This is the software publisher's description.

The World Wide Web is a vast repository of information containing billions of documents. This presents two problems: how do you find what you need to know and what do you do with it when you have found it? Google and its competitors have a workable but by no means perfect solution to the first problem. ScrapBook is the solution to the second problem for those who want to record what they have found.

ScrapBook is aptly named: it is your personal repository for the text snippets, PDFs, web pages, movies and whatever you collect in your travels around the web.

These three points are important. ScrapBook is personal in the sense that your scrapbook is part of your Firefox profile by default1. It is also personal in that you decide how you want to organise your collection. It is a repository since it makes a local snapshot of the web content you specify. Finally ScrapBook items can be in virtually any format: almost anything Firefox can display, ScrapBook can capture.

Prior to ScrapBook or similar programs, you either book-marked interesting web pages or saved them to your hard disk. In the former case you saved where the page was but not its content; in the latter you saved the content but not where it was from. ScrapBook automatically saves the content, the source URL and the date it was captured. For researchers, who have to document the content they cite, there is no contest. Even the casual user benefits since there are no more saved files lost somewhere on the hard disk: they are all there in the scrapbook!

Add to this that ScrapBook allows selective capture: from a snippet of selected text, to a frame, through a whole page with or without embedded images, to a page together with its linked pages to various depths. Add, furthermore, that ScrapBook items can be searched and sorted, edited and annotated, classified and filed away.