Networking Personal Collections: Portable Bibliographies

Like many of my multi-tasking Programs colleagues, I use a variety of online notebooks, bookmarking and citation management tools to keep track of documents and references related to ongoing projects. It’s a source of continual frustration to me that the reference lists and bibliographies that I develop in one space can’t be easily moved to another for re-use in a new context. There are plenty of tools to help me capture data as I move around the Web, but few that enable me to pool or integrate the sources that I’ve squirreled away in different folders and lists. My Google Notebook of shared print policies can’t be swapped into my Zotero library on collaborative collection management; a collection of tagged resources in NINES can’t be pushed into a personal bibliography in WorldCat; an iTunes library of RLG podcasts can’t be merged with a list of publications. These collections live quite near the surface of the Web, but they don’t exhibit the kinds of elective affinities that one might expect in the “social” network environment.

Of course, lists built with most citation management services can be exported and moved around in a number of standard formats (RIS, BibTeX) — but in a world where research data, services and social practices are moving onto the network, one might wish for a better solution. A solution that would enable students and researchers (and people like me) to create and exchange references and citations for the full range of information objects that we use, support context-aware resolution and delivery services, and allow references to be decoupled from the environment in which they were created and reintegrated into other work products. I want my references to free-associate with “others like this” in ways that will enhance and enrich my discovery experience. I want my references and citations (and yours) to rise to the surface and make themselves known.

Some standard tools and protocols exist to make this kind of network-aware personal collection management possible. The COinS specification for embedding OpenURL linking (citation) data in Web pages has been around for several years and has already been implemented in a variety of library service environments, from local OPACs to WorldCat record displays. (Lorcan has noted various advances in this area, here, here and here.) In combination with an OpenURL Resolver, it enables Web users to link to library-owned or licensed content. For example, COinS enables Wikipedia users to locate library-owned titles from citations in articles, like this bibliography of W.H. Auden:

and it also powers Zotero’s seemingly magical ability to lift citation metadata directly out of certain kinds of superficially unstructured Web content, including blog postings:

As for my dreams of a day when references and bibliographies are freed from the various boxes, drawers, folders and files in which they languish…well, we’re a step closer now that WorldCat lists have implemented a COinS compliant view of bibliographies built within Thanks to some energetic OCLC colleagues, all of the citations in the tens of thousands of public WorldCat lists, can now be easily mixed and commingled, set free to move about the Web as resolvable, actionable references. If you’re a Zotero user, you’ll see the change immediately: WorldCat lists now self-identify as content that’s ready for exchange and circulation

— shiny new COinS of the network realm.

Tweet about this on TwitterShare on TumblrShare on LinkedInShare on FacebookBuffer this pageShare on Google+Email this to someone


  1. Constance, this is very interesting. I would be extremely interested to get a feel for how COinS relate to the apparently stalled citation microformats effort, or more generally to RDFa.

    COinS feels like a rather special case; going back to your initial plea, which seems to relate to more general portability, wouldn’t a RDF-based approach be more widely applicable? Perhaps Web2 applications might find it harder to resist the siren call of the semantic web, than a very library-centric technology like OpenURL?

  2. Chris — you make an excellent point about OpenURL’s arguably parochial appeal. I think we do (in the library world) tend to regard it as a magic bullet, when other more general standards / tools / technologies might do the job. And, as you say, my starting point was really about wanting the generic information environment to exhibit a kind of ’surface tension’ that would enable references to bead up more or less naturally, as a property of the network. (I may be pushing that metaphor a bit far!) The advantage of OpenURL in this context, I think, is that it honors both the traditional scholarly practice of referenceability and contemporary expectations for immediate reosolution and access. Of course, one needs some additional bits, like persistent identifiers (as Stu keeps reminding us, viz., and community agreements around what content will be preserved. I don’t think OpenURL is ‘the’ answer, but I’m still glad to find more COinS scattered about. If nothing else, their proliferation can serve as a demonstration of need…traces of a culture that depended on a currency of exchange The Bibliographic Ontology group has some interesting stuff on RDF and rich citation formats, and is sensibly focused on the requirements for moving citations into the authoring flow:

Comments are closed.