refdb was designed with the needs of a scientist in mind. At least one third of a scientist's work hours are spent reading and evaluating other scientist's publications. In no time you have a pile of offprints or PDF files that you would like to archive in a way that keeps the information accessible. refdb helps you to keep track of those papers by creating a database with the essential bibliographic information, keywords for easy retrieval, your personal comments, the reprint status, a hint where you can find the offprint (or a link to the electronic version), and additional information for your personal categorization.
Extended notes help you organize your references according to topics and allow you to associate additional information with references, keywords, authors, or periodicals. This goes far beyond the one-note per reference approach of commercial tools, although this is of course still supported. Extended notes are associated to the user that created them, but they are visible to all users unless you restrict the display accordingly.
Scientists rarely work alone on a subject. More commonly they are members of a group with similar research interests. It would be a waste of time and effort if each member of the group hunts offprints individually, resulting in duplicate copies, and if each member keeps his personal reference database, making it hard to consolidate an information pool of the group. Instead of selling a complete database engine to each scientist and "allowing" the users to peek at each other's databases, refdb uses the inverse approach: designed as a client/server application from the ground up, it allows a group to maintain a common reference database, but still allows each individual to view only those references that she added or explicitly selected. At the same time, each group member can find out in an instant which offprints are available in the whole group. However, there is no problem if you want to use refdb as an individual user.
Eventually you'll want to publish your own data. You'll have to cite dozens of papers that you've already read. Your refdb database will make it a breeze to retrieve those papers by author, by keyword, or by any other field or combination of fields. If you write your document using one of the supported SGML or XML DTDs or as a LaTeX document, refdb can create the bibliography based on the citations in your document. It will also help you to format the citations and the bibliography according to the instructions of your publisher. Your paper got rejected and you want to submit it to a different journal? Fear not, refdb can easily re-transform your document and format it for a different publisher.
Of course it is also feasible to manage your personal library of contemporary literature. Or your favourite cookie recipes. refdb wasn't precisely designed for these purposes, but it's up to you to make the best use of the package.
If all this sounds useful for you, you should give refdb a try.