.. for example you want to be able to scrape many pages of data for a google search result then you need a hack in the url itself. This is difficult to accomplish but I am think I can add support to some of the pagination patterns to start with

3. to be able to give more control over the HTML, maybe a panel which helps extract the relevant one from the target page