This morning's CTO Breakfast seemed to keep coming back to the same idea expressed in different ways. We started off talking about pornography filters and the possibility of using something like real-time blacklisting for SPAM to identify URLs and IP numbers that publish pornography to make it easier for filters to pick it out. At another point we talked about using information from logfilter to create blacklists for portscanning script kiddies and their zombies. Yet another conversation dealt with a Microsoft project for attaching comments to products based on their barcode.
What do all these conversations have in common? The use of an identifier to enable tracking some activity back to a source document or location. For example, another thing we talked about was using Del.icio.us for annotating Web pages. Imagine a Firefox plug-in that puts a pane on the side of the browser. Whenever you browse to a page, the pane shows all of the people who've bookmarked that page and what they put in their comments on Del.icio.us. Again, the URL provides a unique identifier for the page so that you can find every reference to it on Del.icio.us. Of course, Technorati and even Google are doing the same kind of thing.
As I was playing with Google's new academic search tool yesterday, I realized that one of the problems that academic papers have in the online world is the lack of a canonical identifier. Because journals do not routinely assign a URL to a paper, its difficult to refer to it in a way that always works.
Digital Object Identifiers are a possible solution to this problem. Here's an overview of how it works. Of course, DOIs are not just for academic papers, but anything represented digitally. This means that you can create a canonical name for lots of things that might not otherwise have them and then use those canonical names to do mundane things like track the digital object, or more exciting things, like creating trackback systems that find out who's referencing it, using it, commenting on it, and so on.