decentralized bibliographic services


Jill Walker has had an interesting series of posts related to the privileging of Amazon through book links, and asked about developing methods of neutral linking.

I did some poking around, because it seemed to me that somebody must already be doing this. Yup. There's a group called working on a "bibp://" protocol for links to books. They call it a "decentralized bibliographic service network." Basically, users could instruct their clients on what resource(s) to use for retrieving bibp links--from libraries to booksellers. (In the same way they can now specify mail clients for mailto: protocols, or LDAP servers for directory lookups.)

Problem is, it doesn't seem to be terribly active or visible. The IETF draft (Bibliographic Protocol Level 1: Link Resolution and Metapage Retrieval) was last updated in August of 2000. One of the participants in this project, Robert Cameron, wrote an article about it in 1997 for First Monday, entitled "A Universal citation Database as a Catalyst for Reform of Scholarly Publication."

So the question for me becomes how to get this type of project a little higher in the public consciousness, and more actively into the development pipeline. The potential is there, but if it's so far under the radar that nobody builds it into their systems, it won't help anyone.


I found this entry through your comments on Jill's site. What you're talking about (the consequences of relying on retailers for information) already has some relevance on the way allconsuming relies on Amazon for information. For example, long out-of-print books that are no longer available for purchase are not kept in the catalog. Although it may not be a long-term or scalable solution (since I'm the only person running allconsuming, and do not have the financial backing or authority of a company or standards organization), I have been collecting all isbns around the web regardless of whether or not they're in Amazon's catalog, and could probably create a way for this information to be collected and aggregated outside of Amazon's clutches, still within the domain of books. At the same time, as much as I wish non-profit organizations with no ulterior motives were actively finding better solutions to this problem, it's much easier for (semi-)profitable, biased, businesses to do so initially, as they have a reason to encourage adoption, provide stability, and invest in innovation. The fact that we can rely on ISBNs at all is a huge luxury that we shouldn't take for granted, and has already helped a lot. I'm not sure a new protocol for book links alone will ever take off, as it naturally begs the question, what about music links? What about news links?

Liz, I'm so impressed you found this (you librarians really ARE good at finding stuff, aren't you?) - this is a great discussion. Erik's points are interesting too... And Noah and .... (sorry, sleepy, goodnight)

Erik, thanks for chiming in. (And thanks, too, for, which I really do love!)

The bibp protocol isn't just for books. It's for "bibliographic information," which--despite the "biblio" prefix--goes way beyond just books. Libraries have been cataloging non-book materials for a long time...and there are lots of accepted standards for that markup. The Z39.50 protocol that bibp is designed to interact with is capable of pointing to any sort of material that a library could hold--including serials (news), music, films, etc.

I really didn't spend enough time looking at bibp to know how well it interfaces with Z39.50, but I am going to post a message to my favorite library-tech mailing list (not a public list), and see what the folks there know about what's going on in library-land these days on this front.

About this Entry

This page contains a single entry published on November 15, 2002 4:16 PM.

monolithic information structures was the previous entry in this blog.

immersive experiences is the next entry in this blog.

Find recent content on the main index or look in the archives to find all content.


Category Archives