Free Jazz Information Science

By Roy Tennant -- Library Journal, 10/15/2005

This column was inspired and named by a respected colleague in an Internet chat room. A chat room is basically instant messaging in a crowd—well, a crowd of your closest friends, or at least those who show up. In this case, it was a chat room for library software developers. Since these people are creating our future, when they talk, I tend to listen. But as soon as Ross Singer from Georgia Tech announced the eponymous phrase, I knew exactly what he meant.

Free jazz, according to the Wikipedia, “is a movement of jazz music characterized by diminished dependence on formal constraints.” So, Free Jazz Information Science would be “a movement of information science characterized by diminished dependence on formal constraints.” This doesn’t mean that librarians should not rely on any formal constraints, since some limitations are useful and required. For example, it takes some formal constraints, such as compliance with the TCP/IP protocols, to even get on the Internet.

Aesthetic appeal?

But, nonetheless, librarians seem to be overly constrained at times. We mostly bring it on ourselves. In this particular case, my colleague used an analogy first in the chat room conversation, then in a blog posting, that I found particularly compelling. He characterized the SRW/U standard (which replaces Z39.50 to provide a way to search a catalog or index and get back structured results for reformatting and display) as a musician who is “technically adept, but aesthetically unappealing.” He then contrasted this with the OpenSearch specification from Amazon’s A9 search service (it does almost the same thing), which he characterized as “Keith Richards—untrained, but rocks your socks off.” He called this a “Metasearch Metadata Metaphor” in a recent blog posting.

He has been joined by others in his unflattering comparison of SRW/U to OpenSearch. A recent blog posting by Daniel Chudnov of Yale compared these two search protocols side by side. The comparison points out some key differences that support my colleague’s point about SRW/U being “technically adept, but aesthetically unappealing.” Unfortunately, people don’t tend to implement what they don’t like. Would you?

At the heart of this issue is how we develop standards in libraries. We tend to develop them in a top-down mode; we attempt to identify all possible cases and then create a large, complex standard that can accommodate all imagined uses and purposes. In the end, the standard is far too complicated to implement without further work to define a simplified profile for specific uses.

By contrast, the OpenSearch specification seeks to solve simply one specific purpose: to allow a user to perform a search using the A9.com search portal and get back results that can be integrated with the results from other sources. Complication will only creep in as working practice makes clear that alterations are needed. It is a bottom-up process, rather than a top-down one.

While librarians are involved with creating complex orchestras with which rich symphonic music may eventually be played, others are already out on the curb improvising with whatever instruments may be handy.

Learning to improvise

The essential problem is that we now live in a world that is no longer waiting for us to finish our standards development processes—change marches on, with or (increasingly) without us. We can either get with it or get left further behind. We need to learn to improvise.

Thankfully, those who are in a position to do something about this situation understand what’s at stake. The Board of Directors of the National Information Standards Organization (NISO) is in the midst of redesigning how NISO fosters, creates, and manages library standards. Although the changes are not yet implemented, the process of suggesting a standard (such as for OpenURL autodiscovery) and moving it into a successful life will soon be much easier and faster.

But the board also realizes that not all activities must result in a NISO/ANSI/ISO–accredited standard. Draft specifications, best practices, and implementation guidelines may be all that is required to support useful products and services.

Regardless of whether NISO gets its act together or not, my chat room colleagues would argue, line librarians, library systems staff, and others need to get their chops down. They need to think imaginatively about future directions for library services and steal riffs from whoever has something cool to emulate or use. If OpenSearch does the job you need accomplished, by all means use it to expose your catalog or index for others to search. When it no longer serves your needs, then move on to something else.

Improvisation plays a significant part in our lives—and it should play a significant part in our professional lives. If we wait for someone to hand us the score, it may already be too late. A good jazz musician may not know from where the next note is coming. Our services are more complex processes, certainly, but we could definitely stand more free jazz in our information science.